Figure 4.1	Assessment Area and Sites of Conservation Importance
Figure 4.2	Management Compartments at Lut Chau as prescribed in the Mai Po Inner Deep Bay Ramsar Site Management Plan
Figure 4.3	Survey Transect, Location of Infra-red Camera and Flightline Vantage Points
Figure 4.4	Pond Number and Habitats at Nam Sang Wai, Lut Chau and Surrounding 500m in 2021
Figure 4.4a	Pond Number and Habitats at Nam Sang Wai, Lut Chau and Surrounding 500m in 2021 (Zoom-in 1 of 6)
Figure 4.4b	Pond Number and Habitats at Nam Sang Wai, Lut Chau and Surrounding 500m in 2021 (Zoom-in 2 of 6)
Figure 4.4c	Pond Number and Habitats at Nam Sang Wai, Lut Chau and Surrounding 500m in 2021 (Zoom-in 3 of 6)
Figure 4.4d	Pond Number and Habitats at Nam Sang Wai, Lut Chau and Surrounding 500m in 2021 (Zoom-in 4 of 6)
Figure 4.4e	Pond Number and Habitats at Nam Sang Wai, Lut Chau and Surrounding 500m in 2021 (Zoom-in 5 of 6)
Figure 4.4f	Pond Number and Habitats at Nam Sang Wai, Lut Chau and Surrounding 500m in 2021 (Zoom-in 6 of 6)
Figure 4.5	Waterbird Utilization of Drainage Channel between April 2021 and March 2022
Figure 4.6	Significant Ecological Constraints in and around Nam Sang Wai site, Bird Flightlines, Roosts and Concentrations
Figure 4.7	Major Flightline of Great Cormorant Departing Roost in 2021-2022
Figure 4.8	Identified Flightlines Identified over Proposed SPR Bridge in Surveys conducted between April 2021 and March 2022
Figure 4.9	Identified Flightlines at Shan Pui River Egretry during Surveys in Breeding Season 2022 (Apr-Jul)
Figure 4.10	Species of Conservation Importance Recorded and Cormorant Roosting Locations at Nam Sang Wai, Lut Chau and Surrounding 500m in 2021 – 2022
Figure 4.10a	Species of Conservation Importance Recorded and Cormorant Roosting Locations at Nam Sang Wai, Lut Chau and Surrounding 500m in 2021 – 2022 (Zoom-in 1 of 6)
Figure 4.10b	Species of Conservation Importance Recorded and Cormorant Roosting Locations at Nam Sang Wai, Lut Chau and Surrounding 500m in 2021 – 2022 (Zoom-in 2 of 6)
Figure 4.10c	Species of Conservation Importance Recorded and Cormorant Roosting Locations at Nam Sang Wai, Lut Chau and Surrounding 500m in 2021 – 2022 (Zoom-in 3 of 6)
Figure 4.10d	Species of Conservation Importance Recorded and Cormorant Roosting Locations at Nam Sang Wai, Lut Chau and Surrounding 500m in 2021 – 2022 (Zoom-in 4 of 6)
Figure 4.10e	Species of Conservation Importance Recorded and Cormorant Roosting Locations at Nam Sang Wai, Lut Chau and Surrounding 500m in 2021 – 2022 (Zoom-in 5 of 6)
Figure 4.10f	Species of Conservation Importance Recorded and Cormorant Roosting Locations at Nam Sang Wai, Lut Chau and Surrounding 500m in 2021 – 2022 (Zoom-in 6 of 6)
Figure 4.11	Habitat Map, Species of Conservation Importance Recorded and Cormorant Roosting Locations at Nam Sang Wai, Lut Chau and Surrounding 500m in 2021
Figure 4.11a	Habitat Map, Species of Conservation Importance Recorded and Cormorant Roosting Locations at Nam Sang Wai, Lut Chau and Surrounding 500m in 2021 (Zoom-in 1 of 6)
Figure 4.11b	Habitat Map, Species of Conservation Importance Recorded and Cormorant Roosting Locations at Nam Sang Wai, Lut Chau and Surrounding 500m in 2021 (Zoom-in 2 of 6)
Figure 4.11c	Habitat Map, Species of Conservation Importance Recorded and Cormorant Roosting Locations at Nam Sang Wai, Lut Chau and Surrounding 500m in 2021 (Zoom-in 3 of 6)
Figure 4.11d	Habitat Map, Species of Conservation Importance Recorded and Cormorant Roosting Locations at Nam Sang Wai, Lut Chau and Surrounding 500m in 2021 (Zoom-in 4 of 6)
Figure 4.11e	Habitat Map, Species of Conservation Importance Recorded and Cormorant Roosting Locations at Nam Sang Wai, Lut Chau and Surrounding 500m in 2021 (Zoom-in 5 of 6)
Figure 4.11f	Habitat Map, Species of Conservation Importance Recorded and Cormorant Roosting Locations at Nam Sang Wai, Lut Chau and Surrounding 500m in 2021 (Zoom-in 6 of 6)
Figure 4.12a	Transect used for MPBWF Surveys
Figure 4.12b	Distribution of MPBWF in 2021 Surveys
Figure 4.12c	Relative Distribution of MPBWF Observations during 2019 and 2021 Surveys
Figure 4.13	Habitats under the Development Site
Figure 4.14	Proposed Development Site and the Proposed Habitats outside Development Site
Figure 4.15	Wetland Enhancement Phasing Plan and Schedule
Figure 4.16a	Maximum exclusion zone and zone of reduced density due to construction phase disturbance impacts of the proposed bridge
Figure 4.16b	Maximum exclusion zone and zone of reduced density due to the disturbance impacts during the construction phase of the Development Site and operational phase of the proposed bridge
Figure 4.16c	Maximum exclusion zone and zone of reduce density due to operational phase disturbance impacts of the project (Development Site and proposed bridge)
Figure 4.16d	Maximum exclusion zone due to the disturbance impacts during the construction phase of the high-rise buildings
Figure 4.17	Areas of Intertidal Habitat at LCNR to be restored and enhanced

Appendices

Appendix 4.1a	List of Flora Species Recorded (Nam Sang Wai)
Appendix 4.1b	List of Flora Species Recorded (Lut Chau and 500m Assessment Area)
Appendix 4.1c	List of Plant Species Recorded in Ponds in Nam Sang Wai Project Site (DAFOR Scale)
Appendix 4.2a	Survey records at Nam Sang Wai, Lut Chau and the 500m Assessment Area during 2011 to 2015
Appendix 4.2b	Fauna Species Recorded at Nam Sang Wai, Lut Chau and the 500m Assessment Area in April 2021 – March 2022
Appendix 4.2c	Mitigation Target Species and Numbers (Birds): Calculation Details
Appendix 4.3	Representative Photos of Habitats at Nam Sang Wai, Lut Chau and 500m Assessment Area
Appendix 4.4	Current population, distribution and habitat preferences of Mai Po Bent-winged Firefly Pteroptyx maipo in the Inner Deep Bay area of Hong Kong (Prepared by AEC as part of a Further Supplementary Witness Statement submitted to the Town Planning Appeal Board in 2020 (TBP Appeal No. 1 of 2017 (1/17))
Appendix 4.5	Conservation and Management Plan

Glossary

AFCD	Agriculture, Fisheries and Conservation Department
CDWE	Comprehensive Development with Wetland Enhancement
CMP	Conservation and Management Plan
DAFOR	Dominant, Abundant, Frequent, Occasional, Rare
EcolA	Ecological Impact Assessment
EIA	Environmental Impact Assessment
EIAO	Environmental Impact Assessment Ordinance
EIAO-TM	Technical Memorandum on Environmental Impact Assessment Process
EPD	Environmental Protection Department
ER	Enhancement Ratio
HKPSG	Hong Kong Planning Standards and Guidelines
IUCN	International Union for Conservation of Nature
KFBG	Kadoorie Farm and Botanic Garden
KTMDC	Kam Tin Main Drainage Channel
LCNR	Lut Chau Nature Reserve
LMC	Lok Ma Chau
LMC EEA	Lok Ma Chau Ecological Enhancement Area
MA	Management Agreement
MER	Minimum Enhancement Ratio
MLP	Master Layout Plan
MPNR	Mai Po Nature Reserve
MPBWF	Mai Po Bent-winged Firefly
MTR	Mass Transit Railway
NCO	Noise Control Ordinance
NNCP	New Nature Conservation Policy
NPOs	Non-profit-making Organisations
NSRs	Noise Sensitive Receivers
NSW	Nam Sang Wai
NSW WEA	Nam Sang Wai Wetland Enhancement Area
OZP	Outline Zoning Plan
PELB	Planning, Environment and Lands Bureau
PME	Powered Mechanical Equipment
PPP	Public-Private Partnership
PRC	People’s Republic of China
SPR	Shan Pui River
SSSI	Site of Special Scientific Interest
TPAB	Town Planning Appeal Board
TPB	Town Planning Board
TPB-PG	Town Planning Board Guideline
TSL	Tung Shing Lane
WBA	Wetland Buffer Area
WCA	Wetland Conservation Area
WEA	Wetland Enhancement Area

4 Ecological Impact (Terrestrial and Aquatic)

4.1 Introduction

4.1.1 In this Section, the Ecological Impact Assessment, the ecological baseline conditions of the Assessment Area are described, and the potential ecological impacts of the proposed Project are detailed and evaluated; proposed measures to mitigate all significant potential adverse ecological impacts of the construction and operation of the Project are described, together with the projected permanent ecological benefits of its sustainable operation in the long term.

4.1.2 The Project proposes to develop the Project Site (‘the Site”) for comprehensive residential development with wetland enhancement and connecting road bridge between the CDWE and Wang Lok Street in Yuen Long. The Project Site is divided into two portions, Nam Sang Wai (NSW) and Lut Chau (LC) site which occupy about 121.9ha and 55.4ha respectively.

4.1.3 This Ecological Impact Assessment has been prepared to examine the flora, fauna and other components of the ecological habitats within the assessment area, including those highlighted in section 3.2.1(vi) in the Study Brief (ESB-244/2012). The aim of this Assessment is to protect, maintain or rehabilitate the natural environment. In particular, the Project and associated works have aimed to avoid or minimise impacts on recognised sites of conservation importance and other ecologically sensitive areas, or to mitigate for any impacts that cannot be avoided or minimised. The assessment has also identified and quantified as far as possible the potential ecological impacts and benefits arising from the construction and operation of the Project and associated works.

4.1.4 There are numerous historical Applications relevant to the site under the Town Planning Board Ordinance. In 2015 an application was lodged with a Development Site area of about 11.6ha, which had the same Master Layout Plan as current submission. The proposed development consisted of a mixture of towers and houses, included a large public park, an elderly centre, a visitors’ centre and an interpretive trail through the edge of the NSW WEA. A full EcoIA and CMP were submitted. The application, (No. A/YL/NSW/242) was rejected by the TPB and an appeal was lodged with the Town Planning Appeal Board (TPAB).

4.1.5 The TPAB considered the appeal commencing in November 2020 and ending in March 2021. The decision was issued on the 31 December 2021 and granted approval to application No. A/YL/NSW/242 subject to 15 conditions. The TPAB decision can be found at their website.

https://www.tpab.gov.hk/filemanager/en/content_47/Decision%20TPA%20No.1%20of%202017.pdf

4.1.6 Submissions are now being made under the conditions of approval for the approved development as part of the implementation of the project. These are being made in parallel with the submission of this application under the EIAO.

4.2 Environmental Legislation, Standards and Guidelines, and Related Documents

4.2.1 Legislation, standards, guidelines and related documents relevant to and consulted in the preparation of this ecological assessment include the following:

● EIA Study Brief (ESB-244/2012) issued for the proposed project, Proposed Comprehensive Development with Wetland Enhancement (CDWE) at Nam Sang Wai and Lut Chau

● Environmental Impact Assessment Ordinance (EIAO) (Cap. 499) and subsidiary legislation and guidelines

● EIAO Technical Memorandum on Environmental Impact Assessment Process (EIAO-TM), in particular Annexes 8 and 16

● EIAO Guidance Note No. 3/2010 - Flexibility and Enforceability of Mitigation Measures Proposed in an Environmental Impact Assessment Report

● EIAO Guidance Note No. 6/2010 - Some Observations on Ecological Assessment from the Environmental Impact Assessment Ordinance Perspective

● EIAO Guidance Note No. 7/2010 – Ecological Baseline Survey for Ecological Assessment

● EIAO Guidance Note No. 10/2010 – Methodologies for Terrestrial and Freshwater Ecological Baseline Surveys

● EIAO Guidance Note No. 11/2010 – Methodologies for Marine Ecological Baseline Surveys

● AFCD – Strategic Feasibility Study on the Development of Wetland Conservation Parks System Under the Northern Metropolis Development Strategy. Study Report October 2024.

● Country Parks Ordinance (Cap. 208) and its subsidiary legislation

● Forests and Countryside Ordinance (Cap. 96) and its subsidiary legislation, the Forestry Regulations

● Hong Kong Planning Standards and Guidelines (HKPSG) Chapter 10, "Conservation"

● Mai Po Inner Deep Bay Ramsar Site Conservation Strategy and Management

● Marine Parks Ordinance (Cap. 476)

● New Nature Conservation Policy (priority sites present include the Mai Po Inner Deep Bay Ramsar Site and Deep Bay Wetland outside Ramsar Site)

● PELB Technical Circular 1/97 / Works Branch Technical Circular 4/97 “Guidelines for Implementing the Policy on Off-site Ecological Mitigation Measures”

● Protection of Endangered Species of Animals and Plants Ordinance (Cap. 586)

● Town Planning Board Guideline No. 12C – Application for developments within Deep Bay area under Section 16 of the Town Planning Ordinance

● Wild Animals Protection Ordinance (Cap. 170)

4.2.2 International, National and Hong Kong SAR conventions, regulations and guidelines that are relevant and have been referenced in this study include the following:

● IUCN “Redlist” of globally threatened species of wild plants and animals (http://www.iucnredlist.org/ static/introduction ).

● Study on the Ecological Value of Fish Ponds in Deep Bay Area (1997)

● United Nations Convention on Biological Diversity. This convention requires parties to regulate or manage biological resources important for the conservation of biological diversity whether within or outside protected areas, with a view to ensuring their conservation and sustainable use. It also requires parties to promote the protection of ecosystems, natural habitats and the maintenance of viable populations of species in natural surroundings. The People’s Republic of China (PRC) ratified the Convention on Biological Diversity on 5th January 1993. The convention came into force in Hong Kong during 2011.

● Convention on International Trade in Endangered Species of Wild Fauna and Flora (“CITES”). This Convention regulates international trade in certain animal and plant species. Their trade is subject to permits or certificates of origin. Hong Kong’s obligations under this Convention are enforced via the Protection of Endangered Species of Animals and Plants Ordinance (Cap. 586)*

● Convention on Wetlands of International Importance (Ramsar Convention). The Ramsar Convention relates to the protection and wise use of wetland ecosystems for the protection of biological diversity and sustainable development. The convention requires signatories to designate at least one wetland for inclusion in a list of Wetlands of International Importance (Ramsar sites); Mai Po Inner Deep Bay Ramsar Site is designated under this convention and supports internationally important numbers of several bird species.

● Jiang et al. (2016). Red List of China’s Vertebrates. Biodiversity Science. 24. 500–551.

● The List of Wild Plants and Wild Animals Under Special State Protection under the Law of the People’s Republic of China on the Protection of Wildlife (promulgated by State Forestry Administration and Ministry of Agriculture on 14 January 1989)*

● Wild Animals Protection Ordinance (Cap. 170)*

● Fellowes et al. (2002). Wild animals to watch: terrestrial and freshwater fauna of conservation concern in Hong Kong. Memoirs of the Hong Kong Natural History Society 25: 123-159

● Note that referenced sources indicated thus (*) are primarily related to wildlife protection and may not reflect the level of conservation concern or importance attached to the wild population of a species in Hong Kong.

4.3 Assessment Methodology

General

4.3.1 The assessment area for the purpose of terrestrial ecological assessment includes all areas within 500 metres from the Project Boundary and any associated access road/facilities as well as the areas likely to be impacted by the Project. For aquatic ecology, the assessment area is the same as that of the water quality impact assessment described in section 3.4.4.2 in the EIA Study Brief.

4.3.2 The ecological surveys addressed the requirements to establish the ecological baseline profile and followed the requirements of the pertinent of Annexes 8 and 16 of the Technical Memorandum as well as Environmental Impact Assessment Ordinance Guidance Notes No.7/2010 “Ecological Baseline Survey for Ecological Assessment”, No.10/2010 “Methodologies for Terrestrial and Freshwater Ecological Baseline Survey” and No.11/2010 “Methodologies for Marine Ecological Baseline Surveys.

Ecological Field Surveys

4.3.3 Whilst no specific information gap has identified, to provide an updated ecological baseline condition and to satisfy the requirements on ecological surveys set out in the EIA Study Brief, field surveys were undertaken between April 2021 and July 2022, covering habitats, vegetation, mammals, birds, herpetofauna, odonates, butterflies and fireflies. Details of the survey methodologies are presented in the following sections, with transects and counting points shown in Figure 4.3. The programme of field surveys conducted is presented below:

Table 4.1: Ecological Field Survey Programme

Surveys	2021									2022
	Wet Season							Dry Season					Wet Season
	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec	Jan	Feb	Mar	Apr	May	Jun	Jul
Habitat and Vegetation
Walk-Over and Drone Surveys	✓							✓
Focused Vegetation Transect Survey			✓	✓	✓					✓
Mammal
Terrestrial Mammal Transect Survey	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
Camera Trapping	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
Avifauna
Avifauna Transect Survey	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
Avifauna Flightline Survey	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
Egretry Count													✓	✓	✓	✓
Egretry Flightline Survey													✓	✓	✓	✓
Cormorant Roost Count								✓	✓	✓	✓	✓
Ardeid Night Roost Count								✓	✓	✓	✓	✓
Herpetofauna
Herpetofauna Transect Survey	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
Terrestrial Invertebrates
Butterfly Transect Survey	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
Odonate Transect Survey	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
MPBWF Transect Survey	✓	✓	✓	✓	✓	✓	✓

Habitat and Vegetation Surveys

4.3.4 Drone survey flights conducted in August 2021 and January 2022 generated high resolution photographs taken at approximately 15m above ground level. From these habitats were identified and mapped. A ground-truthing survey was then performed to verify the habitats and species identified. The habitat of the entire site and the surrounding 500m were mapped using ArcGIS and the area of each habitat was calculated. The habitat map is shown in Figures 4.4a-f. Representative photographs of habitats and any ecologically important features are provided in Appendix 4.3.

4.3.5 Vegetation surveys were conducted for all habitats via transects along accessible routes. Surveys were also conducted in surrounding areas to characterise the habitats present. Effort was concentrated immediately surrounding the site and with greatest potential to be impacted by the development. Surveys were conducted during the wet season in June, July and August 2021 and the dry season in January 2022. Floral species observed during the surveys were identified to species level, with their relative abundance recorded.

4.3.6 A focused vegetation survey was conducted in wetland habitats at NSW including the proposed Development Site to confirm baseline conditions, especially where particularly important changes had taken place since the time of previous surveys detailed in AEC (2016). Surveys were carried out by walking along accessible bunds. For each defined key habitat of each waterbody, all notable plant species and their relative species abundance were recorded using the DAFOR scale. The DAFOR scale is a semi-quantitative sampling method used for recording the relative abundance of plant species. It categorises the species’ relative abundance according to the approximate percentage cover. The relative abundance of all recorded plant species were categorised into five groups: Dominant (D) species, with more than 75% coverage and generally covering more than three quarters of the habitat; Abundant (A) species, with ~51%-75% coverage, and typically common; Frequent (F) species, with ~26-50% coverage, frequently encountered when walking through a habitat; Occasional (O) species, with ~11-25% coverage, and relatively low frequency and low cover; and Rare (R) species, with ~1-10%, and found once or a very few times in a habitat.

Mammals

4.3.7 Surveys for terrestrial mammals were conducted along transects once per month. Since most mammal species in Hong Kong occur at low densities, surveys for mammals included both direct observation and active searching for signs of mammal occurrence (including potential roost, footprints, and droppings). Night-time surveys were also conducted monthly to supplement the findings from daytime surveys.

4.3.8 Direct observation of terrestrial mammals is typically difficult due to their secretive nature and largely nocturnal behaviour. Furthermore, as required in the EIA Study Brief, special attention was paid to the potential presence of Eurasian Otter within the Assessment Area. Accordingly, remote camera traps were set up to survey mammals. These traps were set up in wooded areas on bunds and were set to record when an infra-red beam was triggered. Three camera traps were set at various locations at NSW between April 2021 and March 2022.

4.3.9 Although a relatively high density of bats may be expected at a wetland such as this, bat diversity around Deep Bay is low (Shek 2006). Though not specifically required under the EIA Study Brief, bat surveys were conducted in parallel with other night-time transect surveys, with bats detected and identified during their active period after sunset by hand-held ultrasonic bat detector (Anabat Walkabout Active Bat Detector or Wildlife Acoutsics Echo Meter Touch 2 Pro).

Birds

4.3.10 Bird surveys were conducted at NSW and LC monthly from April 2021 to March 2022. During the surveys, a transect covering all identified habitat types, was followed which permitted observation of all waterbodies within the site, as well as the adjacent tidal creeks. Transect bird surveys were conducted in the early morning to cover the period when birds are most active, while night-time surveys were also conducted to detect nocturnal bird species in parallel with those for other fauna. All bird species seen or heard were recorded, and the number of individuals was recorded.

4.3.11 Evening surveys were conducted to investigate the number and distribution of Great Cormorants roosting at NSW. Surveys were conducted monthly, between November 2021 and March 2022. All surveys started approximately one hour before sunset to permit a count of all roosting sites before it became too dark to accurately count the number of birds, although care was taken to avoid underestimating numbers on days when birds were still arriving at the start of the count. During the surveys, all groups of trees providing suitable conditions for roosting cormorants were observed and the total number of birds seen in each group of trees was recorded. Simultaneous counts were also conducted at the other Deep Bay roosts at MPNR, Lok Ma Chau (LMC) and a newly established roost at Mai Po San Tsuen (MPST) to ascertain the relative importance of the roosts and estimate the total Deep Bay population of this species.

4.3.12 To assess whether the proposed development would have the potential to cause indirect impacts to birds by impedance of flightlines, flightline surveys were conducted at NSW once per month from April 2021 and March 2022. These were conducted in the early morning, starting at sunrise, and lasted for a period of two hours. During the surveys, the abundance, estimated flight height and flight directions of all significant bird species (including Great Cormorant and Black-faced Spoonbill, ardeids, ducks, shorebirds and raptors) seen flying over the site were recorded and the flightline recorded onto a map of the site.

4.3.13 Egretry surveys at SPR egretry were conducted monthly between April and July 2022. Active nests, determined by the presence of nest-building adults, incubating adults and/or chicks, were counted using 8-10x binoculars. Flightline surveys were also conducted for birds leaving the SPR egretry to investigate the flightline directions to allow for potential impacts to flightline to be considered. During egretry flightline counts individual birds seen leaving the egretry were followed until they landed or were lost from view; these were marked on a basemap and estimated flight height was recorded. Surveys were conducted in the early morning, starting at sunrise, and lasted for a period of two hours. Monthly visits were conducted between April and July 2022 at the former TSL Egretry (which was last occupied in 2020) confirming that this egretry remained abandoned. A search for any additional newly established egretry within the Assessment Area was also conducted during the ardeid breeding season; none were found.

4.3.14 Wintertime ardeid night roost surveys at the SPR egretry were conducted monthly between November 2021 and March 2022. Surveys were conducted 30mins prior to sunset, the period when maximum abundance of ardeid were recorded at the roost.

Herpetofauna

4.3.15 Herpetofauna (amphibian and reptile) surveys were conducted at NSW and LC once per month from April 2021 and March 2022. As required in the EIA Study Brief, special reference was given to the potential presence of Common Rat Snake and Chinese Bullfrog. Both daytime and night-time surveys were conducted. A transect was followed during sunny weather (for daytime surveys, reptiles being more active in such conditions), recording all species of herpetofauna seen or heard. Potential habitats for amphibian and reptile were actively searched throughout the survey, including e.g. stones, crevices or rotten logs to investigate the presence of any amphibians and reptiles.

4.3.16 Active searching for eggs and tadpoles of amphibians in aquatic habitats was conducted to investigate breeding activity. All life forms of amphibians, including adult, tadpole, juvenile and egg, were recorded.

4.3.17 Amphibians are most easily located by recording breeding vocalisations, which are given most often during the evening, especially in the early wet season. During the surveys, all amphibian species heard calling were recorded and their abundance counted or estimated.

Odonates and Butterflies

4.3.18 Odonates (dragonflies and damselflies) and butterflies were surveyed simultaneously in daytime on a monthly basis from April 2021 and March 2022. Surveys were conducted in suitable weather to avoid cloudy, overcast, or rainy conditions when most odonates and butterflies are inactive. During the surveys, all species of odonates and butterflies were recorded, and their abundance counted or estimated. For butterflies, active searching for larvae and pupae within 5m of the transects was conducted. Particular effort was given to habitats which typically support greater abundance of odonata, such as watercourses and abandoned fishponds with emergent vegetation. Species occurring outside 5m wide transects but within the Assessment Area were also recorded.

4.3.19 Reedbeds, brackish marsh and mangroves in the Deep Bay area (including Mai Po and Sha Po) are known to provide habitat for Four-spot Midget Mortonagrion hirosei, a damselfly species which is listed as Near Threatened by IUCN (Wilson & Reels 2011). During the odonata surveys, specific attention was paid to the possible presence of this species.

Mai Po Bent-winged Firefly (MPBWF)

4.3.20 MPBWF surveys were conducted monthly from April to October 2021. Surveys started shortly after sunset and continued until 90 minutes after sunset, the period when the MPBWF are most active. All MPBWF observed were recorded and their locations were marked onto a map. Particular effort was given to mangrove and mangrove fringe areas as this species is associated with these habitats. These surveys covered the perimeter of NSW and LC, the mangroves along the KTMDC and the southern end of MPNR (Figure 4.12a). In addition, surveys were conducted by boat along the KTMDC and SPR to check for the presence of MPBWF on the seaward side of the mangroves. Surveys conducted in 2015 (AEC 2016) and 2019 (AEC 2020) shared the same survey methodology and effort. In addition to the dedicated MPBWF surveys, MPBWF were checked for during other night time surveys (Figure 4.3).

4.3.21 Light levels were measured at sites used by MPBWF. Measurements were taken using a light meter (FLUKE 941 Light Meter) directed vertically towards the sky and were measured to the nearest 0.01 lux. The number of MPBWF present was also recorded.

4.4 Description of the Environment

Recognised Sites of Conservation Importance and Ecologically Sensitive Resources

4.4.1 The following section describes all recognised sites of conservation importance within and in the vicinity of the Project Site in accordance with the requirements of Annex 16 of the TM. The sites described in this section are mapped on Figure 4.1.

Mai Po Inner Deep Bay Ramsar Site

4.4.2 The designation of Mai Po Inner Deep Bay Ramsar Site was accepted by the Hong Kong SAR Government in 1995, under the “Convention on Wetlands of International Importance especially as Waterfowl Habitat” (the Ramsar Convention). The Mai Po Inner Deep Bay Ramsar Site covers an area of about 1500ha and includes an area of intertidal mudflats and mangroves in Deep Bay as well as MPNR and some of the nearby fishponds. The mangroves and fishponds at LC are included within the Ramsar Site.

4.4.3 Management of the Mai Po Inner Deep Bay Ramsar Site is determined by a management plan maintained by AFCD. Under the management plan, the Ramsar Site is divided into zones, determining the management actions for the area. Most fishponds at LC are located within the Wise Use Zone (Figure 4.2), where the intention is “to allow ecologically sustainable use of wetland and other natural resources to be carried out in a way compatible with the Ramsar Site’s management goals”. Part of LC is included in the Private Land Zone, where the intention is to conduct “management in an ecologically sustainable manner consistent with the surrounding or adjacent management zones” (in this case the Wise Use Zone described above). Mangroves at the north of LC that are connected to the mangrove stand in the intertidal area of the MPNR are included in the Core Zone where the intention is “to provide an undisturbed reference area where the biological interest of the Ramsar Site is concentrated”.

Mai Po Marshes SSSI

4.4.4 Mai Po Marshes SSSI was designated in 1976 in recognition of the ecological importance of the mangrove communities and gei wai, and its importance for ducks and shorebirds. The SSSI includes the existing MPNR and the fishpond area at LC.

Inner Deep Bay SSSI

4.4.5 Inner Deep Bay SSSI was designated in 1986. Inner Deep Bay SSSI contains the largest and most important dwarf mangrove communities in Hong Kong and extensive natural intertidal mudflats. Both the dwarf mangroves and mudflats provide an important feeding and resting ground for waterbirds. Much of the mangrove at LC lies within this SSSI.

Mai Po Inner Deep Bay and Deep Bay Wetland outside Ramsar Site Priority Sites for Enhanced Conservation

4.4.6 The New Nature Conservation Policy (NNCP) of the HKSAR Government identified 12 priority sites for enhanced conservation, two of which are relevant to the current submission, namely the Ramsar Site and the Deep Bay wetland outside the Ramsar Site. Under the NNCP, two schemes, namely the Nature Conservation Management Agreement (MA) Scheme and the Public-Private Partnership (PPP) Scheme, were introduced to enhance the conservation of ecologically important sites, in particular those on land under private ownership. The PPP Scheme allows limited development on the ecologically less sensitive portion of a site (Developable Portion) provided that the landowner will, inter alia, formulate a conservation and management plan (CMP) to set out the long-term conservation programmes for the ecologically more sensitive portion of the site (Conservation Portion) and provide a lump sum contribution to support the long-term conservation work for the Conservation Portion, as well as comply with other requirements. With effect from 6 October 2021, an additional option became available, allowing landowners to surrender the Conservation Portion to the Government for proactive conservation and management by the Government. Under this option, landowners are still required to provide to the Government a lump sum contribution that is sufficient to generate recurrent income to support the long-term conservation work for the Conservation Portion.

4.4.7 The MA scheme encourages non-profit-making organisations (NPOs) to enter into management agreements with landowners and/or tenants to enhance ecological or nature conservation of the priority sites for enhanced conservation. Two MA projects for fishponds in Northwest New Territories have been in effect since January 2012; the area covered by these MA projects include both fishponds in LC and NSW; participation is at the discretion of individual fishpond operators. The fishponds in LC fall with the Mai Po Inner Deep Bay Ramsar Site while the fishponds in NSW fall within the Deep Bay Wetland outside Ramsar Site.

Wetland Conservation Area (WCA)

4.4.8 Wetland Conservation Area was designated for all existing continuous and adjoining active/abandoned fishponds in the Deep Bay Area. LC lies completely within the WCA while most of NSW also lies within the WCA. New development within the WCA would not be allowed unless it is required to support the conservation of the ecological value of the area, or the development is an essential infrastructural project with overriding public interest. Any such development should be supported by an ecological impact assessment to demonstrate that the development would not result in a net loss in wetland function and negative disturbance impact.

Wetland Buffer Area (WBA)

4.4.9 Wetland Buffer Area (WBA) of about 500m along the landward boundary of the WCA, was designated to protect the ecological integrity of the WCA. Residential/recreational development may be considered here, to remove the open storage use and/or to restore wetland lost. The southern edge of the NSW Project Site abuts the WBA, whilst the proposed bridge providing access to the proposed development falls within it (Figure 4.1).

Mai Po Marshes Restricted Area and Mai Po Nature Reserve

4.4.10 As is noted above, AFCD has formulated and implemented a Ramsar Site Management Plan (RSMP) to guide the conservation management of the Ramsar Site. The RSMP divides the Ramsar Site into four Management Zones with different management objectives based on habitat type, ecological value and land use. The four Management Zones are Core Zone, Biodiversity Management Zone, Wise Use Zone and Private Land Zone. The Core Zone and the Biodiversity Zone are the main zones covering key habitats of the Ramsar Site, namely the intertidal mudflat, mangals and gei wai. These two main zones are designated as a Restricted Area under the Cap.170, where access is prohibited, except those with permits issued by the AFCD. The strip of mangrove areas to the northwest of LC overlaps with the Mai Po Marshes Restricted Area. (Figure 4.1).

4.4.11 The Mai Po Nature Reserve (MPNR), which mainly comprises gei wai, falls within the Biodiversity Management Zone. AFCD collaborates with the World Wide Fund for Nature Hong Kong (WWFHK) in the conservation management of MPNR and provides funding to support the work of the WWFHK. LC is located 70m from Mai Po Nature Reserve (Figure 4.1).

Important Bird Area namely Inner Deep Bay and Shenzhen River catchment area

4.4.12 An Important Bird Area (IBA) namely Inner Deep Bay and Shenzhen River catchment (BirdLife International 2004) is an estuarine area comprising a variety of habitats, including freshwater wetland, marine-coastal (intertidal mudflats and mangroves) and man-made (aquaculture fishponds, tidal shrimp ponds (gei wai) and oyster farms) habitats. The freshwater wetlands are situated at Mai Po and at a flood plain area at the southern side of Shenzhen River namely Long Valley, this latter area is now managed as the Long Valley Nature Park by AFCD. The mudflats of inner Deep Bay are situated along the Shenzhen River, from the Mai Po/Tsim Bei Tsui peninsula on the Hong Kong side to Fu Tian on the Shenzhen side. A thick belt of coastal mangroves encloses these sites. The inland area consists mainly of farmland, fishponds, and tidal shrimp ponds. Surrounding and among these are residential and industrial areas. The IBA includes the entire NSW and LC, as well as Tai Sang Wai and Fung Lok Wai, which are within the 500m Assessment Area.

Conservation Area Zone

4.4.13 Within the 500m Assessment Area, wetland areas at Fung Lok Wai and Tai Sang Wai, ponds to the south of the former alignment of Kam Tin River including the watercourse, and the ponds between Pok Wai South Road and Pok Wai West Road, are zoned as Conservation Area (CA), which are gazetted under Approved Mai Po and Fairview Park Outline Zoning Plan No.S/YL-MP/8, Approved Nam Sang Wai, Outline Zoning Plan No. S/YL-NSW/9 and Approved Lau Fau Shan and Tsim Bei Tsui Outline Zoning Plan No. S/YL-LFS/11. The planning intention of CA zone is to conserve the ecological value of wetland and fishponds which form an integral part of the wetland ecosystem in the Deep Bay Area. The “no-net-loss in wetland” principle is adopted for any change in use within this zone. The primary intention is to discourage new development unless it is required to support the conservation of the ecological integrity of the wetland ecosystem, or the development is an essential infrastructure project with overriding public interest. LC adjoins the CA zone at Tai Sang Wai and NSW adjoins the CA zone covering the former alignment of Kam Tin River and the Yuen Long Bypass Floodway mitigation wetland.

Yuen Long Bypass Floodway mitigation wetland

4.4.14 Yuen Long Bypass Floodway (YLBF) mitigation wetland (AEIAR-003/1998) falls within the southern fringe of the Assessment Area. This engineered wetland was an ecological compensatory measure that addressed the loss of fishponds and agricultural land under the YLBF project. The construction of this engineered wetland and channel shallow pond commenced in January 2003 and was completed at the end of 2005. It was handed over to Agriculture, Fisheries and Conservation Department (AFCD) in December 2006 after establishment by DSD and is currently managed and maintained by AFCD.

Egretries

4.4.15 There was one active egretry in the Assessment Area in 2022, the Shan Pui River egretry, and one former egretry, the Tung Shing Lane egretry, which was last occupied in 2020.

4.4.16 The SPR egretry, which was first recorded in 2019 (Anon 2020), is in Chinese Banyan Ficus microcarpa trees on the west side of the SPR at its confluence with the KTMDC, north of the Yuen Long Industrial Estate (Figure 4.1). It has largely been used by Chinese Pond Herons, but Little Egrets also bred at the egretry in 2019 and 2020. Data are presented in Table 4.2 for the number of birds breeding at the egretry during 2019-2022.

Table 4.2: Number of ardeid nests in the Shan Pui River egretry during 2019-2022

Species	2019	2020	2021	2022
Little Egret	5	10	0	0
Chinese Pond Heron	86	129	151	38
Total	91	139	151	38

Data from Anon (2020, 2021a, 2021b, 2022)

4.4.17 The former Tung Shing Lane egretry was first noted in 2001. It was largely utilised by Chinese Pond Herons and Little Egrets, with small numbers of Eastern Cattle Egrets in some years. It was the fifth largest egretry in Hong Kong in 2015 and held a peak number of 84 nests of Little Egrets and Chinese Pond Herons in 2018. The number of nests declined thereafter, and it was last occupied in 2020. The reasons for this decline are not certain; Anon (2021) suggested that the condition of the nesting trees was adversely affected by Typhoon Mangkhut in 2018, and the nesting birds may have relocated to the Kam Po Road egretry, 1.9 km to the south from 2021. Whilst it is not impossible that the former egretry might be used again in the future, experience from other abandoned egretries in Hong Kong suggests that this is highly improbable. Historical counts of ardeid nests at this former egretry are detailed in Appendix 4.2a.

Shan Pui River Ardeid Night Roost

4.4.18 The SPR ardeid night roost appears to have become established after the natural expansion of mangrove area near the intersection of SPR and KTMDC; it is located inside this mangrove area, which has largely developed since 2020. Species composition and abundance of roosting ardeids varies seasonally.

NSW Great Cormorant Night Roost

4.4.19 NSW contains a large overnight roost of wintering Great Cormorants, one of four in the Deep Bay area of Hong Kong, the cormorants mainly roost on large mature Eucalyptus trees. The roost is divided into two sub-roosts at the northern and southern ends of NSW.

4.4.20 DSD (2021) referred to a Great Cormorant ‘day roost’ at NSW where Great Cormorants were recorded loafing after foraging in the river channels. Daytime loafing sites of Great Cormorant are found across the Deep Bay wetland area, covering multiple habitat types. These are highly related to preferred foraging areas and the state of tides and water levels in ponds, and as such are ephemeral in nature. Daytime loafing sites are therefore not considered to be sensitive resources in the same manner as are night roosts which are used consistently throughout the winter and across multiple winters.

Black-winged Stilt High Tide Roost

4.4.21 A Black-winged Stilt high tide roost was recorded (DSD 2021) at the western hard bank of the channel, close to the Yuen Long Industrial Estate. It was located approximately 250m away from the proposed Development Site. The site has now been developed and no longer exists, hence is not considered further in this Assessment.

Shan Pui River, former alignment of Kam Tin River and KTMDC

4.4.22 The SPR originates from Tai Lam Country Park, passing through villages and Yuen Long Town, into the channelised section where it is under tidal influence between Yuen Long Industrial Estate and NSW, and which falls within the 500m Assessment Area. The SPR intersects with the KTMDC and eventually feeds into Deep Bay.

4.4.23 Kam Tin River originates from Tai Mo Shan, flowing into streams and tributaries, eventually feeding the KTMDC and Deep Bay. Part of the former alignment of Kam Tin River remains along the south of NSW but is no longer connected to the KTMDC. The KTMDC was constructed in 1997 to solve the flooding problem in the Kam Tin area. The channelised section, where it is under tidal influence, between Tai Sang Wai and NSW, falls within the 500m Assessment Area.

4.4.24 The channel design retains the original natural riverbed, providing habitats similar to the original wetland environment. The soft-mud bottoms of the SPR and KTMDC are exposed under low tide conditions, providing foraging habitats for waterbirds. However, the open areas of both channelised watercourses are reducing in area due to the expansion of mangrove (especially exotic species).

4.5 Ecological Baseline Conditions

Literature Review

4.5.1 A desktop literature review of the ecological (and relevant fisheries) baseline conditions was carried out upon commencement of the study. Findings of relevant studies or surveys, including but not limited to relevant Environmental Impact Assessment reports, publications related to local ecology and conservation (such as ‘Hong Kong Biodiversity’ (AFCD) and ‘Connections’ (Kadoorie Farm and Botanic Garden (KFBG)), published scientific papers, as well as the Biodiversity Database and other publications on the conservation status and distribution of local flora and fauna etc. were reviewed (see Table 4.3 below).

Table 4.3: Baseline Information on Ecological Resources in the Assessment Area

Relevant Literature	Habitat and Vegetation	Avifauna	Mammal	Herpetofauna	Butterfly and Odonate	Firefly
Ecological Impact Assessment Report submitted to the Town Planning Appeal Board for the same Project (AEC 2016)	ü	ü	ü	ü	ü	ü
*Current population, distribution and habitat preferences of Mai Po Bent-winged Firefly Pteroptyx maipo* in the Inner Deep Bay area of Hong Kong (AEC 2020)**						ü
Hong Kong Biodiversity Information Hub (AFCD 2022)		ü	ü	ü	ü
Yuen Long Effluent Polishing Plant (AEIAR-220/2019)		ü	ü	ü	ü
Yuen Long Barrage Scheme (AEIAR-228/2021)		ü	ü	ü	ü
Mai Po Inner Deep Bay Ramsar Site Waterbird Monitoring Programme, Egretry Counts in Hong Kong, Summer Report 2011-2022 (Anon 2021 to 2022)		ü
Mai Po Nature Reserve Management Plan: 2019-2024 (WWF, 2021)	ü
MTRC Contract No. M1016-09C Ecological Monitoring and Adaptive Management Advice Services for Lok Ma Chau and West Rail Wetlands. Report on Ecological Monitoring and Management, and Works Programme		ü
Mai Po Nature Reserve Infrastructure Upgrade Project (AEIAR-233/2022)		ü
Monthly Waterbird Monitoring Winter Report 2021-2022. Mai Po Inner Deep Bay Ramsar Site Waterbird Monitoring Programme, Monthly Waterbird Monitoring		ü
Monthly Waterbird Monitoring Summer Report 2021. Mai Po Inner Deep Bay Ramsar Site Waterbird Monitoring Programme, Monthly Waterbird Monitoring		ü
Construction of Cycle Tracks and the Associated Supporting Facilities from Sha Po Tsuen to Shek Sheung River (AEIAR-133/2009)	ü	ü
Yuen Long Bypass Floodway (AEIAR-003/1998)	ü	ü
The List of Wild Plants and Wild Animals Under Special State Protection under the Law of the People’s Republic of China on the Protection of Wildlife		ü	ü	ü

4.5.2 All available information was collated and evaluated to identify any information gaps relating to the establishment of the ecological profile of the aquatic and terrestrial environment, and to determine the ecological surveys needed for an ecological impact assessment.

4.5.3 In the literature review, reference was made to the Ecological Impact Assessment Report submitted to the Town Planning Appeal Board for the same Project (AEC 2016), which detailed findings from comprehensive ecological field surveys and studies carried out from 2011 to 2015. These surveys included a detailed study of MPBWF, including its population, distribution and habitat preferences (hereafter referred to as AEC (2020). The report of that study is attached here as Appendix 4.4. Ecological surveys were also undertaken longer ago by Aspinwall (1997) and CH2M Hill (2006) in NSW. These included surveys of birds, mammals (including small mammal trapping), herpetofauna, odonates and aquatic invertebrate fauna. However, these surveys were carried over 18 years ago, hence the findings from these surveys are considered obsolete, and are not reviewed in this assessment.

Habitat and Vegetation

4.5.4 The following section presents and reviews the general findings from AEC (2016) on habitat and vegetation in the Project Site and its environs. The review focuses on findings which differ from those from the recent (2021-22) surveys.

4.5.5 NSW mostly comprised active and abandoned fishponds (some of which had been colonised by reedbed or grassland), while other habitats present (largely or around the perimeter of the site) included mangrove, watercourses, terrestrial grassland, plantation, temporary structures (houses and other buildings) and other developed areas (roads).

4.5.6 Active fishponds at NSW were maintained with mostly open water and little emergent vegetation, whilst abandoned fishponds had been progressively colonised by vegetation to varying extents, so that some retained open water while others were overgrown with grasses, reeds Phragmites australis and/or reedmace Typha angustifolia. Some were connected to nearby channels via sluices or damaged bunds, and as a result the water within these was tidal and brackish. Bund vegetation was mostly grassy, dominated by Brachiaria mutica and Panicum maximum, whereas some bunds supported groups of trees, including tall individuals of Eucalyptus spp., and isolated individuals of other tree species (such as Macaranga tanarius var. tomentosa, Mangifera indica and Melia azedarach). Overall, this pattern is not dissimilar to that found in 2021-22, albeit there was more open water in some of the abandoned ponds and the areas of exotic Typha and Cyclosorus fern dominated marsh appear to have been less extensive.

4.5.7 Terrestrial habitats within, or immediately adjacent to, the site included grassland and plantation. Grassland was dominated by common grass species, including Cynodon dactylon, Chloris barbata and Eleusine indica. Plantation along NSW Road comprised common plantation tree species (especially Casuarina equisetifolia). Grassland and plantation were regularly disturbed by human recreational activity. A small area of secondary woodland was present to the south of Shan Pui Tsuen and Shan Pui Hung Tin Tsuen, this area was also heavily influenced by human activity.

4.5.8 Habitats at LC were less diverse and comprised active fishponds, with associated houses and other temporary structures, intertidal mangroves and small areas of tidal watercourse and plantation. The vegetation was typical of fishponds in the Deep Bay, with simple vegetation structure and low vegetative diversity dominated by grasses (such as Brachiaria mutica and Panicum maximum) (AEC 2016). Many of the bunds were maintained by the import of hardcore fill material and consequently supported limited vegetation.

4.5.9 Surrounding the fishponds at LC was an area of intertidal mangrove, some of which is contained within the Site Boundary. This was continuous with the mangrove surrounding Deep Bay and supported a similar diverse mixture of mangrove shrub species.

4.5.10 No plant species of conservation importance were recorded at either NSW or LC.

Mammals

4.5.11 Five terrestrial mammal species (three wild and two domestic/feral) were recorded at NSW by camera trapping and direct observation during a study period from 2011 to 2015 AEC (2016): Small Indian Civet, Small Asian Mongoose, Leopard Cat, Domestic Dog and Domestic Cat. Brown Rat (an exotic pest species which is commensal with man and is of no conservation importance) and Domestic Dog were the only mammal species recorded at LC.

4.5.12 Although not recorded during the surveys in AEC (2016), McMillan et al. (2019) mapped Eurasian Otter in the southern NSW area (although mapped at a scale that makes it impossible to determine if the record(s) were within the Project Site). AFCD has also recorded Eurasian Otter between 2012 to 2017 at the northern tip of NSW (AFCD biodiversity survey data (unpublished)). The most recent published record from NSW is from 2011 and there appear to be no records whatsoever from LC, despite there being records from adjacent areas at Tai Sang Wai and especially the Mai Po Nature Reserve (Li & Chan, 2017). The lack of records at LC is attributed to human disturbance and the presence of Domestic Dogs.

4.5.13 No dedicated bat surveys were conducted by AEC (2016), but bats were recorded during night-time surveys for other species groups. Call structure and frequency recorded by use of bat detectors indicated that these were Japanese Pipistrelle, a species which is very common in lowland Hong Kong. No literature was found to suggest significant bat roosts occur within either LC or NSW.

Avifauna

4.5.14 A total of 138 bird species was recorded at NSW in 2011-2014 (AEC 2016), including 57 species of conservation importance; the diversity of LC was lower, with a total of 79 bird species, including 38 species of conservation importance.

4.5.15 AEC (2016) found that numbers of ardeids and Black-faced Spoonbills were boosted on occasion by large numbers present during fishpond draindown at both NSW and LC; similarly, a flock of 125 Red-necked Stints (a passage migrant shorebird that largely feeds on the intertidal mudflats) was attracted to a drained fishpond at LC in May 2011.

4.5.16 AEC (2016) found the reedbed habitat at NSW to be of importance for a suite of bird species that favours or is often associated with this habitat including Purple Heron, Yellow Bittern, Eurasian Bittern, Eastern Marsh Harrier, Black-browed Reed Warbler, Oriental Reed Warbler and Chinese Penduline Tit.

4.5.17 Whilst noting that the avifauna of LC was less diverse than that of NSW, AEC (2016) did observe that LC regularly held flocks of Collared Crow, a globally Vulnerable species (BirdLife International 2024). The species was recorded at LC on 24 of the 25 surveys, with a peak count of 30 individuals in August 2012. Data from the monthly waterbird counts also show that Collared Crow is regularly present at LC in significant numbers. The Deep Bay population of this species is around 200 individuals (Stanton et al. 2014) and is probably of global importance. Collared Crows were also seen in smaller numbers at NSW, and it is likely that the undisturbed tall trees present at NSW provide suitable breeding locations for this species.

4.5.18 Other globally threatened species recorded by AEC (2016) included Black-faced Spoonbill, Greater Spotted Eagle, Eastern Imperial Eagle and Yellow-breasted Bunting. All these species occur regularly in the Deep Bay area during migration and/or in the winter.

Birds Using Intertidal Shan Pui River and KTMDC

4.5.19 Wetland birds were recorded in large numbers in the KTMDC and SPR adjacent to the site in AEC (2016), with peak counts of 4435 at high tide and 5495 at low tide, both during December 2011. Waterbird utilisation of the channels was not even, with the most heavily utilised channel sections being the downstream and wider sections. Numbers in the downstream and wider sections differed little between high and low tides, however in the narrower upstream sections numbers were significantly higher at low tide. Some species were recorded in higher numbers at high tide and others at low tide, no doubt a consequence of different species having different foraging and roosting strategies. Species recorded in notable numbers relative to the Deep Bay populations in the KTMDC and SPR included Northern Shoveler, Eurasian Teal, Tufted Duck, Grey Heron, Great Egret, Little Egret, Chinese Pond Heron, Black-faced Spoonbill, Black-winged Stilt, Pied Avocet, Marsh Sandpiper, Common Greenshank and Black-headed Gull.

Roosting Birds at NSW

4.5.20 AEC (2016) detailed counts undertaken at the Great Cormorant roost at NSW during the winters of 2010-11 to 2014-15 (see Appendix 4.2a); the peak count of Great Cormorants roosting at NSW during this period was of 7,204 birds during the 2011-12 winter. As in more recent years, marked inter and intra-season changes in the relative numbers using the NSW and MPNR roosts were found. Again reflecting the pattern noted in the 2021-22 surveys undertaken for this Assessment, AEC (2016) described how there were two main roost areas at the north and south of NSW; the changes in the number of birds using different parts of the roost suggested that individual birds may change their roost location during the winter; this was supported by reported findings from a radio-telemetry study undertaken in 2011-12 (Ma 2014), in which the single tracked individual was recorded roosting mostly at NSW but sometimes changed to roost at MPNR.

4.5.21 AEC (2016) provided details of the flightlines of Great Cormorants leaving the roost towards their feeding areas. These showed that the majority of birds departed towards the north, north-west or north-east, towards Deep Bay, Fung Lok Wai, LC and MPNR. Some birds were recorded departing to the south or south-east over Tin Fook Wai. Very few birds departed towards the west or south-west (towards Yuen Long and the Industrial Estate), probably because there was (and is) little suitable foraging habitat in that direction.

4.5.22 Egrets were observed roosting overnight at NSW regularly during early 2012 AEC (2016). Approximately 300 egrets were observed leaving the roost on 3 February 2012 and 192 egrets (comprising 109 Great Egrets and 83 Little Egrets) were observed leaving the roost on 8 February 2012. However, according to AEC (2016) this does not appear to have been used consistently as a night roost prior to winter 2011/12 or in subsequent winters, suggesting that these observations are largely of historical interest.

4.5.23 Other bird species were also reported forming communal roosts around the abandoned fishponds at NSW in AEC (2016), especially in reedbeds; species recorded roosting in the reedbeds included Crested Myna (200 entering roost on 15 February 2012), White-cheeked Starling (280 on 15 February 2012), Red-billed Starling (120 on 15 February 2012), White-shouldered Starling (20 on 15 February 2012), Eastern Yellow Wagtail and Barn Swallow. Greater Spotted Eagle and Eastern Imperial Eagle were also occasionally recorded roosting on trees during evening cormorant counts.

Bird Flightlines and Flight Paths over NSW and LC

4.5.24 Flightline surveys in AEC (2016) recorded moderate numbers of waterbirds flying over NSW and Tin Fook Wai throughout the year. Large numbers of cormorants were observed leaving the night roost on surveys conducted during winter. Some ardeid species were regularly recorded (especially Great and Little Egrets) and flocks of ducks were also recorded flying over the site occasionally during the winter months.

4.5.25 During winter surveys in AEC (2016) the direction of flight paths did not indicate any preferred direction of travel for the local bird population, but rather showed the movement of birds between different sites used for foraging and/or roosting. Thus, no regularly used and well-defined flightlines were identified for wintering birds except for those of Great Cormorants leaving the night roost.

Former Tung Shing Lane egretry

4.5.26 Egretry surveys in AEC (2016) recorded the flightlines to and from this former egretry during the 2014 breeding season; a north-south flightline was identified at the east of NSW. As this egretry has not been utilised since 2020 these observations are now only of historical interest.

Herpetofauna

4.5.27 Nine amphibian species were recorded at NSW during surveys reported in AEC (2016). Most of these were present in very low numbers, although moderate abundance of Günther’s Frog was recorded. All species are common throughout Hong Kong and none of the species is of conservation importance. Only Günther’s Frog was recorded at LC during these surveys (AEC 2016).

4.5.28 Three reptile species were recorded during systematic herpetofauna surveys at NSW reported in AEC (2016): Bowring’s Gecko, Long-tailed Skink and Red-eared Slider. None of these species are of conservation importance. Three other reptile species, Reeves’ Turtle, Chinese Gecko and Burmese Python, were recorded during other surveys of which two are of conservation concern. Reeves’ Turtle was seen at NSW in April 2011, this species is globally Endangered, IUCN (2024). Burmese Python was seen twice during December 2011; the global population is Vulnerable, IUCN (2024).

Odonates

4.5.29 A moderately diverse odonate community, with 19 species recorded during 2011 and two additional species found in 2014 was reported by AEC (2016). The species recorded from the NSW Project Site are widespread in lowland wetlands in Hong Kong. The only species listed in Fellowes et al. (2002) were Ruby Darter and Scarlet Basker, however, both are common in Hong Kong. Despite searches for the species, Four-spot Midget was not recorded in the reedbeds or other habitats at NSW.

4.5.30 In contrast to NSW, only eight odonate species were recorded at LC, including one species, Amber-winged Glider, not recorded at NSW (AEC 2016). The lower diversity at LC is likely to result from differences in habitat, especially the relative lack of emergent vegetation. Coastal Glider was found at nearby Tai Sang Wai in the Assessment Area, and it could potentially occur at LC on occasion; this species is listed as being of Local Concern in Hong Kong by Fellowes et al. (2002) but is common in Hong Kong.

Butterflies

4.5.31 The diversity and abundance of butterflies recorded in two seasons of surveys in 2011 and 2014 were both low at NSW, where 29 species were recorded, and exceptionally low at LC, where only two species were recorded, albeit in only a single survey season in 2011 (AEC 2016).

4.5.32 Two species of conservation importance were recorded at NSW during these surveys (AEC 2016). Two individuals of Pale Palm Dart were seen in July 2011. This is a grassland and shrubland species found in a number of Country Parks in Hong Kong (Lo and Hui 2010). A single individual of the migrant Painted Lady was seen in April 2014. This is another open country species. In addition, one species, Grass Demon, not listed by Fellowes et al. (2002) but which is Rare in Hong Kong (Chan et al. 2011, AFCD 2022), was recorded in July 2014. This species is wetland-associated as the larval food plants are gingers Zingiberaceae.

4.5.33 In general, however, the habitat types present, supporting a low diversity of native plant species and relatively low vegetative structural diversity, were not considered suitable for most butterflies, providing few potential sources of food for adults or larvae.

MPBWF

4.5.34 The MPBWF was first described in 2011 from Hong Kong (Ballantyne et al. 2011, Yiu 2011), and was subsequently found in western Guangdong and Hainan (Fu 2014, Fan and Fu 2017, AFCD 2022). The species appears to be restricted to the vicinity of intertidal mangrove habitats (Cheng et al. 2010, Ballantyne et al. 2011, Yiu 2011). It was recorded in the mangrove between MPNR and LC, and lower numbers of MPBWF were also recorded in the mangrove at the KTMDC (Yiu 2011). Detailed studies of its habitat use, and requirements were conducted by AEC to provide further information in the Town Planning Appeal (AEC 2020).

4.5.35 The following sections present the findings from AEC (2016) on the MPBWF in the Project Site and its environs.

4.5.36 Concentrations of MPBWF were recorded in mangrove and wet brackish grassland along the SPR (up to 120 individuals) and around the southern end of MPNR (up to 187 individuals) (AEC 2016). Moderate numbers were recorded in the channel to the east of LC (up to 27 individuals), but very few were recorded on the northern and western side of LC, despite the large numbers on the opposite side of the channel at MPNR. Small numbers of the MPBWF were observed in the mangroves along KTMDC and SPR. No MPBWF were observed in other habitats at NSW or LC.

4.5.37 Light levels recorded at sites used by MPBWF in May 2015 were in the range of 0.01-0.49 lux. Light levels measured at other sites where no MPBWF were recorded ranged from 0.01-0.27 lux. Weather conditions seemed to affect the light level recorded, with higher light levels recorded on overcast nights than on clear nights, presumably because of light from nearby developments reflected off overhead clouds.

Table 4.4: Peak Counts of MPBWF recorded at NSW and LC, 2012 to 2015

Species	NSW (Shan Pui River)	LC	South of MPNR	KTMDC
MPBWF	120	27	187	<10

4.6 Ecological Baseline Conditions in 2021-22

4.6.1 The ecological baseline conditions, as determined by the surveys undertaken in fulfilment of the requirements of the SB in 2021-22, of the two parts of the Project Site, NSW and LC, and the remainder of the Assessment Area, are described below.

NSW

4.6.2 NSW comprises an area of fishponds, including fishponds which are actively managed for fish farming activities as well as fishponds which have been abandoned. In addition to the active and more recently abandoned fishponds, NSW also contains a number of fishponds which were abandoned many years ago and which have undergone ecological succession due to siltation and colonisation of vegetation.

4.6.3 Some of these former fishponds now support permanently wet reedbeds, but with ongoing siltation and vegetation litter deposition, some reedbeds are only seasonally wet. Many of these reedbeds have been invaded by the invasive exotic Typha angustifolia, together with terrestrial vegetation and even seedling trees, such as Eucalyptus spp. and Leucaena leucocephala. A particular feature of NSW is the presence of large, mature Eucalyptus trees, which form an unusual landscape element in a Hong Kong context and provide an important roost for Great Cormorants.

4.6.4 NSW lies at the intertidal confluence of the Kam Tin Main Drainage Channel (KTMDC) (to the east) and the Shan Pui River (SPR) (to the west). Both rivers have been channelised with construction completed in 1997. The former alignment of the Kam Tin River is located to the south of NSW. To the west of the realigned SPR lies the Yuen Long Industrial Estate. The area to the east of KTMDC and to the south of the former alignment of the Kam Tin River is mostly comprised of active fishponds, although small, developed areas such as village or other uses are also present.

4.6.5 The area surrounding NSW was formerly active fishponds, but these were filled over the course of the channelisation works and it is now largely grassland and plantation woodland. This area is currently subject to significant levels of human disturbance, especially from regular use by large numbers of photographers, cyclists, and model aircraft hobbyists. Part of NSW which is occupied by active fishponds is also subject to disturbance from fish farming activities and from the residents of on-farm domestic structures. A small area of natural mangrove and Brachiaria marsh is located at the southwest of NSW beside the SPR channel, while the KTMDC is lined by mangrove. To the southeast of NSW is a large area of fishponds.

LC

4.6.6 LC is an island surrounded by intertidal habitats. To the west is the intertidal mouth of KTMDC and SPR, while the rest of the site is surrounded by intertidal, mangrove-lined creeks. At the north of LC, an area of intertidal mangroves is contiguous with the extensive mangrove areas of MPNR, which together form the largest stand of intertidal mangrove in Hong Kong. It is also the sixth largest protected mangrove area in China (WWF 2006).

4.6.7 LC largely comprises fishponds which are actively managed for fish production. While fish farming activities at LC are superficially similar to those practiced elsewhere in the Deep Bay area, practices not typical of fish farming elsewhere have been observed during the survey period (2021-2022). These activities include dumping of waste (including inorganic materials and construction waste), suction dredging of fishponds into the adjacent creeks and extension of fishponds at the expense of mangrove habitat – the latter practice has resulted in the loss of approximately 4.1ha of mangrove since 2005 (see Appendix 4.5 Annex B). In addition to such activities, habitats at LC are highly disturbed due to the presence of numerous domestic structures and equipment stores, as well as by farmers, other residents and many unrestrained dogs and cats.

4.6.8 There was extensive discussion during the TPAB hearings regarding the condition of LC and NSW. In their decision the Appeal Board stated of Nam Sang Wai (page 36) that it was ‘picturesque but in limbo from lack of proper management and upkeep’. They had far more to say about the condition of LC, noting ‘There was much dumping of waste including construction waste, several abandoned cars (and a minibus), and all manner of waste’; ‘The situation was most disappointing and shocking, especially for a Ramsar Site’; and’…the sorry state of the LC Site indicates that as a matter of common sense and fact, the Site is functioning well below its ecological potential, there is significant room for enhancement of ecological value and functions’. Subsequently, on p. 87, it also noted ‘From the Site visit and we so find, the conditions concerning dumping and ecological deterioration at the LC Site were ‘serious’ or ‘very serious’.

Survey Findings

Habitats and Vegetation

4.6.9 A habitat map of the 500m Assessment Area and the Project Site is provided in Figures 4.4, 4.4a, 4.4b, 4.4c and 4.4d. A total of 16 habitat types were identified within the Assessment Area. Soft mud exposed as channel bed under low tide conditions was classified as part of tidal watercourse.

4.6.10 Most habitat types in both the Project Site and Assessment Area are wetland. Except for the developed areas of Yuen Long and Fairview Park, areas of non-wetland habitats are small and fragmented. The small area of secondary woodland to the south of Shan Pui Tsuen and Shan Pui Hung Tin Tsuen recorded in AEC (2016) is included in the plantation area in the current study as it was heavily influenced by human activities. In addition, the Assessment Area includes a part of MPNR. Areas of habitats present are detailed in Table 4.5. Photographs showing the typical views of habitats identified are provided in Appendix 4.3. A list of plant species recorded at NSW and LC during the surveys is provided in Appendix 4.1a and 4.1b. Survey findings of the focused vegetation survey conducted in vegetated wetland habitats at NSW are presented in Appendix 4.1c.

Table 4.5: Habitats (ha) present in the Project Sites and the 500m Assessment Area

Habitat	Project Site		500m Assessment Area (Excluding Project Site)	Total (ha)
Habitat	NSW	LC	500m Assessment Area (Excluding Project Site)	Total (ha)
Tidal Watercourse	0.4	0.2	76.8	77.5
Fishpond	22.0	36.0	119.6	177.7
Open Water	2.6	-	14.2	16.8
Reedbed	41.0	-	8.5	49.4
Brachiaria Marsh	7.0	-	-	7.0
Cyclosorus Marsh	4.1	-	-	4.1
Typha Marsh	9.8	-	-	9.8
Grassy Bund	7.3	9.2	66.4	82.8
Wooded Bund	9.5	1.0	0.5	11.0
Bare Soil	0.8	0.2	0.9	1.8
Mangrove	2.6	8.5	118.2	129.3
Intertidal Mudflat	-	-	3.2	3.2
Grassland	0.8	-	7.2	8.1
Managed Grassland	1.3	-	-	1.3
Plantation	9.7	-	25.5	35.2
Developed Area	3.1	0.4	107.1	110.8
Total	121.9	55.4	548.3	725.8

Note: Habitat areas are rounded to the nearest 0.1ha, thus the totals may not match the sum of the breakdowns due to rounding. Part of the proposed bridge falls outside of the Project Site but is shown on all relevant plans and included within all relevant assessments. The Proposed Visitor Centre and Public Park fall within the Project Site.

4.6.11 No plant species of conservation importance were recorded at either NSW or LC. Descriptions of each habitat type and the evaluation of habitat value are included in Section 4.7.

Mammals

4.6.12 A total of nine species was recorded in the 500m Assessment Area including the Project Site. Of these, five species were recorded in the 500m Assessment Area (excluding the Project Site), two of which were domestic/feral dog and cat, one was an invasive exotic pest (Brown Rat) and one was of conservation importance (Japanese Pipistrelle). Nine species were recorded within NSW, including four species of conservation importance. Five species were detected by camera trapping: Small Indian Civet, Small Asian Mongoose, Leopard Cat, Eurasian Wild Pig and Domestic Dog. In addition, Brown Rat and Domestic Cat were recorded during night-time surveys, and Pallas’s Squirrel was seen during daytime bird surveys. Three species were recorded within LC: Brown Rat, Domestic Dog and Domestic Cat. None of these are species of conservation importance. No suitable locations for camera trapping were found at LC due to the high levels of human activity.

4.6.13 Bat surveys were conducted during night-time surveys for other species groups. Call structure and frequency recorded by use of bat detectors indicated that these were Japanese Pipistrelles, which are very common in lowland Hong Kong. Japanese Pipistrelles were recorded along Yau Pok Road, NSW Road, and the pier to the south of NSW. No evidence was found to suggest significant bat roosts occur within either LC or NSW.

4.6.14 Survey findings are similar to those of AEC (2016), albeit Pallas’s Squirrel appears to be a recent arrival at NSW. This species has become more widespread in the Northwest New Territories in recent years (M.R. Leven pers. obs.). Eurasian Otter was not recorded during surveys. A full list of all species recorded at NSW and LC together with the maximum count of each species recorded is provided in Table 1 of Appendix 4.2b.

Table 4.6: Mammal species recorded at NSW, LC and elsewhere in the Assessment Area during Apr 2021 to Mar 2022

Species	NSW	LC	500m Assessment Area (excluding Project Sites)
Pallas’s Squirrel	ü		ü
Brown Rat	ü	ü	ü
Small Indian Civet^	ü
Small Asian Mongoose^	ü
Domestic Dog	ü	ü	ü
Leopard Cat^	ü
Domestic Cat	ü	ü	ü
Eurasian Wild Pig	ü
Japanese Pipistrelle^	ü		ü
^species of conservation importance

Avifauna

4.6.15 During the surveys conducted from April 2021 to March 2022, 152 species were recorded in the Assessment Area as a whole, including 73 species of conservation importance. Of these, 122 species (55 species of conservation importance) were recorded at NSW; 107 species (48 species of conservation importance) were recorded at LC; and 79 species (37 species of conservation importance) were recorded within the 500m Assessment Area excluding the Project Sites.

4.6.16 Observations of the species of conservation importance are summarised and the maximum count of each species at NSW and LC and the 500m Assessment Area excluding the Project Sites are detailed in Table 4.7 below.

4.6.17 Survey findings are similar to the findings of AEC (2016). A full list of all species recorded at NSW and LC together with the maximum count of each species recorded by habitat during avifauna transect surveys is provided in Table 2 of Appendix 4.2b.

Table 4.7: Maximum count of bird species of conservation importance recorded at NSW and LC Project Sites and the 500m Assessment area excluding the Project Sites, April 2021 – March 2022

Species	NSW	LC	500m Assessment Area (excluding Project Sites)
Northern Shoveler	40	35	79
Eurasian Wigeon	-	300	16
Northern Pintail	-	20	1
Eurasian Teal	45	12	33
Tufted Duck	12	1	36
Little Grebe	15	15	3
Eurasian Spoonbill	1	-	-
Black-faced Spoonbill	8	3	8
Yellow Bittern	2	-	-
Black-crowned Night Heron	13	3	3
Striated Heron	1	1	-
Chinese Pond Heron	7	12	36
Eastern Cattle Egret	12	52	-
Grey Heron	8	19	17
Purple Heron	3	-	-
Great Egret	15	124	21
Intermediate Egret	1	28	1
Little Egret	185	263	31
Great Cormorant	128	43	47
Western Osprey	1	1	1
Black-winged Kite	3	-	-
Crested Serpent Eagle	1	-	-
Greater Spotted Eagle	1	-	-
Besra	1	-	-
Eastern Marsh Harrier	-	1	-
Pied Harrier	1	-	-
Black Kite	3	11	3
Eastern Buzzard	1	1	-
Eurasian Coot	1	1	-
Black-winged Stilt	3	11	131
Pied Avocet	16	90	71
Pacific Golden Plover	-	-	3
Grey Plover	-	-	3
Little Ringed Plover	11	15	5
Whimbrel	-	-	1
Black-tailed Godwit	-	1	4
Curlew Sandpiper	-	1	-
Temminck’s Stint	1	10	-
Long-toed Stint	-	1	2
Red-necked Stint	-	7	-
Terek Sandpiper	-	-	1
Common Redshank	4	6	54
Marsh Sandpiper	13	19	14
Wood Sandpiper	3	6	2
Spotted Redshank	1	-	35
Common Greenshank	1	2	18
Oriental Pratincole	-	1	-
Black-headed Gull	46	600	135
Lesser Black-backed Gull	-	-	2
Greater Coucal	2	2	-
Lesser Coucal	-	2	-
Collared Scops Owl	1	-	1
Asian Barred Owlet	1	-	-
White-throated Kingfisher	6	2	2
Pied Kingfisher	5	4	1
Common Kestrel	1	-	-
Peregrine Falcon	1	-	-
Ashy Drongo	2	-	-
Collared Crow	5	14	2
Chinese Penduline Tit	60	-	-
Thick-billed Warbler	1	-	-
Pallas’s Grasshopper Warbler	4	-	-
Lanceolated Warbler	1	-	-
Zitting Cisticola	1	15	-
Red-billed Starling	25	104	-
White-cheeked Starling	8	120	-
White-shouldered Starling	6	10	8
Common Starling	-	10	-
Bluethroat	1	-	-
Red-throated Pipit	-	20	-
Buff-bellied Pipit	-	1	-
Chestnut-eared Bunting	1	-	-
Yellow-breasted Bunting	-	2	-

Birds in the NSW and LC Project Sites (Foraging, Roosting and Breeding)

4.6.18 There was more bird species recorded in NSW (especially wetland-dependent and wetland-associated species foraging or roosting at fishponds and reedbeds), as compared to the remainder of the Assessment Area. Notable species, either because of their numbers relative to the Deep Bay or Hong Kong populations or because of their conservation importance, included raptors (albeit all in small numbers), ardeids in and around fishponds, and Chinese Penduline Tit in reedbeds. Other than a communal night roost used by ardeids at the SPR and the long-established Great Cormorant roosts (see following sections), no other communal roosting sites of birds were recorded at NSW. Further details of foraging and roosting behaviour of wetland-dependent and wetland-associated birds in NSW Site is provided in Table 3 of Appendix 4.2b.

4.6.19 In LC, which is dominated by fishpond habitats, relatively high numbers of wetland birds were recorded foraging or roosting, notably Wigeon, ardeids, Pied Avocet and Black-headed Gull, together with the wetland-associated Collared Crow and starlings. No communal roosting sites of birds were recorded in LC. Further details are provided in Table 3 of Appendix 4.2b.

4.6.20 Some waterbird species recorded, including Little Grebe, Yellow Bittern, Common Moorhen and White-breasted Waterhen, are known to breed in fishponds and reedbeds in Hong Kong and it is likely that all these species bred in the Project Area, at least at NSW where there are more areas with limited human disturbance. However, the only confirmed breeding behaviour took place in the Shan Pui River Egretry, as is discussed in the following sections.

Birds using the Intertidal Shan Pui River and KTMDC

4.6.21 Wetland birds recorded in the KTMDC and SPR adjacent to the site were surveyed at high and low tide throughout the 2021-22 survey period; both high and low tide counts were undertaken as different species have differing foraging and roosting strategies with some species, notably waders, feeding on the exposed mudflats at low tide and others, such as ducks, feeding and/or loafing on the water when the tide is higher. Results are summarised in Table 4.7. Waterbird utilisation of the channels was not even, with the most heavily utilised channel sections being the downstream and wider sections (Figure 4.5).

4.6.22 Relative to the Deep Bay populations, the waterbird assemblage using the channels was more notable for its diversity than the number of individuals present, especially at high tides, when no species were recorded in moderate or large numbers, albeit a few species (including Great Cormorant, ducks and kingfishers) were still observed foraging. During low tides, the abundance of wetland-dependent and wetland-associated bird species was higher (moderate in a Deep Bay intertidal context), notably that of Common Moorhens and Black-winged Stilt and, to a lesser extent, Pied Avocet, sandpipers Tringa spp. and Black-headed Gull, though for all these species except Common Moorhen, Black-winged Stilt and Wood Sandpiper, the primary foraging habitats are the intertidal mudflats of Deep Bay. Further details regarding wetland-dependent and wetland-associated bird species in SPR and KTMDC are provided in Table 4 of Appendix 4.2b.

4.6.23 Apart from the birds using the SPR and KTMDC and the Shan Pui River egretry, no significant bird use was recorded in the other habitats in the 500m Assessment Area outside the Project Site.

Table 4.8: Maximum count of wetland-dependent bird species and other bird species of conservation importance recorded in KTMDC and Shan Pui River, April 2021 – March 2022

Species	High Tide	Low Tide
Great Cormorant	30	47
Grey Heron	8	17
Great Egret	8	21
Intermediate Egret	1	1
Little Egret	9	31
Chinese Pond Heron	15	36
Black-crowned Night Heron	4	3
Yellow Bittern	1	-
Black-faced Spoonbill	3	8
Eurasian Wigeon	-	16
Eurasian Teal	40	33
Northern Pintail	1	1
Northern Shoveler	30	79
Tufted Duck	5	36
Western Osprey	-	1
Black Kite	2	3
White-breasted Waterhen	3	6
Common Moorhen	8	25
Black-winged Stilt	9	131
Pied Avocet	32	71
Pacific Golden Plover	-	3
Grey Plover	-	3
Little Ringed Plover	3	2
Black-tailed Godwit	3	-
Whimbrel	1	1
Spotted Redshank	2	35
Common Redshank	5	54
Marsh Sandpiper	-	14
Common Greenshank	8	18
Green Sandpiper	1	1
Wood Sandpiper	4	2
Terek Sandpiper	-	1
Common Sandpiper	3	10
Common Snipe	2	-
Black-headed Gull	4	135
Pied Kingfisher	2	1
Common Kingfisher	2	1
White-throated Kingfisher	2	2
Collared Crow	-	2

Cormorant Roost Count

4.6.24 Evening surveys were conducted monthly from November 2021 to March 2022. Surveys recorded both the number and the distribution of cormorants roosting at the NSW site. Simultaneous counts were conducted at MPNR and LMC (the other known well-established roosting sites for Great Cormorant in the Deep Bay area) to ascertain the total number of roosting birds and relative number at each of the sites. A new roost site at Mai Po San Tsuen (MPST) near the border fence was counted in November 2021 to March 2022.

4.6.25 Up to 9,088 Great Cormorants were recorded roosting in Deep Bay area during the 2021-2022 winter. The peak count at the NSW site during the 2021-22 winter was of 7,405 birds on 16 December.

Table 4.9: Roost Counts of Great Cormorants at NSW, MPNR Lok Ma Chau and MPST during the 2021-2022 winter

Date	NSW Site			MPNR	MPST	LMC	Total in Deep Bay	% at NSW of Deep Bay Total
Date	Northern NSW	Southern NSW	Total at NSW	MPNR	MPST	LMC	Total in Deep Bay	% at NSW of Deep Bay Total
29 Nov 2021	708	4143	4851	420	N/A	770	6041	80%
16 Dec 2021	488	6917	7405	419	745	519	9088	81%
19 Jan 2022	256	5539	5795	1292	708	328	8123	71%
24 Feb 2022	278	4470	4748	1205	547	0	6500	73%

Note: Great Cormorants left Hong Kong on their northward migration earlier than in previous years in the 2021-2022 winter; thus none were recorded during the survey on 30 March 2022 at all sites.

Ardeid Night Roost

4.6.26 An ardeid night roost was found in mangroves at SPR, between the NSW and LC Project Sites (see Figure 4.1). Although birds have been observed roosting at this location subsequently, it does not appear to have been used consistently as a night roost in subsequent winters.

Table 4.10: Ardeids observed entering night roost during 2021-2022 winter

Species	23 Nov 2021	28 Dec 2021	27 Jan 2022	24 Feb 2022	30 Mar 2022
Chinese Pond Heron	11	17	99	62	25
Grey Heron	2	3	1	0	0
Great Egret	23	33	88	64	42
Little Egret	67	87	97	54	24
Eastern Cattle Egret	0	0	12	3	1
Total	103	140	297	183	92

Shan Pui River egretry

4.6.27 The number of nests at the SPR egretry were counted monthly from April to July 2022; the highest number of nests recorded was 38, all of Chinese Pond Heron, on 30 May. Monthly maximum counts are presented in table below. The total number of nests at the egretry has decreased significantly since this egretry was first recorded in the 2019 breeding season (see Table 4.4), the cause of the decline is unknown but similar declines were noted at other egretries in the Deep Bay area in 2022 (Anon 2022).

Table 4.11: Maximum monthly count of nests at Shan Pui River egretry between April and July 2022

Species	28 Apr 2022	30 May 2022	13 Jun 2022	17 Jul 2022
Chinese Pond Heron	31	38	36	28

4.6.28 Flightline surveys were undertaken to establish the routes used from the egretry by foraging Chinese Pond Herons (Figure 4.9). Most birds leaving the egretry were seen to fly to LC (Flightline 1: 33% of birds) or land in the KTMDC (Flightline 2: 32%), with smaller numbers (Flightlines 3 - 5: 10 – 13% of the total) flying down the SPR or towards FLW. No birds seen leaving the egretry were observed to land within the NSW site, or to fly towards or over Yuen Long Industrial Estate.

Tung Shing Lane egretry (not occupied since 2020)

4.6.29 The egretry survey confirmed that the TSL egretry remained abandoned (it was last utilised in 2020, see para. 4.4.17). Accordingly, no flightline survey was conducted.

Bird Flightlines over NSW, Proposed Development Site, and Proposed Bridge over the SPR

4.6.30 Large numbers of Great Cormorants were observed leaving the night roost on surveys during winter (Figure 4.7). The direction of flight was north-north-west (i.e. towards Deep Bay, as expected and as found in previous surveys), with cormorants using the southern roost trees following a flight-line that passed over the northern roost trees.

Surveys confirmed that the numbers of waterbirds flying over the NSW Development Site were very small and were insignificant in the context of the populations present in the Assessment Area, furthermore no defined flightlines (as distinguished from the flight paths of individual birds) were discernible (Table 4.12).

Table 4.12: Mean number of birds recorded per survey hour flying over the Development Site

Species	Mean per survey hour
Black-faced Spoonbill	0.04
Chinese Pond Heron	0.75
Great Cormorant	0.33
Great Egret	1.83
Grey Heron	0.25
Little Egret	1.08
Black-crowned Night Heron	0.46
Total	4.63

4.6.31 Surveys were conducted to determine the routes and flight elevations of waterbirds flying along the SPR to the west of the Proposed NSW Development Site that might potentially by impacted by the bridge that is proposed to be constructed over the channel (Figure 4.8). Most of the birds (84%) observed to use this route followed the channel itself, whilst 16% took a slightly more western route over eastern edge of the Industrial Estate, perhaps to reduce their flight distance. Approximately 9% of birds recorded flew below 10m; 14% flew at a height of 10-20m above the channel; 11% flew at a height of 20-30m; and the remaining 65% flew above 30m. The number of birds flying along the channel was small (only 4.7 birds per hour in total including those taking the more western route), furthermore it is reasonable to assume that those flying at above 30m would not need to alter their flight paths; the total number of birds that might have to make a very minor adjustment to their flight paths due to the bridge by flying higher or taking the western route would thus be very small.

Herpetofauna

Survey Findings

4.6.32 Seven amphibian species were recorded in the Assessment Area, all of which were noted at NSW, during surveys conducted in 2021-22. Most of these were present in very low numbers, although moderate numbers of Günther’s Frog were recorded. All species are common throughout Hong Kong, and none are of conservation importance.

4.6.33 Five amphibian species were recorded at LC during surveys. All were present in very low numbers. All species are common throughout Hong Kong, and none are of conservation importance.

4.6.34 Twelve reptile species were recorded during surveys in the Assessment Area, one of which, Chinese Gecko (not a species of conservation importance), was not recorded in the Project Sites. Ten of these species were found at NSW, all are distributed throughout Hong Kong except Chinese Stripe-necked Turtle, which is an introduced species in Hong Kong (IUCN 2024), the observations of which are considered to relate to ex-captive individuals. Reeves’ Turtle, Chinese Soft-shelled Turtle and Common Rat Snake are species of conservation importance. Locations of these observations are shown on Figure 4.10 and Figure 4.11a.

4.6.35 Three reptile species were recorded at LC during surveys, one of which, Many-banded Krait, was not found at NSW. All of these species are widespread throughout Hong Kong; but Many-banded Krait is of Potential Regional Concern (Fellowes et al. 2002). The location of this observation is shown on Figure 4.10 and Figure 4.11b.

4.6.36 Survey findings are similar those of AEC (2016). A full list of all species recorded at NSW and LC together with the maximum count of each species recorded by habitat is provided in Table 7 and Table 8 of Appendix 4.2b.

4.6.37 As required in the EIA Study Brief (Appendix C, para. 4(v)(h)), special reference was given to Common Rat Snake and Chinese Bullfrog. Common Rat Snake was recorded at NSW; this species is common and widespread throughout Hong Kong (Karsen et al. 1998), although it is listed as of Potential Regional Concern (Fellowes et al. 2002) and considered Endangered in China (Jiang et al. 2016). No recent records of Chinese Bullfrog were noted from previous studies, and this species was not recorded in the current surveys. Although considered uncommon (Karsen et al. 1998), Chinese Bullfrog is listed as widespread in Hong Kong (AFCD 2022).

Table 4.13: Herpetofauna species recorded at NSW and LC, April 2021 to March 2022

Species	NSW	LC	500m Assessment Area (excluding Project Sites)
Amphibians
Asian Common Toad	ü	ü	ü
Asiatic Painted Frog	ü		ü
Ornate Pigmy Frog	ü
Paddy Frog	ü	ü	ü
Günther’s Frog	ü	ü	ü
Brown Tree Frog	ü	ü	ü
Greenhouse Frog	ü	ü	ü
Reptiles
Reeves’ Turtle^	ü
Red-eared Slider	ü		ü
Chinese Stripe-necked Turtle *	ü
Chinese Soft-shelled Turtle^	ü
Changeable Lizard	ü
Chinese Gecko			ü
Bowring's Gecko	ü	ü	ü
Long-tailed Skink	ü
Reeve’s Smooth Skink	ü
Common Rat Snake^	ü
Many-banded Krait^		ü	ü
Checkered Keelback	ü	ü	ü

* Introduced species (IUCN 2024).

^ species of conservation importance

Odonates

Survey Findings

4.6.38 A total of 31 species was recorded in the 500m Assessment Area (including the Project Sites) during the baseline surveys conducted in 2021-22. Of these, four are of conservation importance. A total of 24 species was recorded in the 500m Assessment Area (excluding the Project Site), including two species of conservation importance namely Dingy Dusk-hawker and Scarlet Basker. The NSW site supported a moderately diverse odonate community, also with 24 species recorded during the surveys in 2021-22. The species recorded at NSW are widespread in lowland wetlands in Hong Kong except two species, Mangrove Skimmer, which is Uncommon (AFCD 2022) and Vulnerable (IUCN 2024); and Ruby Darter which is of Local Concern (Fellowes et al. 2002). LC supported a lower diversity of odonata, with 12 species recorded during surveys in 2021-22. All species are Abundant or Common in Hong Kong (AFCD 2022).

4.6.39 Locations of these observations are shown on Figure 4.10 and Figure 4.11. Survey findings are similar to those of AEC (2016). A full list of all species recorded at NSW and LC together with the maximum count of each species recorded by habitat is provided in Table 9 of Appendix 4.2b.

Table 4.14: Odonate species recorded at Nam Sang Wai and LC, April 2021 to March 2022

Species	NSW	LC	500m Assessment Area (excluding Project Sites)
Orange-tailed Midget			ü
Orange-tailed Sprite	ü		ü
Common Bluetail	ü	ü	ü
Common Evening Hawker			ü
Pale-spotted Emperor	ü		ü
Lesser Emperor	ü	ü	ü
Dingy Dusk-hawker^			ü
Common Flangetail	ü	ü	ü
Golden Flangetail		ü
Regal Pond Cruiser			ü
Asian Pintail	ü
Blue Dasher	ü	ü	ü
Asian Amberwing	ü	ü	ü
Crimson Darter	ü	ü	ü
Forest Chaser	ü
Russet Percher	ü	ü	ü
Pied Percher	ü
Red-faced Skimmer	ü		ü
Common Blue Skimmer	ü		ü
Marsh Skimmer	ü
Mangrove Skimmer^	ü
Common Red Skimmer	ü		ü
Green Skimmer	ü	ü	ü
Wandering Glider	ü	ü	ü
Pied Skimmer	ü		ü
Ruby Darter^	ü
Variegated Flutterer	ü	ü	ü
Evening Skimmer	ü		ü
Saddlebag Glider	ü	ü	ü
Scarlet Basker^			ü
Dingy Dusk-darter			ü
^species of conservation importance

Butterflies

Survey Findings

4.6.40 A total of 41 species was recorded in the 500m Assessment Area (including the Project Sites) during the baseline surveys conducted in 2021-22. Of these, four species are of conservation importance. A total of 32 species was recorded in the 500m Assessment Area (excluding the Project Site), including three species of conservation importance.

4.6.41 A total of 25 species was recorded at NSW during the 2021-22 survey periods, including three species of conservation importance namely Common Awl, Pale Palm Dart and Danaid Egg-fly, all of which are considered of Local Concern (Fellowes et al. 2002). A total of 11 species was recorded at LC, including 1 species of conservation importance namely Small Cabbage White, which is Rare in Hong Kong (AFCD 2022), albeit it is globally distributed in association with vegetable crops and is considered a pest species in many parts of the world. Locations of these observations are shown on Figure 4.10 and Figure 4.11a.

4.6.42 Survey findings are similar to those of AEC (2016). A full list of all species recorded at NSW and LC together with the maximum count of each species recorded by habitat is provided in Table 10 of Appendix 4.2b.

Table 4.15: Butterfly species recorded at the Nam Sang Wai and Lut Chau and in the 500m Assessment Area excluding the Project Sites, April 2021 to March 2022

Species	NSW	LC	500m Assessment Area (excluding Project Sites)
Common Awl^	ü		ü
Forest Hopper	ü
Formosan Swift	ü		ü
Common Straight Swift	ü
Pale Palm Dart^	ü		ü
Common Hedge Blue			ü
Lime Blue			ü
Plains Cupid			ü
Gram Blue		ü	-
Dark Cerulean			ü
Pale Grass Blue	ü		ü
Lesser Grass Blue			ü
Plum Judy			ü
Large Faun	ü		-
Common Tiger			ü
Common Indian Crow	ü		ü
Blue-spotted Crow	ü		ü
Blue Tiger		ü	ü
Angled Castor	ü	ü	ü
Red-ring Skirt	ü		ü
Great Egg-fly	ü	ü	ü
Danaid Egg-fly^	ü		ü
Lemon Pansy			ü
Common Sailer	ü	ü
Common Palmfly	ü		ü
Dark Evening Brown			ü
Dark-brand bush Brown	ü		ü
Common Five-ring	ü		ü
Tailed Jay	ü	ü
Common Bluebottle	ü
Red Helen	ü
Great Mormon			ü
Paris Peacock			ü
Common Mormon	ü		ü
Spangle			ü
Lemon Emigrant	ü	ü	ü
Common Grass Yellow	ü	ü	ü
Red-base Jezebel	ü	ü	ü
Great Orange Tip			ü
Indian Cabbage White	ü	ü	ü
Small Cabbage White^		ü
^species of conservation importance

MPBWF

Survey Findings

4.6.43 MPBWF was surveyed in the Assessment Area during April to October 2021. Concentrations of MPBWF were recorded in mangrove and Brachiaria marsh to the south of NSW and mangrove around the southern end of MPNR. Moderate numbers were recorded in the channel to the south of LC. Small numbers of MPBWF were also observed in the mangroves along KTMDC and SPR. No MPBWF was observed in other habitats at NSW or LC. Locations of the MPBWF recorded in 2021 are shown in Figure 4.12b. A comparison between surveys in 2019 (AEC 2020) and surveys in 2021 is shown in Figure 4.12c. The number of MPBWF recorded at NSW decreased, perhaps because the Brachiaria marsh to the south of NSW has been heavily disturbed by people and light levels have increased because of streetlamps that have been installed near the Brachiaria marsh. The number of MPBWF recorded at the south of MPNR, and between LC and Tai Sang Wai also decreased; this may be due to the growth of mangrove stands along the tidal watercourse - the study on habitat preference by AEC (2020), Appendix 4.4) suggested that extensive pure mangrove stands may not be favoured by this species, albeit further study is required.

4.6.44 Light levels recorded at sites used by MPBWF in 2021 were in the range of 0.01-0.51 lux. Light levels measured at other sites where no MPBWF were recorded ranged from 0.01-0.65 lux. On a few occasions single MPBWF were recorded near streetlamps along the KTMDC but these observations are not thought to suggest that MPBWF does not avoid brightly-lit locations.

Table 4.16: Peak Counts of MPBWF recorded at Nam Sang Wai and Lut Chau, April 2021 to October 2021

Species	NSW (Shan Pui River)	LC	South of MPNR	KTMDC
MPBWF	47	25	83	13

4.7 Evaluation of Habitats

4.7.1 The following section provides an ecological evaluation of habitats at NSW and LC and the Assessment Area (i.e. within 500m of the Project Boundary).

Tidal Watercourse

4.7.2 NSW is bounded to the north-east and west by the KTMDC and the SPR, respectively. These join and continue north along the western side of LC to reach Deep Bay. Both rivers have been channelised and were designed to be environmentally friendly using an unlined muddy bottom for the trapezoidal channel and planting of mangroves along both sides of the channel (Lai et al. 2007); these mangrove areas are evaluated under Mangrove below. The channel is intertidal, with the muddy bottom exposed at low tide. These intertidal areas support an abundant and diverse community of wetland birds and are particularly important for ardeids, spoonbills, ducks (especially Northern Shoveler, Common Teal and Tufted Duck), shorebirds (especially Black-winged Stilt and Pied Avocet) and gulls. The intertidal river channels are largely devoid of vegetation, but the sides of the river support mangrove communities as described above.

4.7.3 In addition, a small section of tidal channelised watercourse is present at the southwest of the plantation area connecting NSW fishponds to the SPR. This has artificial sides but mud bottom and has some colonizing mangrove species present. Fauna includes feeding egrets and is similar to that of intertidal mangrove-lined creeks.

4.7.4 Smaller natural intertidal, mangrove-lined watercourses are present to the south of NSW (the former alignment of Kam Tin River) and to the southeast and north of LC. Being narrower, with less exposed mudflats, these channels support a lower abundance of waterbirds.

Table 4.17: Ecological evaluation of Tidal Watercourse

Criteria	KTMDC and SPR	Channelised Watercourse linking NSW and SPR	Natural Watercourses including the former alignment of Kam Tin River and near LC
Naturalness	Man-made channels with artificial banks but natural base.	Man-made channel with artificial banks and mud bottom.	Semi-natural streams and rivers.
Size	Relatively large within the Assessment Area (76.7ha) and wide (>100m wide for KTMDC, 30m-40m wide for SPR). None is actually within the Project Site, but 0.1ha is under the footprint of the proposed bridge over the SPR.	Very small, entirely within the Project Site. (0.2ha)	Fairly small and narrow (around 5.0m wide). Small areas fall within the Project Site at NSW and to the north of LC (0.4ha and 0.2ha respectively).
Diversity	Low diversity of microhabitat types but moderate faunal diversity, especially birds.	Low diversity of fauna present but includes mangrove-associated species.	Fairly low diversity of microhabitat types but mangroves support moderate faunal diversity.
Rarity	Most intertidal mudflat habitat in Hong Kong is in Deep Bay, but habitat is common there. Supports moderate numbers of wetland bird species of conservation importance, but all are present in small numbers in a Deep Bay context, and none are restricted to this habitat.	Small channelised watercourses are common in Hong Kong. Supports small numbers of a few wetland bird species of conservation importance, but in insignificant numbers.	Several similar small channels exist in Deep Bay area but otherwise fairly restricted in Hong Kong. Supports small numbers of a few wetland bird species of conservation importance, but in insignificant numbers.
Re-creatability	Channel could be re-created at a suitable location.	Channel could be re-created at a suitable location.	Could be re-created at a suitable location.
Fragmentation	Not fragmented.	Not fragmented.	Not fragmented.
Ecological linkage	Ecologically linked to mudflats and mangroves around Deep Bay and mangrove stands lining the channels, fishpond areas, and to upstream sections of the rivers.	Ecologically linked to SPR and adjacent mangroves and NSW fishponds.	Ecologically linked to KTMDC and SPR, mangroves around Deep Bay and adjacent pond areas.
Potential value	Little scope for further enhancement.	Little scope for enhancement due to small size.	Little scope for further improvement.
Nursery / breeding ground	No known breeding sites but provides foraging habitat for birds breeding at SPR egretry.	Not a known nursery/breeding ground.	No known breeding sites.
Age	Relatively recent; construction completed in 1997.	Not known. Perhaps same as SPR.	Relatively old habitat: LC supports mature mangrove stands (see Evaluation of Mangrove below).
Abundance / Richness of wildlife	High abundance and diversity of wetland birds in the northern section of the SPR, low to moderate abundance elsewhere.	Limited abundance and richness of wildlife due to small size but utilised by some wetland fauna.	Mangrove community at LC is relatively good (see Evaluation of Mangrove below) but otherwise these streams support a relatively low faunal diversity.
Ecological value	The northern section of the SPR is of Very High ecological value due to its importance for large waterbirds, but the southern section, is of Low to Moderate ecological value as it is used by relatively few waterbirds and is already disturbed because its proximity to Yuen Long. The KTMDC is of Moderate ecological value but is also somewhat disturbed by adjacent village housing	This watercourse is of Low to Moderate Ecological Value.	These watercourses are generally of Moderate Ecological Value

Fishpond

4.7.5 This habitat type comprises fishponds. Most of these fishponds at NSW and almost all fishponds at LC are actively maintained for commercial aquaculture. Management includes periodic drainage for harvesting of fish, management of water quality and adjustment of fishpond profiles. Large waterbirds (including egrets and spoonbills) are frequently attracted into fishponds during drain-down undertaken to facilitate fish harvesting. Bund vegetation is regularly managed and is mostly maintained with very low vegetation.

4.7.6 Due to the presence of large numbers of fish, there is little vegetation in the ponds, whilst the dominant plant species on the pond bunds are common grasses (such as Brachiaria mutica, Panicum maximum and Phragmites australis) and ruderal herbs such as Bidens alba and climbers (such as Paederia scandens and Mikania micrantha). Some trees are also present, especially fruit trees such as Litchi chinensis, Dimocarpus longan and Artocarpus macrocarpon. To enable vehicular access, some fishpond bunds have been strengthened with fill material, limiting the colonisation of vegetation.

4.7.7 Whilst active fishponds are well known for their value to wildlife, especially waterbirds, active management requires human and vehicular activity, this leads to some disturbance to large waterbirds and other more disturbance-sensitive wildlife. Fishpond operators are resident at both NSW and LC, so some degree of disturbance is present at all times (albeit at significantly lower levels overnight). Dogs are present at both sites, creating an additional source of disturbance to wildlife. Man-made structures and cables around the fishponds further increase the levels of disturbance.

4.7.8 There were a handful of inactive fishponds in the Project Site at NSW, a single such pond at LC. Whilst similar to active fishponds in most respects, in the absence of management such ponds eventually tend to become too acidic to support fish and much aquatic fauna and hence offer little food to birds and other foraging species. As such, they generally support few foraging waterbirds, however, as they tend to be relatively undisturbed, they may be utilised as nesting, loafing or roosting sites. Whilst not considered as an ecologically distinct habitat, inactive fishponds are considered separately in the Fisheries Impact Assessment and are mapped at Figure 5.1.

4.7.9 A total of 36 bird species of conservation importance were recorded in fishponds from the NSW Project Site in the recent twelve-month survey period. Amongst these, ardeids and Great Cormorants, as well as Red-billed Starling occurred in higher numbers in active fishpond habitat. Increased numbers of ardeids were often present during fishpond drain-down followed by aquaculture harvest. The shallow water habitat formed during drain-down was also found to attract Black-faced Spoonbills (assessed as Endangered by Birdlife International (2017b)) and waders, albeit numbers of both Black-faced Spoonbills and waders were small in a Deep Bay context. Red-billed Starling is an opportunistic species which scavenges around fishponds; it is widespread in the Deep Bay area during the winter months.

Table 4.18: Ecological evaluation of Fishpond

Criteria	Fishpond
Naturalness	A man-made habitat comprised of actively managed fishpond areas located in the northeastern and northwestern parts of NSW.
Size	~22 ha in the NSW site and ~36 ha in LC; ~120 ha in the 500m Assessment Area. Out of these totals, inactive fishponds comprise 6ha at NSW, 0.5ha at LC and 52.9ha elsewhere in the Assessment Area.
Diversity	Low habitat and vegetation diversity but moderate diversity of fauna, especially birds.
Rarity	Fishponds are a common habitat in the Deep Bay area but are becoming less common throughout Hong Kong. Frequent records of several groups of waterbirds of conservation importance, including ardeids, waders and cormorants, especially during fishpond drain-down, but numbers are small in a Deep Bay context.
Re-creatability	Easily re-creatable.
Fragmentation	Not fragmented.
Ecological linkage	Fishponds show strong ecological linkage to nearby wetland habitats, including reedbeds, marshes and tidal watercourses.
Potential value	Value could be increased by more ecologically friendly management methods. Management Agreement(s) operating under the New Nature Conservation Policy (NNCP) may be effective in this respect. However, value may also decrease if fisheries management becomes more intensive.
Nursery/ breeding ground	Shan Pui River egretry is located in pond bund area north of the Yuen Long Sewage Treatment Works. Fishponds in general may also be used by breeding ardeids as a foraging ground and inactive fishponds may provide nesting sites for a few species of wetland birds.
Age	Not known but moderately old.
Abundance/ Richness of wildlife	Abundance of most bird species lower than at many other fishpond sites in Deep Bay area, but some species are present in high numbers and some waterbirds may be abundant during fishpond drain-down, especially where these are carried out under Management Agreements. Low abundance and diversity of other fauna (odonates and amphibians).
Ecological value	In their current state these fishponds attract relatively low numbers and diversity of wetland species, although some wetland birds are present in good numbers and the ecological linkages are good; these active fishponds are therefore considered currently to be of High Ecological Value while the inactive fishponds are of Moderate to High Ecological Value.

Open Water

4.7.10 Most of the former fishponds within the NSW site and in the surrounding area have long been abandoned, potentially more than thirty years. Some of these have open water areas, but all have been at least partially overgrown with other vegetation (including reedbed Phragmites sp., bullrush Typha sp., ferns Cyclosorus sp. and grasses, especially Brachiaria sp.) and some have no open water remaining; these overgrown areas are assessed separately in the following sections. As is typical of freshwater and brackish open water habitats in Hong Kong, surveys confirmed that there were no submerged plants growing in these open water areas.

4.7.11 Compared to the active fishponds, the open water areas in the abandoned fishponds receive considerably less human disturbance, increasing their value to disturbance-sensitive species. Conversely, they lack the drain-down period of actively managed fishponds, and the fish stocks are lower than in active fishponds; also, in the absence of management some of these fishponds are acidic. Both factors are likely to result in a reduction in their ecological function, especially for foraging piscivorous birds.

4.7.12 However, open water areas support a distinctly different faunal community to those utilising active fishponds. Bird species associated with this habitat included Yellow Bittern and Purple Heron, species which were not found to utilise active fishponds. In addition, several passerine species were observed on the interface habitat between open water and reedbed; however, this is likely to be a function of these species being more readily detected on the edge of reedbeds rather than their being particularly associated with this ecotone.

4.7.13 There is evidence from surveys that open water habitat in abandoned fishponds may be of importance for turtle/terrapin species of conservation importance at NSW site; with observations of two species: Reeves’ Turtle and Chinese Soft-shelled Turtle.

Table 4.19: Ecological evaluation of Open Water

Criteria	Open Water
Naturalness	A semi-natural habitat largely resulting from the abandonment of aquaculture and comprising a seral stage in the process of natural succession from inactive fishpond to reedbed or another vegetated wetland.
Size	~2.6 ha in the NSW site; ~14.2 ha in the 500m Assessment Area
Diversity	Low habitat and vegetation diversity but moderate diversity of fauna, especially birds.
Rarity	A relatively rare habitat, both in Hong Kong and in the Deep Bay area; area has probably declined significantly in recent years, as either succession to vegetated wetland or resumption of aquaculture has occurred. Has declined in area in the Project Site for the same reasons. Fauna recorded includes two native turtle species of conservation importance.
Re-creatability	Easily re-creatable.
Fragmentation	Intrinsically fragmented, but this does not necessarily detract from its value so long as the intervening habitat is wetland.
Ecological linkage	Intrinsically linked to nearby wetland habitats, including reedbeds, marshes and tidal watercourses.
Potential value	Value could be increased by active management.
Nursery/ breeding ground	No significant nursery or breeding grounds, but potential used by breeding ardeids in the nearby egretries.
Age	Not known but moderately old.
Abundance/ Richness of wildlife	Abundance of birds low but some of the species, notably Purple Herons and Yellow Bitterns occur at low densities in Hong Kong. Low abundance and diversity of reptiles, four species, including two species of conservation importance. Low abundance and diversity of other fauna (odonates and amphibians).
Ecological value	In their current state these open water areas attract relatively low numbers and diversity of wetland species, but some of the species are locally infrequent; the ecological linkages with adjacent vegetated wetlands are good; hence considered to be of Moderate to High Ecological Value.

Reedbed

4.7.14 Many of the abandoned fishponds at the NSW site have been colonised by reeds Phragmites australis and have evolved into extensive reedbed habitat. However, perhaps the single most important finding from the updated habitat and vegetation and faunal surveys conducted in 2021-22 is the extent to which the NSW reedbed has undergone changes to other wetland habitats. Overall, the area of reedbed within NSW site declined from 49.5ha to 40.96ha (17.3%) from 2015 to 2021-22. Some of this decline is due to successional change to drier wetland habitats (Cyclosorus and Brachiaria marsh); and some is due to the spread of invasive Typha angustifolia.

4.7.15 Reedbed habitat is generally rare in Hong Kong, being largely restricted to the Deep Bay area. Elsewhere in the Assessment Area reedbed areas are small and rather fragmented. The reduction in reedbed due to succession and invasion by Typha is a matter of concern.

4.7.16 Over 25 avifauna species of conservation importance have been found in reedbed habitat at the NSW site. Purple Heron, which is a reedbed specialist species, has been regularly recorded at NSW (including in the recent 12-month survey). Purple Heron is of Regional Concern (Fellowes et al. 2002).

4.7.17 In addition, all the three bittern species known to favour reed marshes in Hong Kong, namely Eurasian Bittern, Yellow Bittern and Cinnamon Bittern (Carey et al. 2001), have been found in the reedbed in the Nang Sang Wai Project Site, though only Yellow Bittern was noted in the recent survey period. Yellow Bittern has been recorded from the reedbed at NSW during the breeding season, suggesting that it may breed there - reedbed is one of the main breeding habitats for the species (Carey et al. 2001). Eurasian Bittern, which is an uncommon winter visitor and spring passage migrant in Hong Kong, is of Regional Concern (Fellowes et al. 2002). Cinnamon Bitterns are present in Hong Kong largely as passage migrants with small numbers in summer and winter.

4.7.18 Chinese Penduline Tit is of Regional Concern by Fellowes et al. (2002), though it is a common autumn migrant and winter visitor to the Deep Bay area in Hong Kong. In the recent survey period Chinese Penduline Tits were regularly recorded during the winter months in flocks of up to 60 individuals, indicating that the site continues to be of importance for this species.

4.7.19 Pallas’s Grasshopper Warbler and Lanceolated Warbler have occasionally been recorded from the reedbed at NSW site and were also recorded in the recent 12-month surveys. Reedbed in Hong Kong has been shown to be an important foraging habitat for these species, and it is also utilised by other, rarer, species in the genera Locustella and Helopsaltes (Allcock et al. 2013, 2018). Both species are passage migrants to densely vegetated habitats in Hong Kong, though Pallas’s Grasshopper Warbler is commoner and considered of Local Concern (Fellowes et al. 2002). Lanceolated Warbler is assessed as Near Threatened in China (Jiang et al. 2016).

4.7.20 The globally threatened Yellow-breasted Bunting was occasionally recorded in low numbers at NSW, though not in the recent twelve-month period. It is Critically Endangered on the IUCN Red List (BirdLife International 2017a). It is also Endangered in China, Jiang et al. (2016) and is of Regional Concern by Fellowes et al. 2002).

4.7.21 Overall, whilst it is difficult to be definitive as the number of individuals of many species is rather small, and many of these species are cryptic, the relatively low numbers of most of the bird species of conservation importance recorded during the recent twelve-month survey period, suggests that the value of reedbed at NSW as habitat for reedbed-dependent or reedbed-associated bird species has declined in the last five years.

4.7.22 Apart from avifauna, Leopard Cat and Small Indian Civet have been recorded from reedbed areas, albeit probably both species are more likely to use bund areas, rather than the actual reedbeds. Similarly, Burmese Python, Common Rat Snake and Reeves' Turtle were previously recorded in this habitat, but in view of these species’ habitat requirements it is probable that their use of reedbeds is, at most, incidental. Though Burmese Python and Common Rat Snake are widespread throughout Hong Kong (AFCD 2022), they are Critically Endangered and Endangered in China respectively (Jiang et al. 2016). Burmese Python is also a globally threatened species listed as Vulnerable on IUCN Red List (2024). Reeves’s Turtle used to be a common species throughout the territory, but sightings have been rare in recent years (AFCD 2022). This species is Endangered globally and nationally, IUCN (2024) and Jiang et al. (2016) respectively. An amphibian species, Spotted Narrow-mouthed Frog, assessed as Near Threatened in China (Jiang et al. 2016), was recorded once from reedbed habitat. This species is widespread from low to moderate altitudes in northern and central New Territories (AFCD 2022). Small numbers of odonates Ruby Darter and Scarlet Basker were found from the reedbed areas. Both species are listed as of Local Concern by Fellowes et al. (2002) but are currently assessed as Common in Hong Kong (AFCD 2022).

Table 4.20: Ecological evaluation of Reedbed

Criteria	Reedbed
Naturalness	Habitat has evolved from natural colonisation of vegetation into man-made fishponds. Some areas gradually drying because of natural processes and being colonised by terrestrial vegetation; others are being invaded by the exotic Typha angustifolia.
Size	~41ha in NSW; 8.5ha in the 500m Assessment Area
Diversity	Very low vegetation diversity but moderately high diversity of fauna, especially birds. In Hong Kong most reedbed-dependent species favour wet reedbeds (Allcock et al. (2013, 2018).
Rarity	Reedbed is a rare habitat in Hong Kong, and that at NSW is one of the most extensive areas present. Some reedbed-dependent bird species of conservation importance are present.
Re-creatability	Easily re-creatable.
Fragmentation	Somewhat fragmented by dry bunds and increasingly fragmented by other vegetated wetland habitats including invasive Typha marsh.
Ecological linkage	Strong ecological linkage to nearby fishponds, and other wetland habitats; also, linkages to non-wetland habitats for mammals and reptiles.
Potential value	Value could be increased by more ecologically friendly management methods such as increasing the extent of permanently wet areas and removal of invading terrestrial vegetation and Typha.
Nursery/ breeding ground	Apparently, a breeding ground for common amphibian species. May also provide breeding habitat for uncommon breeding birds such as Yellow Bittern.
Age	Reedbed has formed since abandonment of fishponds in the last 15-20 years.
Abundance/ Richness of wildlife	Abundance of fauna relatively low compared to fishponds and intertidal wetland habitats but diversity relatively high. There is some evidence from the recent 12-month survey period that both numbers and diversity of reedbed bird species have declined since 2015.
Ecological value	NSW contains one of the largest remaining areas of reedbed in Hong Kong and supports reedbed-dependent species. Whilst there is evidence of the reedbed habitat becoming fragmented and its value for fauna having declined, reedbed at the site is considered to be of High Ecological Value.

Brachiaria Marsh

4.7.23 Brachiaria marsh is scattered throughout the NSW site. This habitat is dominated by the exotic grass Brachiaria mutica and it is commonly found at the drier fringes of the abandoned fishponds in the NSW site. Other terrestrial vegetation, including tree seedlings of Eucalyptus tereticornis and Melia azedarach; terrestrial weedy climbers Paederia scandens and Mikania micrantha were also recorded on bunds adjoining this habitat. The flora composition of this habitat is simple, and the habitat is mainly dominated by grass Brachiaria mutica.

4.7.24 Low faunal diversity and numbers were recorded from Brachiaria marsh. No species appeared to rely on this habitat. A few bird species of conservation importance (Purple Heron, Little Egret and Chinese Penduline Tit) were found in low numbers in this habitat.

4.7.25 Most of the Brachiaria marsh at NSW has formed in former fishpond areas. However, a seminatural Brachiaria marsh is located in the extreme southwest of NSW within the Project Site. This marsh area was found to be particularly utilised by MPBWF. The abundance of MPBWF in this area was higher than elsewhere in NSW.

Table 4.21: Ecological evaluation of Brachiaria Marsh

Criteria	Brachiaria Marsh in the southwest of NSW (within Project Site)	Brachiaria Marsh elsewhere
Naturalness	Seminatural: habitat has probably developed from sediments deposited by the adjacent seminatural watercourse.	Habitat has evolved from exotic vegetation colonisation into man-made fishponds.
Size	Approximately 1.3ha in NSW.	Approximately 5.7ha
Diversity	Low fauna and flora diversity.	Low fauna and flora diversity.
Rarity	Common in lowland areas. No records of faunal species of conservation importance except MPBWF This species occurred in this area in relatively high numbers.	Common in lowland areas. Occasional records of a few mammals, bird and odonate species of conservation importance.
Re-creatability	Re-creatable, but due to tidal influences in its location, levels, erosion/sedimentation and vegetation establishment would require careful consideration and implementation.	Easily re-creatable.
Fragmentation	Moderately fragmented.	Moderately fragmented.
Ecological linkage	Ecologically linked with adjacent wetland habitats, linkages with adjacent mangrove stands probably important for MPBWF.	Ecologically linked with adjacent wetland habitats
Potential value	Potential for ecological enhancement reducing light levels and adjusting vegetation mix to increase its habitat value for MPBWF. Would also benefit from removal of dumped rubbish and a reduction in disturbance.	Little potential for ecological enhancement, overall ecological value of the Site would be enhanced by reduction of the area of this habitat.
Nursery/ breeding ground	Although MPBWF abundance recorded in this area was relatively high, no signs of breeding (such as presence of larvae) were noted, however assumed to be a breeding site on a precautionary basis.	No significant nursery or breeding grounds known.
Age	Unknown, though may be a relatively old seral habitat maintained by tidal influences and storm flows in the channel.	A young, successional habitat.
Abundance/ Richness of wildlife	Low	Low
Ecological value	Supports a significant population of MPBWF but currently somewhat degraded by rubbish dumping and other disturbance, hence currently considered to be of Moderate ecological value.	Low to Moderate

Cyclosorus Marsh

4.7.26 The wetland fern Cyclosorus interruptus is dominant in some of the abandoned fishponds forming patches of Cyclosorus marsh. This is habitat scattered throughout the NSW site and appeared to be largely present in very shallow water or seasonally wet areas occurring as a transitional successional stage between reedbed or open water and Brachiaria marsh. Terrestrial plant species, including grass Brachiaria mutica and climber Mikania micrantha were occasionally recorded in this habitat. Faunal diversity and numbers in Cyclosorus marsh were low and no species appeared to rely on or favour this habitat.

Table 4.22: Ecological evaluation of Cyclosorus Marsh

Criteria	Cyclosorus Marsh
Naturalness	Habitat has evolved from natural vegetation colonisation into man-made fishponds.
Size	4.1ha in NSW.
Diversity	Low plant diversity.
Rarity	Not uncommon in Hong Kong. A few records of bird species of conservation importance.
Re-creatability	Easily re-creatable.
Fragmentation	Moderately fragmented.
Ecological linkage	Ecologically linked with adjacent wetland habitats.
Potential value	Little potential for ecological enhancement; overall ecological value of the Site would be enhanced by reduction of the area of this habitat.
Nursery/ breeding ground	No significant nursery or breeding grounds known.
Age	A young, successional, habitat that has largely formed within the last five years.
Abundance/ Richness of wildlife	Low.
Ecological value	Low to Moderate

Typha Marsh

4.7.27 This habitat has formed as a result of colonisation of abandoned fishponds by the invasive and exotic wetland herb Typha angustifolia. This exotic species aggressively outcompetes native wetland plant species, especially Phragmites australis, forming dense monocultures with extremely low floristic diversity.

4.7.28 Low faunal diversity and numbers were recorded from Typha marsh. No species appeared to rely on this habitat. According to the survey findings, Typha marsh is not a favoured habitat for wetland-dependent bird species at NSW, though single records of Black-browed Reed Warbler and Oriental Reed Warbler, and occasional records of other bird species of conservation importance were obtained. These species are common migrants in wetland habitats in Hong Kong, especially reedbeds, thus their use of structurally similar Typha marsh is unsurprising.

Table 4.23: Ecological evaluation of Typha Marsh

Criteria	Typha Marsh
Naturalness	Not natural: Typha angustifolia is an exotic invasive species in Hong Kong wetlands.
Size	9.8ha in NSW.
Diversity	Low fauna and flora diversity.
Rarity	Not uncommon in Deep Bay area with an increasing trend; outcompeting other wetland plants, especially reed.
Re-creatability	Easily re-creatable.
Fragmentation	Moderately fragmented.
Ecological linkage	Ecologically linked with adjacent wetland habitats.
Potential value	Removal of invasive exotic and replacement Typha would be a significant ecological benefit.
Nursery/ breeding ground	No significant nursery or breeding grounds known.
Age	A fairly young, successional, habitat; most stands have formed within the last five years.
Abundance/ Richness of wildlife	Low.
Ecological value	Low

Grassy bund, wooded bund and bare soil bund

4.7.29 The fishponds of NSW and LC were formed with earth bunds between them. These original bunds remain; those around the active fishponds continue to fulfil their original functions of separating the water bodies and providing access for management purposes. Meanwhile, whilst some of the bunds in former fishpond areas are intact (and hence continue to fulfil a hydrological function), many have gaps or partially lowered sections and no longer do so. Some of the bunds around active fishponds are maintained as tracks for vehicle access, others are kept largely free from vegetation or are covered with managed grass with some planted fruit and other trees including large Eucalyptus spp. Conversely, the bunds in the former fishpond areas are largely overgrown with long grass, shrubs and naturally colonising or previously planted trees, again including large Eucalyptus trees in some locations. Accordingly, bunds were categorised according to their vegetation cover as grassy bund, wooded bund and bare soil bund.

4.7.30 Fauna species using bunds are largely those primarily associated with the adjacent wetland habitats, for example ardeids foraging in surrounding fishponds, or are common generalist landbirds, often found in association with people and in village-edge locations including birds such as bulbuls, starlings and munias. Whilst overgrown bunds no doubt provide some food, habitat and cover for such birds, together with various other fauna species in small numbers, no species of conservation importance were found to be strongly associated with or dependent upon bund habitats.

4.7.31 However, whilst the bund habitats, taken as a whole, may have limited ecological value, the large eucalypts planted on some bunds at NSW are of high importance as they provide some of the roost trees for the Great Cormorant roost (as well as providing roosting or perching sites for other bird species including Black Kites and other raptors and Collared Crows). Accordingly, these eucalypt roost trees are evaluated separately below.

Table 4.24: Ecological evaluation of Grassy Bund, Wooded Bund and Bare Soil Bund

Criteria	Grassy Bund, Wooded Bund and Bare Soil Bund
Naturalness	Not natural: formed as an integral element of the construction of the original fishponds. Most herbaceous vegetation has colonised naturally but includes a high proportion of pantropical weed species. Most trees are exotic and planted, notably large Eucalyptus spp. on some bunds at NSW, but some native or naturalised species such as Melia azedarach have grown naturally. Some bunds adjoining active fishponds are under regular management by the operators as vehicular access tracks.
Size	7.3ha of grassy bund within NSW; and 9.2 ha of grassy bund within LC; 9.7 ha of wooded bund (of which the cormorant roost trees comprise 0.7ha) within NSW; and 1.0 ha of wooded bund within LC; 0.8 ha of bare soil bund within NSW and 0.2 ha of bare soil bund within LC.
Diversity	Low fauna and flora diversity.
Rarity	Common habitats in the Deep Bay area. Utilised by fauna associated with adjoining wetland habitats and village-edge habitats in small numbers, but not of importance for any species of conservation importance except for Great Cormorants which utilise eucalypt trees on some of the wooded bunds for roosting.
Re-creatability	Easily re-creatable, but it would take some years for replacement trees to grow large enough to be suitable for roosting cormorants.
Fragmentation	Not fragmented.
Ecological linkage	Ecologically linked with adjacent wetland habitats.
Potential value	Removal of invasive exotic vegetation and regular management would be a significant ecological benefit. Repair of breached bunds would benefit hydrological management of wetland habitats.
Nursery/ breeding ground	No significant nursery or breeding grounds known.
Age	Formed decades ago when the area was largely man-made fishponds. Eucalypts used as the cormorant roost trees are 40+ years old.
Abundance/ Richness of wildlife	Generally low, though numbers of some birds such as ardeids may be moderate on occasion in opportunistic response to food availability (for example when an adjacent pond is drained).
Ecological value	Low except some wooded bunds (a total of 0.7ha) are of High ecological value due to the cormorant roost.

Mangrove

4.7.32 The Assessment Area includes a large mangrove area around LC and to the west of MPNR which is continuous with the extensive mangroves around Deep Bay, and conditions in the LC mangroves are similar to those elsewhere in Deep Bay. The dominant mangrove species are Kandelia obovata and Acanthus ilicifolius but the exotic mangrove Sonneratia spp. is also present. Surveys of intertidal fauna were not conducted in the mangrove stands at LC because these are not expected to be affected by the project and access into this ecosystem would result in unnecessary disturbance and/or damage to the ecosystem. It is expected that the faunal community is the same as that found in mangroves at Mai Po. Part of this mangrove area falls within LC.

4.7.33 A further area of mangrove is present along both sides of the KTMDC. This was originally planted as an element of the mitigation for the ecological impacts of the channelisation project, but now has a largely natural appearance and ecological function, supporting characteristic mangrove invertebrate and bird communities. However, the faunal community appears to be less diverse than in older and larger stands, perhaps because it is narrow and linear, and the landward sides abut public roads rather than more natural habitats. As in the mangrove at LC, the exotic mangrove Sonneretia spp. has colonised and has become quite extensive.

4.7.34 A smaller area of mangrove is located at the southwestern corner of the NSW site. Although there remains some tidal influence, this area of mangrove is drier than is found in most of Deep Bay, parts of the mangrove stand have been colonised by terrestrial vegetation and climbers, especially around the fringes and it is somewhat disturbed by adjacent development. As a result, the mangrove is relatively degraded compared to most mangroves in the Deep Bay area.

4.7.35 Mangrove habitats in Deep Bay are known to support diverse fauna, especially invertebrates and fish including the MPBWF which was only discovered in 2009 and was described in 2011, it was first found inhabiting mangroves around Deep Bay area but has since been also recorded along the coastline of western Guangdong Province and Hainan Island (Fan & Fu 2017, AFCD 2022).

4.7.36 MPBWF occurs in or in association with mangroves in the Assessment Area and in Project Site at LC and along the KTMDC and at the southwest of NSW. MPBWF are sometimes seen perched on mangroves at the fringes of the closed-canopy mangrove habitat, but detailed habitat surveys in 2019 (AEC 2020) suggest that this species is largely dependent on adjacent mangrove associates, rather than the closed-canopy mangroves. The mangrove associate community is found in intertidal areas without closed-canopy mangroves and consists of a plant community dominated by Acanthus ilicifolius, Clerodendrum inerme and Derris trifoliata, typically forming dense stands of vegetation up to around 1.5m tall. This habitat may be part of the natural succession from mangroves to terrestrial habitat. The surveys conducted in 2019 suggested that this is the most important habitat in the Deep Bay area for the MPBWF, although few other species of conservation importance have been recorded (AEC 2020).

Table 4.25: Ecological evaluation of Mangrove

Criteria	Mangroves at LC and to the West of Mai Po Nature Reserve	Mangroves at the Kam Tin Main Drainage Channel	Mangroves at Nam Sang Wai
Naturalness	A natural habitat in intertidal areas in Hong Kong.	A natural habitat in intertidal areas in Hong Kong, but this mangrove area was planted.	A natural habitat in intertidal areas in Hong Kong.
Size	~110ha of which 8.5ha are within the Project Site at LC.	~19ha, none of which are within the Project Site	2.6ha, all within the Project Site.
Diversity	Relatively high diversity of microhabitats, expected to have high diversity of fauna.	Moderate diversity of microhabitats, with a correspondingly moderate faunal diversity.	Relatively low diversity of mangrove species and hence associated fauna.
Rarity	Mangroves are relatively common in Deep Bay intertidal areas. The mangroves in Deep Bay are better than most remaining stands in south China. Habitat for the MPBWF.	Mangroves are relatively common in Deep Bay intertidal areas. Small numbers of MPBWF found during surveys. Also used by ardeids as a night roost (near SPR).	Mangroves are relatively common in Deep Bay intertidal areas. Few MPBWF recorded during surveys, but relatively large numbers of this species were found in adjacent Brachiaria marsh.
Re-creatability	Could be re-created by planting in intertidal area.	Created habitat which could readily be re-created.	Could be re-created by planting in intertidal area.
Fragmentation	Not fragmented.	Fragmented from large intertidal mangrove areas, but linear habitat which is not internally fragmented.	Fragmented from other mangrove areas.
Ecological linkage	Part of and with very strong ecological links to contiguous mangrove area around Deep Bay and intertidal mudflat areas.	Linkages with intertidal areas in river channel and other nearby mangrove areas.	Some linkage to adjacent intertidal areas and nearby marshes, notably brackish Brachiaria marsh which appears to be of importance to MPBWF.
Potential value	Value could be enhanced by controlling exotic Sonneratia spp.	Mangrove habitat is still quite young and may still to have potential to increase over time. Value could also be enhanced by controlling exotic Sonneratia spp.	Value might be enhanced by controlling terrestrial vegetation and climbers and reducing disturbance by people.
Nursery / breeding ground	Presumed to provide breeding ground for mangrove fauna including MPBWF, and probably for some marine fish.	Presumed to provide breeding ground for mangrove fauna including MPBWF.	Appears not to be utilised by breeding adult MPBWF but may have a role in their life cycle. No other nursery or breeding ground known.
Age	Age not known but is a well-established mangrove stand.	Planted approximately 20 years ago, with natural colonisation in the last decades	Age not known.
Abundance / Richness of wildlife	Mangrove communities in Deep Bay are known to support a high diversity of fauna, especially invertebrates.	Mangrove communities in Deep Bay are known to support a high diversity of fauna, especially invertebrates, though limited number of microhabitats and young age in these stands suggest that diversity is likely to be lower than average.	Probably has moderate species diversity, but this is limited by relatively dry conditions.
Ecological value	Part of, and continuous with the extensive mangroves around Deep Bay and considered to be of High Ecological Value.	Small size, relatively young age and linear nature render the ecological value less than the large mangrove stands of Deep Bay, nevertheless still considered to be of Moderate to High Ecological Value.	Ecological value lower than most mangrove areas in Deep Bay due to colonisation of terrestrial vegetation around fringes. Considered to be of Moderate Ecological Value.

Intertidal Mudflat

4.7.37 A small part of the extensive Inner Deep Bay intertidal mudflat falls within the northern part of the Assessment Area. The intertidal mudflats are a major and important element of the Deep Bay wetland ecosystem and are of international importance for wetland birds, especially migrant shorebirds.

Table 4.26: Ecological evaluation of Intertidal Mudflat

Criteria	Intertidal Mudflat
Naturalness	Natural habitat formed by sediment accretion.
Size	3.2ha in the Assessment Area, none within the Project Site
Diversity	Largely unvegetated with a few colonising mangroves.
Rarity	Uncommon habitat in Hong Kong. Utilised by a number of bird species of conservation importance.
Re-creatability	Extremely difficult to recreate artificially but would develop naturally in time assuming currents and pattern of sediment deposition remained unchanged.
Fragmentation	Not fragmented and part of a much larger area in Inner Deep Bay.
Ecological linkage	Ecologically linked to surrounding wetland habitats.
Potential value	Little scope for habitat enhancement but would benefit from reduction in disturbance by anthropogenic activities.
Nursery/ breeding ground	Breeding ground for a number of habitat-restricted invertebrate species and fish.
Age	A natural feature, likely to be several hundred years old, though distribution has doubtless changed due to sedimentation, tides and currents.
Abundance/ Richness of wildlife	Very high abundance and diversity of wildlife, especially wetland birds.
Ecological value	Very High Ecological Value.

Grassland

4.7.38 There are a number of relatively small areas of terrestrial grassland within the Assessment Area, some of which lies within the Project Site. Most of these areas have developed on areas which were formerly fishponds, some of which have been filled. Terrestrial grassland formed by habitat succession is also found in and around some of the inactive fishponds, largely in the south of the Assessment Area. This grassland mostly comprises Panicum maximum and Brachiaria mutica but some damper areas may support small, isolated patches of reed and Typha angustifolia. This terrestrial grassland is utilised by a limited diversity of fauna, mostly comprising species which are widespread in lowland open country habitats. This habitat is an indicator that unmanaged wetland habitat will eventually be lost through habitat succession in the long term.

Table 4.27: Ecological evaluation of Grassland

Criteria	Grassland
Naturalness	Habitat has evolved from natural vegetation colonisation into man-made fishponds some of which have been filled.
Size	8.1ha in the Assessment Area, including 0.8ha with NSW Project Site.
Diversity	Low to moderate diversity of vegetation; some wetland plant species may be present in damper areas.
Rarity	Overgrown and filled fishponds are not uncommon in Hong Kong, especially around Deep Bay. Not utilised by significant numbers of any fauna species of conservation importance.
Re-creatability	Easily re-creatable.
Fragmentation	Fairly fragmented.
Ecological linkage	Some linkage to surrounding fishponds for habitat generalist species.
Potential value	Little scope to increase value, but some areas could readily be converted to fishpond, marsh or reedbed, all of which would have higher ecological value.
Nursery/ breeding ground	No known breeding grounds.
Age	Fairly recently colonised in abandoned fishponds.
Abundance/ Richness of wildlife	Low abundance and diversity of wildlife.
Ecological value	Small, fragmented areas with limited diversity and numbers of flora and fauna, thus this habitat is of Low Ecological Value.

Managed Grassland

4.7.39 An area of grassland, created during the channelisation of the SPR, is located on the western side of the NSW site. Short grass is maintained in this area, which is regularly highly disturbed by human activities, including picnicking, dog walking and flying of model aircraft. The diversity of fauna recorded using this habitat area was very low.

Table 4.28: Ecological evaluation of Managed Grassland

Criteria	Managed Grassland (at West of NSW)
Naturalness	Not natural: terrestrial grassland in this area was created and is managed by human activity.
Size	1.3ha, entirely within NSW.
Diversity	Very low plant species diversity, dominated by a small number of grass species.
Rarity	A common habitat in Hong Kong.
Re-creatability	Easily re-creatable.
Fragmentation	Block not internally fragmented but fragmented from other terrestrial habitats.
Ecological linkage	Grassland shows little ecological linkage to surrounding habitats.
Potential value	Value could be increased with lower levels of human disturbance or by allowing areas to flood seasonally.
Nursery/ breeding ground	No known breeding grounds.
Age	A fairly young, highly managed habitat.
Abundance/ Richness of wildlife	Low abundance and diversity of wildlife.
Ecological value	A small area of common habitat supporting a low faunal diversity; the grassland is considered to be of Low Ecological Value.

Plantation

4.7.40 Plantation located along the banks of the KTMDC and the SPR, and around the perimeter of the NSW site consists largely of non-native species, especially Casuarina equisetifolia and Eucalyptus camaldulensis. Some recently planted areas to the west of the site include a large number of native Ficus microcarpa. Plantation near Shan Pui Tsuen included village associated species, Averrhoa carambola, Dimocarpus longan and Litchi chinensis, Bombax ceiba, Melia azedarach and Platycladus orientalis.

4.7.41 Plantation on bunds within NSW largely comprises non-native Eucalyptus camaldulensis together with some E. tereticornis and Corymbia citriodora, many mature specimens of which are greater than 15m in height. Also present within the plantations are areas of fruit trees (especially Artocarpus macrocarpon, Dimocarpus longan and Litchi chinensis) and some patches of native trees (such as Macaranga tanarius var. tomemtosa) and bamboos.

4.7.42 Although the plantation on site was not found to contain a high diversity of floral or faunal species, it is important as a roosting site for Great Cormorant. The large Eucalyptus trees are utilised by the majority of Great Cormorants roosting at NSW. These trees are reaching the end of their biological life and need to be replanted as a long-term management exercise.

Table 4.29: Ecological evaluation of Plantation

Criteria	Plantation
Naturalness	Man-made habitat created by planting of trees.
Size	35.2ha, of which 9.7ha is within the Project Site at NSW site
Diversity	Relatively low diversity of flora and fauna.
Rarity	Plantation is common in Hong Kong, although such mature specimens of Eucalyptus spp. are uncommon, especially in the Deep Bay area. One of only four major roost sites for Great Cormorant in Deep Bay.
Re-creatability	Can be re-created in suitable locations, although trees would take a long time to reach maturity.
Fragmentation	Slightly fragmented along bunds.
Ecological linkage	Roosting cormorants show linkage to wetland habitats; otherwise, ecological linkages are poor.
Potential value	Value could be enhanced with greater diversity of native tree species.
Nursery/ breeding ground	Some trees probably used by breeding Collared Crow.
Age	Judging by the size of the trees, Eucalyptus plantations are moderately old (c. 40 to 50 years old) while other plantations are generally fairly young.
Abundance/ Richness of wildlife	Generally low diversity of wildlife but some larger trees provide a very important roost site for Great Cormorant.
Ecological value	Most trees are exotic and the plantation generally supports a low diversity of wildlife: hence that around the perimeter of NSW is generally of Low Ecological Value. However, some plantation on bunds within the NSW site is of High Ecological Value due to its importance as a Great Cormorant roost site.

Developed Area

4.7.43 The western part of the Assessment Area includes the developed area of Yuen Long Industrial Estate and in the south of the Area there are areas of urban and village housing. Within the Project Site there are some temporary structures at NSW and LC, some of which are inhabited. Dogs are present at most of these and are likely to cause some disturbance to birds using nearby fishponds. Some of these inhabited structures are surrounded by a small area of cultivation, including small-scale vegetable cultivation as well as fruit trees (for example Artocarpus heterophyllus, Litchi chinensis and Dimocarpus longan). The few other faunal species associated with these buildings are common and widespread species of anthropogenic habitats such as Eurasian Tree Sparrow. Other developed areas present at NSW are a tarmac wetland access road which runs alongside the KTMDC and SPR, together with associated sea wall and maintenance access points. These areas have no significant ecological value.

Table 4.30: Ecological evaluation of Developed Area

Criteria	Developed Area
Naturalness	An artificial, man-made habitat.
Size	Occupies a large part of the Assessment Area, especially to the west but only a small area on the Project Site (3.3ha at NSW and 0.4ha at LC).
Diversity	A low diversity of vegetation managed for cultivation around houses.
Rarity	A common habitat in Hong Kong, not utilised by significant numbers of any species of conservation importance.
Re-creatability	Easily re-creatable.
Fragmentation	Urban area is not fragmented in the Assessment Area, but developed area in the Project Site consists of small, isolated buildings, which are relatively fragmented. Road is not fragmented but is, itself, a cause of fragmentation.
Ecological linkage	Urban developed area has no significant ecological linkages. Some ecological linkage of village areas to adjacent fishponds but occupied buildings (especially with dogs) cause disturbance to nearby habitats. No ecological linkages for road and sea wall.
Potential value	Little scope for an increase in ecological value.
Nursery/ breeding ground	Some temporary structures on the Project Site are known to have been used by breeding White-shouldered Starlings (though breeding was not recorded in the recent surveys).
Age	Major developed areas have been urban for many years. Temporary structures on the Project Site are relatively recent. Any vegetation experiences regular disturbance.
Abundance/ Richness of wildlife	A very few common species (such as Eurasian Tree Sparrow) associated with buildings and anthropogenic habitats.
Ecological value	Developed areas are all of Low Ecological Value.

4.7.44 A summary of species of conservation importance recorded within the Assessment Area from both the literature review and Ecological Baseline surveys is shown in Table 4.31 below. A full list of species is provided in Appendix 4.2a and Appendix 4.2b. Species are further evaluated later (Sections 4.9) to address whether they qualify as target species for mitigation measures.

Table 4.31: Evaluation of Faunal Species of Conservation Importance

Species	Conservation and Protection Status¹	Local Distribution²	Status in HK²
Mammals
Eurasian Otter	RC; RLCV(EN); CSMPS(II); CITES(I); Cap.170; Cap.586	Restricted	Rare
Pallas’s Squirrel	Cap.170	Widespread	Very Common
Small Indian Civet	RLCV(VU); CSMPS(II); CITES(III); Cap.170; Cap.586	Widespread	Very Common
Small Asian Mongoose	RLCV(VU); CITES(III); Cap.170; Cap.586	Widespread	Uncommon
Leopard Cat	RLCV(VU); CSMPS(II); CITES(II); Cap.170; Cap.586	Widespread	Uncommon
Japanese Pipistrelle	Cap.170	Widespread	Very Common
Birds
Mallard	RC	Localized	Uncommon
Northern Shoveler	RC	Localized	Abundant
Eurasian Wigeon	RC	Localized	Abundant
Northern Pintail	RC	Localized	Abundant
Eurasian Teal	RC	Localized	Common
Tufted Duck	LC	Localized	Abundant
Little Grebe	LC	Localized	Common
Eurasian Spoonbill	LC; RLCV(NT); CITES(II); CSMPS(II); Cap.586	Localized	Uncommon
Black-faced Spoonbill	PGC; RLCV(EN); IUCN(EN); CSMPS(II)	Localized	Common
Eurasian Bittern	RC	Localized	Uncommon
Yellow Bittern	(LC)	Localized	Uncommon
Cinnamon Bittern	LC	Localized	Uncommon
Black-crowned Night Heron	(LC)	Widespread	Common
Striated Heron	(LC)	Localized	Uncommon
Chinese Pond Heron	PRC (RC)	Widespread	Common
Eastern Cattle Egret	(LC)	Widespread	Common
Grey Heron	PRC	Localized	Common
Purple Heron	RC	Localized	Uncommon
Great Egret	PRC (RC)	Localized	Common
Intermediate Egret	RC	Localized	Uncommon
Little Egret	PRC (RC)	Widespread	Common
Great Cormorant	PRC	Localized	Common
Western Osprey	RC; RLCV(NT); CITES(II); CSMPS(II); Cap.586	Localized	Common
Black-winged Kite	LC; RLCV(NT); CITES(II); CSMPS(II); Cap.586	Localized	Uncommon
Crested Serpent Eagle	(LC); RLCV(NT); CITES(II); CSMPS(II); Cap.586	Widespread	Uncommon
Greater Spotted Eagle	GC; RLCV(EN); IUCN(VU); CITES(II); CSMPS(II); Cap.586	Localized	Rare
Eastern Imperial Eagle	GC; RLCV(EN); IUCN(VU); CITES(II); CSMPS(I); Cap.586	Localized	Rare
Besra	CSMPS(II); CITES(II); Cap.586	Widespread	Common
Eastern Marsh Harrier	LC; RLCV(NT); CITES(II); CSMPS(II); Cap.586	Localized	Common
Pied Harrier	LC; RLCV(NT); CITES(II); CSMPS(II); Cap.586	Localized	Uncommon
Black Kite	(RC); CITES(II); CSMPS(II); Cap.586	Widespread	Common
Eastern Buzzard	CSMPS(II); CITES(II); Cap.586	Widespread	Common
Watercock	RC	Localized	Uncommon
Eurasian Coot	RC	Localized	Uncommon
Black-winged Stilt	RC	Localized	Common
Grey-headed Lapwing	LC	Localized	Rare
Pied Avocet	RC	Localized	Abundant
Pacific Golden Plover	LC	Localized	Common
Grey Plover	RC; IUCN(VU)	Localized	Abundant
Little Ringed plover	(LC)	Localized	Common
Kentish Plover	RC	Localized	Common
Pheasant-tailed Jacana	LC	Localized	Rare
Whimbrel	LC	Localized	Common
Black-tailed Godwit	RC; IUCN(NT)	Localized	Abundant
Curlew Sandpiper	RC; IUCN(VU)	Localized	Abundant
Temminck's stint	LC	Localized	Uncommon
Long-toed Stint	LC	Localized	Common
Red-necked stint	LC; IUCN(NT)	Localized	Abundant
Terek Sandpiper	RC	Localized	Common
Common Redshank	RC	Localized	Abundant
Marsh Sandpiper	RC	Localized	Abundant
Wood Sandpiper	LC	Localized	Common
Spotted Redshank	RC	Localized	Uncommon
Common Greenshank	RC	Localized	Abundant
Oriental Pratincole	LC	Localized	Common
Black-headed Gull	PRC	Localized	Abundant
Lesser Black-backed Gull	LC	Localized	Common
Greater Coucal	CSMPS(II)	Widespread	Common
Lesser Coucal	CSMPS(II)	Localised	Uncommon
Collared Scops Owl	CITES(II); CSMPS(II); Cap.586	Widespread	Common
Asian Barred Owlet	CITES(II); CSMPS(II); Cap.586	Widespread	Common
White-throated Kingfisher	(LC)	Widespread	Common
Black-caped Kingfisher	(LC); IUCN(VU)	Localized	Uncommon
Pied Kingfisher	(LC)	Localized	Uncommon
Common Kestrel	CITES(II); CSMPS(II); Cap.586	Widespread	Common
Peregrine Falcon	(LC); RLCV(NT); CITES(I); CSMPS(II); Cap.586	Localized	Uncommon
Black-naped Oriole	LC	Widespread	Uncommon
Ashy Drongo	LC	Localized	Uncommon
Collared Crow	LC; RLCV(NT); IUCN(VU)	Localized	Common
Chinese Penduline Tit	RC	Localized	Common
Thick-billed Warbler	LC	Localized	Scarce
Pallas's Grasshopper Warbler	LC	Localized	Common
Lanceolated Warbler	RLCV(NT)	Widespread	Common
Zitting Cisticola	LC	Widespread	Common
Golden-headed Cisticola	GC	Localized	Uncommon
Red-billed Starling	GC	Localized	Abundant
White-cheeked Starling	PRC	Localized	Uncommon
White-shouldered Starling	(LC)	Localized	Common
Common Starling	LC	Localized	Scarce
Bluethroat	LC	Localized	Uncommon
Red-throated Pipit	LC	Localized	Common
Buff-bellied Pipit	LC	Localized	Uncommon
Chestnut-eared Bunting	LC	Localized	Uncommon
Yellow-breasted Bunting	RC; RLCV(EN); IUCN(CR)	Localized	Common
Reptiles
Reeves' Turtle	GC; RLCV(EN); IUCN(EN); CITES(III); Cap.170; Cap.586	Widespread	Common
Chinese Stripe-necked Turtle	RLCV(EN); IUCN(CR); CITES(III); Cap.586	n/a	Introduced
Chinese Soft-shelled Turtle	GC; RLCV(EN); IUCN(VU); Cap.170	Localized	Rare
Common Rat Snake	PRC; RLCV(EN); CITE(II); Cap.586	Widespread	Common
Many-banded Krait	PRC; RLCV(EN)	Widespread	Common
Odonates
Dingy Dusk-hawker	LC	Localized	Uncommon
Mangrove Skimmer	GC; IUCN(VU)	Widespread	Uncommon
Ruby Darter	LC	Widespread	Common
Scarlet Basker	LC	Localized	Common
Butterflies
Common Awl	LC	Localized	Very Rare
Pale Palm Dart	LC	Widespread	Rare
Danaid Egg-fly	LC	Widespread	Uncommon
Small Cabbage White	-	Widespread	Rare
Fireflies
MPBWF	IUCN(EN)	Localized	Uncommon

1. Conservation and protection status refers to Fellowes et al. (2002), RLCV (Jiang et al. 2016), China State Major Protection Status, IUCN (2024), CITES (2024), Cap.170 and Cap.586.

a. Conservation status by Fellowes et al. (2002): LC = Local Concern; PRC = Potential Regional Concern; RC = Regional Concern; PGC = Potential Global Concern. Letters in parentheses indicate assessment is on the basis of restrictedness in breeding and/or roosting sites rather than in general occurrence.

b. Conservation Status by Red List of China’s Vertebrates (RLCV) (Jiang et al. 2016): NT = Near Threatened; EN = Endangered.

c. Conservation status by IUCN (2024): NT = Near Threatened; EN = Endangered; VU = Vulnerable; CR = Critically Endangered.

d. Protection status by China State Major Protection Status (CSMPS): II = Class II Protected Species in China.

e. Protection status by CITES (2024): I = Listed in Appendix I; II = Listed in Appendix II

f. All wild birds in Hong Kong are protected under Cap. 170 Wild Animals Protection Ordinance.

g. Cap.586 Protection of Endangered Species of Animals and Plants Ordinance.

2. Distribution and rarity refer to AFCD (2024a), Chan et al. (2011), Tam et al. (2011). Distribution and status of bird species refers to The Avifauna of Hong Kong, (G.J.Carey, Editor), HKBWS (2023).

4.8 Identification and Evaluation of Potential Impacts

Introduction

4.8.1 This section provides an assessment of the significance of direct and indirect, primary and secondary, on-site and off-site and fragmentation impacts predicted to be caused by the proposed Project during its construction and operational phases in accordance with the requirements of Annex 8 of the EIAO-TM. It details the measures taken at the design stage to avoid and minimise impacts together with the measures proposed to be taken during the construction and operational phases to further avoid, minimise and compensate for any remaining impacts such that the net ecological consequences will be positive. For the purposes of the impact assessment, it is considered that current conditions represent the baseline against which potential impacts are assessed.

Description of the Proposed Development

4.8.2 The proposed development project involves the construction of 2,521 housing units comprising a mixture of three storey houses and apartment blocks ranging from 19 to 25 residential storeys, together with associated facilities including a private underground sewage pumping station and a proposed bridge. The Development Site footprint (Figure 4.13) is proposed to be 11.6ha, 10.7ha of which is currently within the wetland area (i.e. Open Water, Reedbed, Brachiaria Marsh, Typha Marsh, Grassy Bund and Wooded Bund).

4.8.3 On the western side of NSW WEA, approximately 3.3ha existing managed grassland and plantation area currently in public use will be preserved in the proposed public park. A visitor centre will be built, and wooden jetty will be reprovisioned within the proposed public park. There will be no loss of wetland area due to the proposed public park and no significant increase in human disturbance to wetland habitats is anticipated comparing with the current status of the public area. The net effect of the changes to habitat areas and water areas within the Project Site will be ecologically positive as is enumerated in the tables below and is detailed further in the following sections and the CMP.

Table 4.32: Current and proposed habitats in the Project Site (ha)

Habitat	Area of Current and Proposed Habitats (ha)
	Total
	Current	Proposed	Net Change
Tidal Watercourse	0.5	0.5	0
Fishpond	58.1	48.4	-9.7
Open Water	2.6	1.6	-1.0
Lily Pond	0	4.1	+4.1
Shallow Tidal Pond	0	5.1	+5.1
Reedbed	41.0	49.4	+8.4
Brachiaria Marsh	7.0	6.9	-0.1
Cyclosorus Marsh	4.1	4.1	0
Typha Marsh	9.8	0	-9.8
Grassy Bund	16.9	11.1	-5.8
Wooded Bund	10.4	9.7	-0.7
Bare Soil Bund	0.9	0	-0.9
Mangrove	11.0	11.3	+0.3
Grassland	0.8	0	-0.8
Managed Grassland	1.3	1.3	0
Plantation	9.7	9.5	-0.2
Developed Area	3.5	14.8	+11.0
Total	177.3	177.3	0

Note:

1. Totals, subtotals, and net change values may not match the sum of individual habitat areas and current area minus proposed area due to rounding.

2. Part of the proposed bridge falls outside of the Project Site but is shown on all relevant plans and included within all relevant assessments. The Proposed Visitor Centre and Public Park fall within the Project Site.

4.8.4 As is stated in the EIAO-TM Annex 16, 'potential impacts shall be avoided to the maximum extent practicable such as [by] adopting suitable alternatives’. Accordingly, ways to avoid impacts were identified and followed wherever possible during the design stage.

4.8.5 Figure 4.6, which was first produced in AEC (2016) shows the key ecological constraints at NSW, most notably, in this context, the Great Cormorant roost locations, and flightlines from the then active TSL egretry, based on previous surveys conducted between 2011 and 2015. The figure demonstrates how the configuration of the development was carefully planned in the early design stage well before the EIA Study to avoid or minimise impacts on flightlines from the nearby egretries, and on the Great Cormorant roost and flightlines between the roost and the cormorants’ Deep Bay foraging areas.

4.8.6 As is demonstrated in Figures 4.7, 4.8 and 4.9, the surveys conducted between 2021 and 2022 confirmed that the key ecological constraints remained largely the same as those identified in AEC (2016). A significant exception is that the TSL egretry has been abandoned and is no longer considered an ecological constraint; the newly established Shan Pui River egretry, the location and flightlines from which are shown in Figure 4.9, is remote from the development site and neither the egretry nor its flightlines will be adversely impacted by the project.

4.8.7 The following paragraphs detail measures proposed to mitigate for the potential ecological impacts identified. As noted above, proposed mitigation seeks wherever possible to avoid causing the potential impacts. Where this is not possible, measures are proposed to minimise the impact or if necessary to compensate for the impact.

4.8.8 The proposed Development Site is located in the southernmost portion of the site, furthest away from MPNR and Deep Bay, and fishpond filling has been minimised subject to other development constraints and the planning intention for the site. In addition, both the location of the proposed development and location of the proposed bridge have avoided direct impacts to the habitat (mangrove and marsh) of MPBWF in the southwest of NSW (Figure 4.12b). Footings of the proposed bridge will not be within the SPR, there will be no direct loss of mangrove and marsh habitat due its construction, and the alignment avoids the area where most MPBWF were found. Further measures to minimise indirect impacts to this species are considered in Section 4.9.

4.8.9 As required under the OZP in the OU(CDWEA1) zone, the Development Site is proposed to be developed together with the Site of Special Scientific Interest (1) (SSSI(1)) zone on the Mai Po and Fairview Park OZP at LC in a comprehensive and integrated manner. In addition, again as required under the OZP, the development scheme for the project includes a Conservation and Management Plan (CMP) which details a wetland conservation and enhancement scheme, including its detailed design, wetland buffer proposals, a long-term maintenance and management plan, and monitoring and implementation mechanism. This CMP for the NSW WEA and the LCNR is included in this Ecological Impact Assessment (see Appendix 4.5).

4.8.10 As is detailed in the project programme, to minimise the adverse impacts of the project, habitat enhancement measures will commence in advance of the commencement of construction of the Development Site, such that compensatory habitat will be available for fauna which is displaced by loss or disturbance to habitat in or near to the Development Site. In addition, the Development Site will be phased with the works area restricted at any one time. The timing of the implementation of the CMP is critical in mitigating for the construction phase impacts arising from the development.

Direct Habitat Loss

4.8.11 Direct habitat loss arising from the project will include both habitat which will be lost under the footprint of the development (primary impact) and habitat which will be lost due to the conversion of one type of wetland habitat to another to mitigate for the wetland area to be developed (secondary impact). There will be no net permanent loss of the area of plantation utilised by roosting cormorants and reedbed because of the project. There will be no loss of existing mangrove habitat in both NSW and LC. There will also be no loss of tidal watercourse; the proposed bridge to be built over the SPR to provide access to the development will have no piers or other structural elements within the watercourse, and while some disturbance-sensitive fauna may be deterred from using the habitat under and near the proposed bridge, its ecological function for other taxa such as fish and freshwater invertebrates will be unimpaired. The loss of reedbed under the Development Site footprint will be compensated by the provision and enhancement of reedbeds elsewhere within the Project Site, whilst loss of open water area will be compensated by re-profiling fishponds and removing and narrowing bunds of limited ecological function. There will be a permanent loss of fishpond habitat; this will be compensated in terms of ecological function by enhancement of much of the remaining fishpond habitat including conversion of 4.1ha to lily ponds, as well as conversion of 5.1ha to shallow tidal pond habitat.

4.8.12 The decision to compensate any loss in area of non-open water habitats on a like-for-like basis in terms of area as well as function, and to restrict the loss of wetland habitat area to open water habitat, recognises that there are proven examples, notably at Lok Ma Chau Ecological Enhancement Area (LMC EEA), in Hong Kong of the successful implementation of ecological enhancement measures designed to meet quantitative targets for wetland fauna at open water habitat and successful management of tidal ponds at MPNR. Enhancement of open water habitat can be achieved through a variety of techniques including reprofiling of bunds, scheduled drain-down for waterbirds and management of fish stocks. However, unlike open water habitats, evidence of the quantitative achievement of targets in other wetland habitats is more limited and/or more taxon-specific, therefore impacts to these habitats are compensated in terms of both area and function.

4.8.13 The areas of direct habitat loss under the Development Site footprint and direct habitat loss due to changes in wetland habitat type are detailed in the table below. Impacts of habitat loss (permanent and temporary, and primary and secondary) are evaluated in the following tables. The design of the proposed bridge crossing the SPR does not include any footings in the river, and consequently construction of the proposed bridge will only result in loss of plantation (see Table 4.32); whilst some fauna, such as large waterbirds, may be excluded, the ecological function of the channel will not be reduced in other respects. Indirect impacts from the proposed bridge are discussed below.

Table 4.33: Direct permanent habitat loss under the Development Site footprint

Habitat	Development footprint (ha)
Open Water	0.9
Reedbed	6.2
Brachiaria Marsh	0.1
Typha Marsh	2.0
Grassy Bund	0.9
Wooded Bund	0.7
Grassland	0.6
Plantation	0.4
Total	11.6

Table 4.34: Area to be converted to other wetland habitats

Habitat	Converted to other wetland habitats (ha)
Fishpond	17.4
Open Water	0.1
Reedbed	0.0
Brachiaria Marsh	0.0
Typha Marsh	7.8
Grassy Bund	4.9
Bare Soil Bund	0.9
Grassland	0.2
Plantation	0.0
Developed Area	2.3
Total	33.7

Direct Loss of Fishpond and Open Water

4.8.14 Under the proposed development footprint there will be a direct loss of open water habitat of 0.9ha. It is also proposed to convert a small area of open water elsewhere in NSW into reedbed, resulting in a total net loss of 1.0ha of open water. Likewise, whilst there will not be any direct loss of fishpond under the proposed development footprint, a total of 17.4ha of fishpond in LC will be converted to lily pond, shallow tidal pond and reedbed. Proposed enhancements at NSW WEA and LCNR include conversion of unwanted Typha marshes into fishponds and removal of some internal bunds to permit provision of larger wetland areas (see Figure 4.14). The net loss of fishpond area would be 9.7ha, in the form of conversion to other wetland habitats.

4.8.15 The areas which will be converted support a small number but moderate diversity of wetland bird species. Habitat loss to species which are unlikely to utilise reedbed and marshes habitats in significant numbers will be mitigated by their being targets of enhancement measures to be undertaken in areas of retained fishpond habitat. Most bird species that use fishpond habitat will also use shallow tidal pond and lily pond habitats hence would not be impacted by this change. Little Grebe will probably not use shallow tidal ponds, but particularly favours lily ponds.

4.8.16 Impacts of fishpond and open water habitat loss are evaluated below.

Table 4.35: Direct impacts on Fishpond and Open Water

Criteria	Fishpond	Open Water
Habitat Quality	Active fishponds are of High ecological value; inactive fishponds are of Moderate to High ecological vale. .	Open water areas are of Moderate to High ecological value.
Species	Moderate diversity of waterbirds including several species of conservation importance, usually in relatively low numbers but occasional high counts of ardeids when ponds are drained. Diversity of other fauna low.	Moderate diversity of waterbirds including several species of conservation importance, usually in relatively low numbers. Diversity of other fauna low.
Size/Abundance	There will be a net loss of 9.7ha resulting from the conversion into other habitats which will be used by all species, except Little Grebe which is unlikely to use shallow tidal ponds. Utilised by small numbers but a moderate diversity of bird species including species of conservation importance which are unlikely to use reedbed in significant numbers.	0.9ha is under the development footprint will be lost and 0.1ha of current open water area outside of the development footprint will be converted to reedbed. Utilised by small numbers but a moderate diversity of bird species including species of conservation importance which are unlikely to use reedbed in significant numbers.
Duration	Permanent: fishpond habitats to be lost will be converted to lily pond, shallow tidal pond, and reedbed.	Open water within the development footprint would be permanently lost. The open water habitat to be lost outside the development footprint will be converted to reedbed.
Reversibility	Habitats to be converted could readily be returned to their current state if necessary.	Loss of open water within the development footprint would be irreversible. Habitats to be converted could readily be returned to their current state if necessary.
Magnitude	Magnitude will be large, as a significant portion of the fishpond area in the Project Site will be converted to other wetland habitats.	Small magnitude as the open water habitat loss is of limited size.
Impact Severity	Impact would be of High Severity in the absence of mitigation; but there would be a net gain in ecological function during the operational phase with successful implementation of proposed mitigation measures.	Impact would be of Low to Moderate Severity in the absence of mitigation; but there would be a net gain in ecological function during the operational phase with successful implementation of proposed mitigation measures.

Direct Loss of Reedbeds

4.8.17 A total area of 6.2ha of reedbed will be lost under the Development Site footprint. Reedbed will be planted at LC to compensate for that which will be lost under the Development Site footprint, and all existing Typha marsh within the WEA will be converted into reedbed and fishpond. In addition, 0.1ha of current open water area outside of the development footprint will be converted to reedbed. As a result, there will be a net gain of 8.4ha of reedbed habitat. There will, therefore, be no-net-loss in reedbed area, whilst function of existing reedbed will be enhanced by active management. Accordingly, there will be no net negative impact on reedbed-dependent or reedbed-associated fauna species.

Table 4.36: Direct Impacts on Reedbed

Criteria	Reedbed
Habitat Quality	Reedbed on site is of High ecological value.
Species	Utilised by a number of birds of conservation importance, especially large blocks of wet reedbed which support several bird species which are largely dependent on this habitat.
Size/Abundance	6.2ha of reedbed will be permanently lost in the Development Site; however, conversion of fishpond to reedbed at LC, conversion of open water to reedbed and conversion of Typha marsh to reedbed at NSW will result in a net gain of reedbed area of 8.4ha. Reedbed at NSW is used by small numbers and a relatively low diversity of fauna, however some of these species, notably Yellow Bittern are strongly associated with this habitat and are of conservation importance.
Duration	Reedbeds within the Development Site will be permanently lost.
Reversibility	Loss of reedbed in the Development Site is irreversible but will be compensated by reedbed creation elsewhere. Reedbed areas to be created could readily be returned to their current land uses.
Magnitude	Magnitude will be moderate to high as 6.2ha of reedbed will be permanently impacted.
Impact Severity	Impact from the Development Site footprint would be of Moderate to High Severity in the absence of any mitigation. However, with the proposed provision of compensatory planting of reeds prior to impacting reedbeds under the Development Site footprint, impact will be of Low Severity during construction phase and there will be a net gain in ecological function during the operational phase.

Direct Loss of Marshes

4.8.18 Approximately 0.1ha of Brachiaria Marsh and 2.0ha of Typha Marsh will be permanently lost under the current development footprint. An additional 7.8ha of Typha Marsh in NSW outside of the development footprint will be converted into reedbed and fishpond. Hence, the total loss of Typha Marsh would be 9.8ha. Both of these marsh habitats are dominated by exotic and invasive grass or herb species, with low faunal diversity and abundance recorded. No native fauna species were found to be dependent on these habitats. With the overall net increase of wetland ecological function in the Project Site during the Operational Phase, the impact of the loss of these marshes of lower ecological value is not considered significant, especially when such loss will be the result of conversion into reedbed and fishpond, which are of much higher ecological and conservation value. There will be no direct loss of Cyclosorus marsh, hence it is not considered here.

Table 4.37: Direct impacts on Marshes

Criteria	Brachiaria Marsh	Typha Marsh
Habitat Quality	Brachiaria Marsh to be lost is of Low to Moderate ecological value.	Typha Marsh to be lost is of Low ecological value.
Species	A few bird species of conservation importance (Purple Heron, Little Egret and Chinese Penduline Tit) were found in low numbers in this habitat.	Typha marsh is not a favoured habitat for wetland-dependent bird or other species. Occasionally used by small numbers of bird species of conservation importance but these species are common migrants in wetland habitats in Hong Kong, especially reedbeds.,
Size/Abundance	Only 0.1ha will be lost directly. Low faunal diversity and number were recorded from Brachiaria marsh. No species appeared to rely on or favour this habitat.	A total of 2.0ha will be lost to the development, with another 7.8ha to be converted into reedbed and fishpond. Low faunal diversity and numbers were recorded. No species appeared to rely on or favour this habitat.
Duration	Brachiaria Marsh within the development footprint would be permanently lost.	Typha Marsh within the development footprint and the NSW Project Site would be permanently lost.
Reversibility	Irreversible, but of low significance as the area is very small (0.1ha).	Loss under the development footprint of 2.0ha would be irreversible; other loss would be readily reversible (albeit not desirable).
Magnitude	Low magnitude given the small size.	Low to Moderate magnitude but conversion to other wetland habitats would be beneficial.
Impact Severity	Impact of loss of Brachiaria Marsh would be of Low severity given the small size. With the proposed ecological enhancement elsewhere in the Project Site, there will be a net gain in ecological function during the operational phase.	Impact of loss of Typha Marsh would be of Low severity. Conversion of this habitat into reedbed and fishpond will be a net gain in ecological function during the operational phase.

Direct Loss of Grassy Bund, Wooded Bund and Bare Soil Bund

4.8.19 There will be a direct loss of 0.9ha of grassy bund and 0.7ha of wooded bund under the development footprint, and a loss of 4.9ha of grassy bund and 0.9ha of bare soil bund to be converted to other wetland habitats. None of the wooded bund used by roosting cormorants is under the development footprint and none of this will be lost elsewhere. The bund habitats which will be lost are all of low ecological value and are not utilised by significant numbers of any species of conservation importance, their replacement by other wetland habitats will result in a net ecological gain. Direct impacts of bund loss are evaluated below.

Table 4.38: Direct impacts on Grassy Bund, Wooded Bund and Bare Soil

Criteria	Grassy Bund, Wooded Bund and Bare Soil
Habitat Quality	All bund habitats which will be lost are of Low ecological value.
Species	Grassy bunds are dominated by common pantropical grass species, wooded bunds are largely dominated by planted trees including fruit trees and eucalypts, together with some native pioneer species. Habitats are utilised by some species of conservation importance but generally in small numbers and no species depend on them; bare soil bunds, may sometimes be used by moderate numbers of waterbirds, especially ardeids, when adjacent fishponds are drawn down for harvesting.
Size/Abundance	There would be a net loss of 5.8ha of grassy bund, 0.7ha of wooded bund and 0.9ha of bare soil bund. Utilised by small numbers but moderate diversity of fauna species, none of which are dependent on this habitat.
Duration	All loss would be permanent.
Reversibility	Loss within the development site would be irreversible, loss elsewhere could readily be reversed if desired.
Magnitude	Magnitude would be low to moderate in respect to grassy bund and low in respect to wooded and bare soil bund.
Impact Severity	Impact would be of Low Severity, due to the low ecological value of the habitats, during the construction phase. Following conversion to other wetland habitats, there will be a net ecological gain in function during the operational phase.

Direct Loss of Plantation

4.8.20 There will be a direct loss of 0.4ha of plantation under the Development Site footprint (0.35ha loss) and proposed bridge (<0.1ha loss). None of the plantation to be lost is utilised by roosting cormorants. Impacts on plantation habitat are evaluated below. The loss of plantation will be compensated by planting of trees and bamboos on an island to be formed (from existing bunds) at LC, and screen planting at NSW WEA.

Table 4.39: Direct impacts on Plantation

Criteria	Plantation
Habitat Quality	Plantation which will be directly impacted by development is not used by roosting cormorants and is of Low ecological value.
Species	Most tree species are exotic, but some are native. Shrub and field layers are floristically impoverished due to management regime. Faunal numbers and diversity are low and most species are widespread habitat generalists.
Size/Abundance	0.4ha will be permanently lost within the Development Site and the proposed bridge. However, all other areas will be retained and enhanced, with additional plantation be planted at LC, and screen planting at NSW WEA.
Duration	All loss will be permanent.
Reversibility	Loss of plantation in the Development Site and the proposed bridge is irreversible but will be reversed by compensatory planting elsewhere.
Magnitude	Magnitude will be low as only 0.4ha will be impacted.
Impact Severity	Impact will be of Low Severity during both construction and operational phases as loss of a small area of this widespread habitat will have very limited ecological consequences.

Direct Loss of Grassland and Developed Area

4.8.21 There will be a permanent loss of 0.8ha of terrestrial grassland at NSW and 2.3ha of developed area (largely squatter houses and other temporary structures). These changes will result in a small net gain of habitat of ecological value.

Table 4.40: Direct impacts on Grassland and Developed Area

Criteria	Grassland	Developed Area
Habitat Quality	Low ecological value.	Low ecological value.
Species	Low to moderate diversity of flora; low diversity of fauna.	Low diversity of flora and fauna.
Size/Abundance	0.8ha within NSW Project Site will be permanently loss.	2.3ha within NSW and LC will be permanently lost.
Duration	All loss will be permanent.	All loss will be permanent
Reversibility	Loss of grassland in the Development Site is irreversible.	Reversible, albeit not desirable.
Magnitude	Magnitude will be low as only 0.8ha will be impacted	Negligible.
Impact Severity	Impact will be of Low Severity during both construction and operational phases as loss of a small area of this widespread habitat will have very limited ecological consequences. Net gain of ecological value from the conversion into wetland habitat.	Negligible. Net gain of ecological value from the conversion into wetland habitat.

Disturbance Impacts on Habitats

4.8.22 Disturbance occurs when activities within a Development Site result in a reduction of the value of a habitat outside the site, usually because of fauna being deterred from using the habitat. Examples of disturbance include noise disturbance (which can scare certain species away from suitable habitat), visual disturbance from the presence of humans or vehicles within the site or in surrounding areas, or disturbance to nocturnal species or species roosting in the area at night from lighting resulting in glare to surrounding habitats. Disturbance may occur during construction and/or operation, and the nature and severity of the impact may differ at these times.

4.8.23 The severity of disturbance varies considerably between species (generally larger birds and mammals are most disturbance-sensitive, smaller birds and mammals and other vertebrates are less sensitive, whilst most invertebrates are not sensitive) and activity (breeding individuals are often more prone to disturbance). In practice, for purposes of impact assessment, impacts on larger birds and mammals can be used as a proxy for identification of most disturbance impacts. Open habitats such as fishponds and intertidal mudflats provide less screening to sources of disturbance and species living in such habitat are therefore usually more sensitive to disturbance impacts. Shrubland and woodland, on the other hand, provide screening to the source of disturbance and species using these habitats are often less concerned by the presence of human activity nearby. Reedbed, with its dense vegetation, is particularly effective of providing cover for disturbance-sensitive species; however larger reedbed species, such as Eurasian Bittern and Purple Heron, benefit less from this effect.

4.8.24 At its most severe, disturbance might result in a species entirely vacating a habitat over a large area and/or might result in increased mortality (for example if nesting birds are forced to abandon eggs or dependent young). More often, however, it results in species occurring in lower numbers or less regularly. In some cases, there will be an exclusion zone close to the source of disturbance, surrounded by a zone of reduced density where disturbance drops with increasing distance from the source.

4.8.25 Waterbirds can be sensitive to noise disturbance, for which the impacts from within the site are anticipated to be greatest during the construction phase. Construction of the high-rise buildings (19 – 25 storeys; 66.6 – 85.5 mPD) is expected to have the greatest noise disturbance impact. Table 4.56, details the evaluation of disturbance impacts on birds, provides estimated distances for the exclusion zone and the zone of reduced density with regards to the relevant waterbird and other wetland-associated bird species of conservation importance in the area. These are based on previous EIA reports for development projects in the Deep Bay area, together with estimates for those species for which sensitivity to disturbance has not previously been assessed in the area.

4.8.26 Based on Table 4.56, the most disturbance-sensitive species are estimated to have an exclusion zone during construction of high-rise buildings of 0 - 200m and a zone of reduced density of 200 - 400m from the high-rise buildings (the disturbance source). Disturbance from construction of low-rise buildings (3-storeys, 17.5 mPD) and infrastructure including the proposed bridge over the SPR would be of lower magnitude, and the distances for exclusion and reduced density of the most disturbance-sensitive species are estimated to be 0 - 100m and 100 - 200m respectively. Disturbance to the habitats present is estimated based on these maximum distances.

4.8.27 With regards to disturbance to species, within the exclusion zone it is assumed that all individuals are disturbed and displaced. Meanwhile, within the zone of reduced density, as disturbance is considered to decline progressively from complete exclusion to no impact; it is estimated that half of the individuals potentially present in the zone are disturbed. In addition to noise, there would be the potential for visual disturbance during the construction phase from vehicles, machinery and construction workers operating within the site. The impacts of this would be smaller in magnitude than the impacts of noise disturbance and would be restricted to the areas within the exclusion zone for noise disturbance, so construction phase impacts are calculated only in relation to noise disturbance.

4.8.28 The exclusion zones and zones of reduced density for the most disturbance-sensitive species during the construction phase of the project are shown on Figures 4.16a and 4.16b and form the basis for much of the evaluation of disturbance impacts on habitats as well as the calculations of the area of the exclusion zones and zones of reduced density for waterbirds and hence the area of each habitat which will experience disturbance impacts below. The following tables show the disturbance that will take place during the construction of the proposed bridge (which will have to be completed before construction of the development commences as the proposed bridge will be required for construction phase site access), as well as the disturbance during construction of the Development Site, together with disturbance from operation of the proposed bridge (though the operational disturbance zone for the latter is largely contained with the construction phase disturbance envelope of the former).

4.8.29 It should be noted that since these areas are calculated based on the potential disturbance impacts on the most disturbance-sensitive species, they are highly precautionary in that impactson most species, and hence the overall loss of ecological function of the habitats will be considerably less. Furthermore, the areas do not consider the phasing of the Development Site which will also reduce disturbance.

4.8.30 These zones of exclusion and reduced density apply only to disturbance arising from the construction of the development project; disturbance from modification of wetlands during creation of the WEA and LCNR would be little different either quantitatively or qualitatively, to that which currently takes place during routine aquaculture activities and would not therefore impact significantly on adjacent wetlands. As is noted under Section 1.11 of the CMP (see Appendix 4.5), up to four fishponds are routinely drained at LC at present. The proposed habitat modification works will also be restricted to no more than four fishponds at any one time. The exception to this will be the creation of the shallow tidal pond, which requires the conversion of six ponds and can only be created in one phase. Works will be on adjacent ponds, focusing activity into a smaller area and reducing overall disturbance.

Table 4.41: Area of habitats (ha) disturbed during bridge construction (see Figure 4.16a)

Habitat	Area (ha)
Habitat	Exclusion Zone (0 - 100m)	Zone of Reduced Density (100 - 200m)
Tidal Watercourse	2.4	2.8
Open Water	0.3	0.4
Reedbed	0.9	3.2
Brachiaria Marsh	0.2	0.1
Typha Marsh	0.1	-
Grassy Bund, Wooded Bund	0.5	0.7
Mangrove	0.3	0.9
*All Wetland Sub-total*	*4.8*	*8.2*
Grassland	-	0.3
Plantation	0.6	1.0
Total	5.4	9.4
Note: Aggregate totals may differ from individual habitat totals due to rounding.

Habitat	Area (ha)
Habitat	Exclusion Zone (0 - 100m for low rise and 0 - 200m for high rise)	Zone of Reduced Density (100 – 200m for low rise and 200 – 400m for high rise)
Tidal Watercourse	6.6	5.9
Fishpond	2.5	5.2
Open Water	0.2	1.4
Reedbed	7.0	10.0
Brachiaria Marsh	3.3	2.8
Cyclosorus Marsh	0.9	1.7
Typha Marsh	1.0	1.7
Grassy Bund, Wooded Bund	4.4	5.5
Mangrove	2.8	-
*All Wetland Sub-total*	*28.7*	*34.0*
Grassland	0.1	-
Managed Grassland	-	0.3
Plantation	3.4	11.4
Total	32.3	45.7
Note: Aggregate totals may differ from individual habitat totals due to rounding.

4.8.31 Compared to the construction phase, noise impacts during the operation of the development would be considerably reduced. At this time the main source of disturbance would be from the presence of an increased human population in the area, potentially leading to increased visual disturbance from the presence of people close to wetland habitats, not only in the Development Site footprint but potentially also in other nearby areas. Predicted operational phase disturbance from these sources is detailed below.

Habitat	Area (ha)
Habitat	Exclusion Zone (0 - 100m)	Zone of Reduced Density (100 - 200m)
Tidal Watercourse	2.9	5.3
Fishpond	0.9	2.2
Open Water	0.1	0.9
Reedbed	5.9	5.7
Brachiaria Marsh	2.2	2.3
Cyclosorus Marsh	0.5	1.4
Typha Marsh	1.0	0.2
Grassy Bund, Wooded Bund	4.2	2.9
Mangrove	2.2	0.6
*All Wetland Sub-total*	*18.3*	*21.5*
Grassland	0.1	-
Plantation	2.2	1.9
Total	20.6	23.5
Note: Aggregate totals may differ from individual habitat totals due to rounding.

4.8.32 Since the wetlands of the WEA would be secured fusing a suitable chain-link fence with lockable gates at all access points and both public access and aquaculture activities would be regulated during operation, disturbance arising from human activities would be limited and lower than its current unregulated level, and much of the WEA would have no human disturbance except that required for wetland monitoring and maintenance. There may be an increase in human activity along NSW Road leading to a slight increase in disturbance along the SPR and KTMDC, although waterbirds in these channels are already habituated to relatively high levels of human activity compared to other wetlands in Hong Kong, and the presence of mangroves along the sides of the channel provides a degree of screening between the channel and the public road.

4.8.33 Although the proposed bridge will become the main vehicular access to the Development Site, the Visitor Centre and the Public Park, motor vehicle access will not be possible between the existing NSW Road and the proposed bridge across the SPR. A control barrier will be installed to allow only a shuttle bus service to the Visitor Centre, emergency service vehicles, vehicles with permit, cyclists and pedestrians to go from the bridge to NSW Road. The connection will not be available for use by the public as a motor vehicle access. With no significant increase in traffic, it is anticipated there will be no significant increase in disturbance from motor vehicles along NSW Road. For safety reasons it is proposed to provide a cycle lane on the proposed bridge across the SPR, however no cycle track is proposed along, or besides, NSW Road (where there is no existing cycle track) accordingly, numbers of cyclists are not predicted to increase significantly and hence no significant impacts arising from cyclists’ activities are predicted. However, in view of the difficulty of predicting whether numbers of cyclists may change, and/or other increases in human activity may occur for reasons unrelated to the Project, mitigation measures to screen any additional disturbance impacts from these sources are proposed on a precautionary basis.

4.8.34 Although human activity is expected to increase around NSW (but not within the WEA), disturbance at LC would be lower during operation than is currently the case, due to a reduction in human activity (including the presence of people on site at night) and the exclusion of dogs and cats.

4.8.35 Whilst most terrestrial invertebrates are not sensitive to most forms of disturbance, light is known to result in disturbance to MPBWF by disrupting communication. Potential disturbance from light is, therefore, a key concern in the evaluation of ecological impacts of the current project in view of the importance of the Assessment Area for the MPBWF, including a population in the mangrove and mangrove associate are to the southwest of the Development Site. However, this population showed a decline, for reasons unknown, between 2015 and 2019/2021, especially away from the old Kam Tin River channel, both absolutely and relative to the total population in the Assessment Area (Figures 4.12c), hence the magnitude of potential impacts have been adjusted compared to previous assessments. While existing vegetation may provide some localised screening from light pollution, the presence of lighting will cause a general increase in overall illumination of the area, especially in combination with existing light sources at nearby Yuen Long.

4.8.36 Given the importance of NSW as a night-time roost for Great Cormorants, the potential for light disturbance impacts to this roost also warrants consideration. Although direct light may act as a deterrent for roosting birds, most species are relatively tolerant of indirect lighting. The Great Cormorant roosts at LMC and Mai Po San Tsuen (established in the winter of 2020/21) do not show signs of disturbance from nearby Shenzhen, and ardeid roosts in Hong Kong are sometimes located close to street lighting or other potential light disturbance. From observations such as these, human activity close to the roosting location is more likely to cause disturbance than increased light levels.

4.8.37 In the impact evaluation tables below, disturbance impacts on habitats are presented in the following order for ease of reference:

i. Overall disturbance impacts in the Assessment Area from the combined effect of noise, light and avoidance of anthropogenic structures and human activity.

ii. light disturbance impacts on particularly sensitive receivers.

iii. Disturbance impacts on sensitive receivers evaluated by habitat.

Table 4.44: Overall disturbance impacts arising from the project

Criteria	Combined Disturbance Impacts of Noise, Light, Human Activity & Avoidance/ Disturbance by Anthropogenic Structures
Habitat Quality	Would impact wetland habitats including open water, tidal watercourse, reedbed, marshes, etc., as well as plantation.
Species	The most sensitive species to noise impacts are larger waterbirds utilising open habitats, including roosting Great Cormorants. Smaller bird species and those using closed habitats would be impacted to a lesser degree.
Size/Abundance	Bridge construction would lead to exclusion of the most disturbance-sensitive species from 4.8ha of wetland habitat (of which 2.7ha is open wetland habitat (tidal watercourse and open water)) and reduced density in 8.2ha (of which 0.9ha is open water habitat and 2.8ha is channelised watercourse). Residential construction together with concurrent operation of the proposed bridge would lead to exclusion of the most disturbance-sensitive species from a total of 28.7ha of wetland habitat (of which 9.3ha is open wetland habitat (tidal watercourse, fishpond and open water) and reduced density in 34.0ha (of which 12.5ha is open wetland habitat). Operational phase disturbance would lead to exclusion of the most disturbance-sensitive species from a total of 18.3ha of wetland habitat (of which 3.9ha is open wetland habitat (tidal watercourse, fishpond and open water)) and reduced density in 21.5ha of wetland habitats (of which 8.4ha is open wetland habitat). Calculation of the number of birds of conservation importance potentially affected by disturbance during the construction phases of the proposed bridge, the construction phase of the Development Site together with the operation of the proposed bridge, and the operational phase of the development as a whole are detailed in Table 4.53.
Duration	Permanent: impacts would be greatest during the construction phase, especially that of the high-rise housing, but there would be some ongoing impact during operation.
Reversibility	Would be largely reversed during the operational phase as most of the source of noise disturbance will be removed.
Magnitude	The magnitude would be large during construction, especially during piling for the high-rise buildings. Operational impacts would be of lower magnitude.
Impact Severity	Impacts to SPR of High Severity during construction of the proposed bridge, primarily due to disturbance to waterbirds in the channel (but the number of affected individuals would be small to moderate). Impacts to wetland habitats and their fauna of High Severity during construction of the residential area but the number of affected individuals would be small. During operation impacts would be reduced to Moderate Severity because of lower magnitude and smaller area of impact.

Table 4.45: Impacts from light disturbance arising from the project

Criteria	Light Disturbance
Habitat Quality	Would mostly impact habitat close to the Development Site where light levels may increase, including open water, tidal watercourse, Brachiaria marsh, reedbed, mangrove and plantation.
Species	Potential impact to MPBWF and to roosting Great Cormorants.
Size/Abundance	A small population of MPBWF and a large Great Cormorant roost present close to NSW though much of this is beyond the potentially impacted area.
Duration	Impacts would only occur at night. Mostly during the operational phase, when impacts would be permanent.
Reversibility	Readily reversible by removal or effective screening of the source of light disturbance.
Magnitude	The magnitude would be dependent on the brightness and direction of the light and on any screening present (from vegetation or constructed features). Thus, according to location, could be between low and high magnitude. However, most sensitive receivers are not in the areas that may be impacted, hence the impact will be at the lower end of this spectrum.
Impact Severity	Without measures to minimise the magnitude, the impacts would potentially be of Moderate to High Severity if bright lights were directed towards habitat used by the light-sensitive species (mangrove and Brachiaria marsh for the MPBWF, plantation for the Great Cormorant roost); however, with appropriate mitigation measures in place impacts would be of Low Severity.

Table 4.46: Disturbance impacts on Tidal Watercourse

Criteria	Tidal Watercourse
Habitat Quality	The northern section of the SPR is of Very High ecological value due to its importance for large waterbirds, but the southern section, is of Low to Moderate ecological value as it is used by relatively few waterbirds and is already disturbed because its proximity to Yuen Long. The KTMDC is of Moderate ecological value but is also somewhat disturbed by adjacent village housing.
Species	Moderate diversity of waterbirds of conservation importance, especially ducks and ardeids. A favoured habitat for some bird species of conservation importance, but the southern section of the SPR is not a favoured area of this habitat for these species and the number of individuals which may be disturbed is moderate (a mean of approximately 150 individuals of species of conservation importance), see Table 4.57 for details of the numbers of each species which may be disturbed.
Size/Abundance	2.4ha is present in the exclusion zone and 2.8ha in the zone of reduced density during the bridge construction phase; 6.6ha is in the exclusion zone and 5.9ha in the zone of reduced density during residential construction; falling to 2.9ha in the exclusion zone and 5.3ha in the zone of reduced density during the operational phase. However, numbers of birds present in this area are small compared to the section of channel further north (at the confluence of the SPR and KTMDC).
Duration	Permanent: greatest during the construction phase.
Reversibility	Will decrease significantly once the construction phase ends.
Magnitude	Magnitude will be moderate during bridge construction as the total area impacted is small relative to the area of this habitat which is present in the Assessment Area and it is already disturbed significantly and the number of individuals which would be affected is moderate. There would be a moderate increase to disturbance in the watercourses during Development Site construction and during operation because of increased human activity near the development; however this would be limited somewhat by screening provided by adjacent mangroves and the magnitude of the increased disturbance would therefore be low to moderate during operation.
Impact Severity	Impacts to SPR of High Severity during construction of the proposed bridge, primarily due to disturbance to waterbirds. Impacts to wetland habitats and their fauna of High Severity during construction of the residential area. During operation impacts would be reduced to Moderate Severity because of lower magnitude and smaller area of impact.

Table 4.47: Disturbance impacts on Fishpond habitat arising from the project

Criteria	Fishpond
Habitat Quality	Fishpond habitat is of Moderate to High ecological value
Species	Birds of conservation importance which utilise fishpond areas are potentially sensitive to construction and operational phase disturbance due to noise and human activities/presence
Size/Abundance	During the bridge construction phase, no fishpond habitat is present in the exclusion zone or the zone of reduced density; during the residential construction phase, 2.5ha is in the exclusion zone and 5.2ha in the zone of reduced density; during the operational phase 0.9ha is in the exclusion zone and 2.2ha in the zone of reduced density. The number of waterbirds, especially the most disturbance-sensitive species, within these zones is generally low (see Table 4.57).
Duration	Construction phase disturbance would be temporary, operational phase would be permanent.
Reversibility	Will decrease significantly once construction has been completed.
Magnitude	The magnitude of disturbance near NSW during both bridge and residential construction phases would be low because a relatively small area is affected, most of which is already disturbed and supports low numbers of large waterbirds. Increased human activity during operation would have a disturbance impact of low magnitude because public access and fisheries activities would be regulated. There would be a reduction in disturbance of moderate magnitude at LC during both construction and operation.
Impact Severity	Impact severity at NSW would be Low. Moderate improvement at LC due to reduced disturbance from people and dogs on site.

Table 4.48: Disturbance impacts on Open Water arising from the project

Criteria	Open Water
Habitat Quality	Open water habitat is of Moderate to High ecological value.
Species	Birds of conservation importance which utilise open water areas are potentially sensitive to construction and operational phase disturbance due to noise and human activities/presence; however, large waterbirds (which are more sensitive to disturbance) only utilise this microhabitat in small numbers.
Size/Abundance	During the bridge construction phase, 0.3ha of open water habitat is present in the exclusion zone and 0.4ha is in the zone of reduced density; during the residential construction phase 0.2ha is in the exclusion zone and 1.4ha in the zone of reduced density; during the operational phase only 0.1ha is in the exclusion zone and 0.9ha in the zone of reduced density. The number of waterbirds, especially the most disturbance-sensitive species, within these zones is relatively low (see Table 4.57).
Duration	Construction phase disturbance would be temporary, operational phase would be permanent.
Reversibility	Magnitude will decrease significantly once construction has been completed.
Magnitude	The magnitude of disturbance near NSW during both proposed bridge and residential construction phases would be low because a relatively small area is affected, most of which is already disturbed and supports low numbers of large waterbirds. Increased human activity during operation would have a disturbance impact of low magnitude because public access would be regulated.
Impact Severity	Impact severity would be Low.

Table 4.49: Disturbance impacts on Reedbed habitat arising from the project

Criteria	Reedbed
Habitat Quality	Reedbed is of High ecological value.
Species	Most habitat-restricted bird species of conservation importance using reedbeds are smaller birds such as Yellow Bittern and Chinese Penduline Tit that are less sensitive to disturbance. Two larger species, Eurasian Bittern and Purple Heron are sensitive to disturbance.
Size/Abundance	Just 0.9ha is located within the exclusion zone and a further 3.2ha is located in the zone of reduced density during bridge construction; however, 7.0ha is present in the exclusion zone and 10.0ha in the zone of reduced density during residential construction; during the operational phase 5.9ha is in the exclusion zone and 5.7ha in the zone of reduced density. The abundance of disturbance-sensitive faunal species within this habitat (Eurasian Bittern and Purple Heron) is low.
Duration	Construction phase disturbance would be temporary, operational phase would be permanent.
Reversibility	Would decrease significantly once construction has been completed.
Magnitude	Numbers of potentially impacted reedbed species of conservation importance are small and the dense vegetation of reedbed provides a degree of cover for species using this habitat, such that the magnitude of the disturbance during construction is considered to be low, especially during bridge construction which is relatively distant and is screened by plantation. Disturbance from human activity during operation would be of low magnitude because the public would not be able to access reedbeds in the WEA except along the proposed Nature Trail.
Impact Severity	Despite the relatively large area and the high ecological value of the habitat, the magnitude of disturbance impacts in this habitat would generally be low, especially during operation. Overall the impact severity is estimated to be Low during the bridge construction phase, Moderate during the residential construction phase and Low during the operational phase.

Table 4.50: Disturbance impacts on Marshes (Brachiaria Marsh, Cyclosorus Marsh and Typha Marsh) arising from the project

Criteria	Marshes (Brachiaria Marsh, Cyclosorus Marsh and Typha Marsh)
Habitat Quality	Brachiaria marsh is of Low to Moderate ecological value (except for the area immediately to the south of the development footprint, which is of Moderate ecological value), Cyclosorus marsh is of Low to Moderate ecological value; Typha marsh is of Low ecological value.
Species	Birds of conservation importance which utilise these habitats include species that are potentially sensitive to construction and operational phase disturbance due to noise and human activities/presence, but numbers of individuals are low. MPBWF is present in moderate numbers in the brackish Brachiaria marsh to the southwest of the development footprint.
Size/Abundance	During the bridge construction phase 0.3ha is present in the exclusion zone and 0.1ha in the zone of reduced density; during the residential construction phase 5.2ha is in the exclusion zone and 6.2ha in the zone of reduced density; during the operational phase 3.7ha is in the exclusion zone and 3.9ha in the zone of reduced density. The number of waterbirds, especially the most disturbance-sensitive species, within these zones is low.
Duration	Construction phase disturbance would be temporary, operational phase would be permanent.
Reversibility	Will decrease significantly once construction has been completed.
Magnitude	These habitats are located relatively close to the proposed residential area so residential construction phase disturbance would be of moderate to high magnitude. Except for the brackish Brachiaria marsh utilised by MPBWF, habitats are at some distance from the proposed bridge location and are screened by plantation, so bridge construction phase disturbance would be of low magnitude, as would disturbance due to increased human activity in the area during operation because the habitat would be within the WEA with no public access. For the same reason, no disturbance impacts on MPBWF additional to those evaluated under light disturbance impacts above.
Impact Severity	The number of disturbance-sensitive individuals present is low. and the overall impact severity is considered to be Low. No disturbance impacts on MPBWF additional to those evaluated under light disturbance impacts.

Table 4.51: Disturbance impacts on Grassy Bund and Wooded Bund arising from the project

Criteria	Grassy Bund, Wooded Bund
Habitat Quality	Bund habitats are of Low ecological value, with the exception of the wooded bund areas utilised by roosting cormorants which are of High ecological value.
Species	Utilised by some waterbird species of conservation importance, all in small numbers, except Great Cormorant, large numbers of which utilise the roost trees.
Size/Abundance	During the bridge construction phase 0.5ha is present in the exclusion zone and 0.7ha in the zone of reduced density; during the residential construction phase 4.4ha is in the exclusion zone and 5.5ha in the zone of reduced density; during the operational phase 4.2ha is in the exclusion zone and 2.9ha in the zone of reduced density. None of the wooded bund habitat containing cormorant roost trees falls within the exclusion zones, but 0.3ha would be within the zone of reduced density during the high-rise residential construction phase. The number of roosting cormorants using the trees in the reduced density zone are small relative to the overall size of the roost, numbers of other waterbirds, especially the most disturbance-sensitive species, within these zones are low. Note that because the footprint of the roost trees extends beyond the bund area over the adjacent habitats, the mapped areas at tree canopy level differ slightly from the habitat areas at ground level.
Duration	Construction phase disturbance would be temporary, operational phase would be permanent.
Reversibility	Would decrease significantly once construction has been completed.
Magnitude	These habitats are located relatively close to the proposed residential area so residential construction phases disturbance would be of moderate magnitude. Habitats are distant from the proposed bridge location and are screened by plantation, so bridge construction phase disturbance would be of low magnitude, as would, disturbance due to increased human activity in the area during operation because the habitats would be within the WEA with regulated access.
Impact Severity	Magnitude would be of Low Severity in respect of Great Cormorant as only a small area of the roost would be disturbed and any displacement would be energetically minor, and of Very Low Severity for other species as only a small number of individuals would be disturbed.

Table 4.52: Disturbance impacts on Mangrove habitat arising from the project

Criteria	Mangrove
Habitat Quality	Mangrove habitat southwest of NSW is of Moderate ecological value, largely due to the presence of MPBWF. No disturbance impacts to other mangrove areas.
Species	The small area of mangrove supports only small numbers of most disturbance-sensitive species, including a few individuals of bird species of conservation importance, however none of these species are dependent on or strongly associated with mangroves. Accordingly, disturbance impacts on species using this habitat from bridge construction and operation or other construction and operational phase elements are not considered to be significant for most species.
Size/Abundance	Just 0.3ha is located within the exclusion zone and 0.9ha in the zone of reduced density during bridge construction; 2.8ha is within the exclusion zone and none in the zone of reduced density during residential construction; and 2.2ha in the exclusion zone and 0.6ha in the zone of reduced density during operation.
Duration	Permanent, light disturbance greater once the project is operational; other disturbance greatest during the construction phase.
Reversibility	Light disturbance is readily reversible by effective screening of the light source. Other disturbance impacts will decrease significantly once the construction phase ends.
Magnitude	The magnitude will be small, due to the small size of the habitat patch. There would be a slight increase in human activity in adjacent areas during the operational phase; but the magnitude of the increase would be low.
Impact Severity	Low in general; of Moderate Severity to MPBWF, but not additional to that evaluated under impacts of light disturbance and readily addressed by screening of light sources.

Table 4.53: Disturbance impacts on Grassland and Managed Grassland arising from the project

Criteria	Grassland, Managed Grassland
Habitat Quality	Habitats are of Low ecological value.
Species	Used by a few bird species of conservation importance in small numbers.
Size/Abundance	0.3ha of grassland would be in the zone of reduced density during bridge construction, 0.1ha of grassland would be in the exclusion zone and 0.3ha of managed grassland in the zone of reduced density during residential construction, and 0.1ha of grassland would remain in the exclusion zone during the operational phase.
Duration	Permanent, but significantly lower during the operational phase.
Reversibility	Not reversible but will decrease significantly during the operational phase.
Magnitude	Magnitude will be low throughout as the areas impacted are small and very few individuals would be affected.
Impact Severity	Very Low Severity during construction, not significant during operation.

Table 4.54: Disturbance impacts on Plantation habitat arising from the project

Criteria	Plantation
Habitat Quality	Plantation which may experience disturbance is not used by roosting cormorants (no part of the cormorant roost is within 150m of the proposed development) and is of Low ecological value. Note: 'plantation’ areas utilised by roosting cormorants are classified as ''Wooded Bund" c.f. Table 4.51.
Species	Most tree species are exotic, but some are native. Shrub and field layers are floristically impoverished due to management regime. Faunal numbers and diversity are low, and most species are widespread habitat generalists.
Size/Abundance	0.6ha is located in the exclusion zone and another 1.0ha in the zone of reduced density during bridge construction; 3.4ha in the exclusion zone and 11.4ha in the zone of reduced density during residential construction; and 2.2ha in the exclusion zone and 1.9ha in the zone of reduced density during operation; most of this is located at the western side of NSW. Some of the ''plantation'' in the zone of reduced density during Phase 2 of residential construction (but none in the exclusion zone) to the east of the development footprint is used by roosting cormorants, however this is classified as 'wooded bund’ and is assessed accordingly (c.f. Table 4.51). Numbers of other faunal species are generally low and most species are not particularly sensitive to disturbance.
Duration	Permanent, but significantly lower during the operational phase.
Reversibility	Will decrease significantly during the operational phase.
Magnitude	Overall magnitude will be low during bridge construction as area is small and few species/individuals will be affected, but low to moderate during residential construction. During operation there would be increased human activity in plantation west of the development, but plantation within the WEA would not have uncontrolled public access, therefore would have a slight decrease in disturbance (because these areas are currently accessible). However, magnitude is assessed as moderate on a precautionary basis.
Impact Severity	Disturbance impact on plantation is considered to be of Low to Moderate Severity in both the residential construction and operational phases.

Impacts on Fauna of Conservation Importance

Mammals

4.8.38 Four mammal species of conservation importance were recorded at NSW: Small Indian Civet, Small Asian Mongoose, Leopard Cat and Japanese Pipistrelle. Small Indian Civet, Small Asian Mongoose and Leopard Cat have been recorded within the proposed Development Footprint. Japanese Pipistrelle was also recorded foraging along Yau Pok Road, NSW Road, and the pier to the south of NSW, within the 500m Assessment Area. No mammal species of conservation importance were recorded at LC.

4.8.39 Potential impacts on Eurasian Otter and other mammal species of conservation importance include direct habitat loss, disturbance, fragmentation of movement corridors and direct mortality arising from vehicle movements. Since, for the purposes of assessment, all wetland at NSW is regarded as suitable habitat, the direct habitat loss under the Development Site footprint (11.6ha). Large mammals are unlikely to occur close to the developed area during construction and operation; no quantitative data on the exclusion distance and distance of reduced density are available in Hong Kong. Accordingly, these distances were set on a precautionary basis as similar to those of the large, disturbance-sensitive waterbirds as 0 - 100m and 100 - 200m respectively during operation and during construction of the proposed bridge and low-rise buildings, and as 0 - 200m and 200 - 400m respectively during piling for high-rise buildings. This habitat loss would, however, be outweighed during both the construction and operational phases of the project by the removal of people and dogs at LC, thus adding over 55ha to the habitat area which would potentially be suitable for these species, as well as a reduction in disturbance in those parts of NSW outside the disturbance and exclusion zones of the Project. Fragmentation of movement corridors is highly unlikely, as the proposed development is on the edge of the area of potentially suitable habitat. For a similar reason, it is unlikely that Eurasian Otter and other mammal species of conservation importance would seek to crossroads associated with the development on a regular basis, hence direct mortality is unlikely. Furthermore, removal of uncontrolled dogs will have a potential beneficial effect.

4.8.40 Potential impacts on Eurasian Otter and other mammal species of conservation importance are evaluated below:

Table 4.55: Potential impacts on Eurasian Otter and other mammal species of conservation importance arising from the project

Criteria	Mammal species of conservation importance
Habitat Quality	The predominantly wetland habitats in the Assessment Area are not particularly favoured by any of the mammal species recorded there, all of which are generalists not tied a limited range of habitats. Although the most recent published record was in 2011, Eurasian Otter potentially still occurs irregularly within the project area.
Species	Eurasian Otter, Small Indian Civet, Small Asian Mongoose, Leopard Cat and Japanese Pipistrelle.
Size/Abundance	Most of the habitat in the Assessment Area provides potentially suitable habitat, except active fishpond areas such as LC due to the presence of large numbers of uncontrolled dogs. Loss of 11.6ha habitat within the Development Site footprint; bridge construction phase disturbance impacts leading to exclusion from 5.4ha and reduced density in 9.4ha, residential construction phase exclusion from 32.3ha and reduced density in 45.7ha; and operational phase exclusion from 20.6ha and reduced density in 23.5ha.
Duration	Direct habitat loss will be permanent; disturbance will be reduced significantly during the operational phase.
Reversibility	Disturbance will be readily reversible with appropriate mitigation measures; habitat loss will be more than compensated by the addition of more than 55ha of protected additional habitat at LC.
Magnitude	Low, as populations are small.
Impact Severity	Low during construction phase and very low for Eurasian Otter due the lack of recent records within the Project Area. Anet ecological gain for Eurasian Otter and other mammal species of conservation importance is predicted in the operational phase as the area and quality of habitat will increase, and no increase in mortality is predicted.

Based on this evaluation, Leopard Cat has been identified as a Target Species on a precautionary basis, see CMP para 1.10.5.

Birds

4.8.41 A number of bird species of conservation importance occur regularly in the Project Site and the Assessment Area. On and off-site impacts on these species will take place if they are disturbed for the following reasons:

● Displaced due to the direct habitat loss arising from the development

● Displaced due to conversion of fishpond habitat to other wetland habitats

● Displaced due to disturbance arising from bridge construction

● Displaced due to disturbance arising from Development Site construction

● Displaced due to disturbance during project operation

4.8.42 The magnitude and significance of impacts depends not only on the number of individuals affected, but also on the sensitivity of the species to disturbance. As noted in those paragraphs, the following table details the evaluation of disturbance impacts on birds and provides estimated distances for the exclusion zone and the zone of reduced density with regards to the relevant waterbird and other wetland-associated bird species of conservation importance in the area.

Species	Construction Phase (High Rise)		Construction (Low Rise) & Operation Phases (Low & High Rise)		Sensitivity to disturbance
Species	Exclusion distance (m)	Max distance of reduced density (m)	Exclusion distance (m)	Max distance of reduced density (m)	Sensitivity to disturbance
Northern Shoveler	100	300	100	200	Moderate
Eurasian Teal	100	300	50	100	Moderate-High
Tufted Duck	100	300	100	200	Moderate
Little Grebe¹	100	200	20	50	Moderate
Black-faced Spoonbill	200	400	100	200	High
Yellow Bittern	50	50	20	20	Low
Black-crowned Night Heron	50	100	20	30	Low
Chinese Pond Heron	100	300	20	30	Low - Moderate
Eastern Cattle Egret¹	50	100	20	30	Low – Moderate
Grey Heron	100	300	100	200	High
Purple Heron^1,2	100	300	100	200	High
Great Egret	200	400	100	200	High
Little Egret	100	400	20	100	Moderate – High
Great Cormorant	200	400	100	150	High
Black Kite	50	100	20	30	Low
Black-winged Stilt	100	100	50	50	Moderate
Pied Avocet¹	100	100	50	50	Moderate
Little Ringed Plover¹	100	200	20	30	Low - Moderate
Common Redshank	100	100	50	50	Moderate
Marsh Sandpiper	100	100	50	50	Moderate
Wood Sandpiper	100	100	50	50	Moderate
Common Greenshank	100	100	50	50	Moderate
Black-headed Gull¹	100	100	50	50	Moderate
White-throated Kingfisher	50	100	20	20	Low
Pied Kingfisher	50	100	20	20	Low
Collared Crow¹	50	100	100	200	Moderate
Chinese Penduline Tit¹	50	50	20	20	Low
Pallas’s Grasshopper Warbler¹	50	50	20	20	Low
Zitting Cisticola	50	50	20	20	Low
Red-billed Starling	100	200	50	100	Moderate
White-cheeked Starling	100	200	50	100	Moderate
White-shouldered Starling	100	200	50	100	Moderate

Note:

1. The table is based on the disturbance distance estimate methodology originally generated to assess construction phase disturbance impacts of the Lok Ma Chau Spur Line (Binnie, Black & Veatch 2002) and Fung Lok Wai (CH2M Hill 2008). For species not assessed in the course of those studies, disturbance distance estimates have been generated using similar methodology (primarily through empirical observation on their sensitivity to human induced disturbance in the Deep Bay area, as was the case for the Lok Ma Chau Spur Line). Such species are identified by ’1’ except where they are within the same genus and with similar ecological characteristics to species assessed under those studies (Yellow Bittern and Starling spp.). This approach has not been followed for ducks in recognition that Eurasian Teal, which was assessed under Binnie, Black & Veatch (2002), is less sensitive to disturbance than several larger duck species. Distances generated by Binnie, Black & Veatch (2002) assumed the implementation of screen planting, but no other in situ mitigation; application of these distances to impacts from the current development project is broadly comparable in this respect, subject to the assumption that hoardings are in place around any works area.

2. Purple Herons were not found to be of regular occurrence during surveys but are included here as they favour large reedbeds.

3. All of these species were recorded during the 2021-2022 survey.

4.8.43 The table below summarises the numbers of birds of conservation importance predicted to be impacted, because of the factors detailed above. For details, please refer to Table 1 of Appendix 4.2c.

Species	Maximum Mean Number Displaced
Northern Shoveler	6.67
Eurasian Teal	7.65
Tufted Duck	1.00
Little Grebe	1.92
Eurasian Spoonbill	0.04
Black-faced Spoonbill	0.38
Yellow Bittern	0.22
Black-crowned Night Heron	0.44
Striated Heron	0.04
Chinese Pond Heron	8.02
Eastern Cattle Egret	0.67
Grey Heron	5.72
Purple Heron	0.76
Great Egret	5.87
Little Egret	9.50
Great Cormorant	13.69
Western Osprey	0.03
Black-winged Kite	0.19
Crested Serpent Eagle	0.04
Greater Spotted Eagle	0.04
Besra	0.02
Pied Harrier	0.03
Black Kite	0.98
Eastern Buzzard	0.08
Eurasian Coot	0.04
Black-winged Stilt	45.05
Pied Avocet	15.63
Pacific Golden Plover	0.13
Grey Plover	0.04
Little Ringed Plover	1.50
Whimbrel	0.08
Terek Sandpiper	0.08
Common Redshank	6.88
Marsh Sandpiper	0.17
Wood Sandpiper	0.17
Spotted Redshank	0.08
Common Greenshank	3.96
Black-headed Gull	7.71
Collared Scops Owl	0.21
Asian Barred Owlet	0.12
White-throated Kingfisher	0.94
Pied Kingfisher	0.55
Common Kestrel	0.04
Ashy Drongo	0.08
Collared Crow	1.24
Chinese Penduline Tit	4.75
Thick-billed Warbler	0.04
Pallas's Grasshopper Warbler	0.39
Lanceolated Warbler	0.12
Zitting Cisticola	0.09
Red-billed Starling	4.86
White-cheeked Starling	0.54
White-shouldered Starling	0.97
Common Starling	0.21
Red-throated Pipit	0.08
Chestnut-eared Bunting	0.08

4.8.44 The significance of the impacts of displacement on bird species of conservation importance is evaluated below.

Table 4.58: Potential impacts on bird species of conservation importance arising from the project

Criteria	Birds of Conservation Importance
Habitat Quality	Wetland habitat in the Project Site is largely of high value for waterbirds of conservation importance; other habitats are of low or low to moderate value for birds, except for the eucalypts at NSW which are of high value for Great Cormorants as a roost.
Species	A total of 56 bird species of conservation importance are potentially impacted by the Project (see Table 4.57). These include one globally Endangered species (Black-faced Spoonbill) and two globally Vulnerable species (Greater Spotted Eagle and Collared Crow).
Size/Abundance	Numbers of most species that may be affected are low (fewer than 20 individuals for 55 species; however, numbers of one species, Black-winged Stilt (45), are low to moderate relative to the Deep Bay populations.
Duration	Impacts of direct habitat loss would be permanent; disturbance impacts would be permanent but would be significantly reduced during the operational phase.
Reversibility	Readily reversible by habitat enhancement to increase carrying capacity and function of habitats within the Project Area.
Magnitude	Magnitude of impacts is expected to be low for most species but is potentially low to moderate for those occurring in larger numbers; moderate overall due to the large number of species which may be affected.
Impact Severity	Low severity for most species but potentially of Low to Moderate severity for Black-winged Stilt. However, Moderate severity overall due to the large number of species which may be affected.

Potential Impacts on Roosting Great Cormorants at NSW

4.8.45 The development location has been adjusted to take into account the location of the cormorant roost based on available data from 2011-2015 (AEC 2016) and 2021-2022, such that no part of the development is within the 150m distance (a distance known from observations MPNR and Fairview Park to have no effect on cormorant roosts) It is thus considered that the roosting cormorants would not be disturbed during the operational phase. However, Great Cormorants may be more sensitive to disturbance during the residential construction phase, especially during construction of the high-rise buildings, when an exclusion zone is estimated at 0 - 200m and the zone of reduced density at 200 - 400m.

4.8.46 Nonetheless, it should be highlighted here that in the Deep Bay area (including NSW), all cormorant roosts may not be in the exact same trees in each winter, with the roost locations varying to some extent during each winter. As shown in Appendix 4.2a, the number at each roost varies markedly from year-to-year, and within each winter. In particular, the percentage of the entire Deep Bay population utilising roosts in NSW ranged from 25% to 100% (based on past data between 2011 and 2020). This suggests the cormorants' fidelity to individual roosting areas is low and they readily switch between roosting sites even without obvious human disturbance.

4.8.47 To minimise the potential disturbance during the construction phase, the disposition of the residential towers has been adjusted so that no part of the roost based on available data from 2011-2015 (AEC 2016) and 2021-2022 is located within the exclusion zone and most of the roost is located outside the zone of reduced density (see Figures 4.16a-d). According to the recent surveys in 2021-22, the maximum number of Great Cormorants that would potentially be displaced by the construction works is 984 birds, approximately 12% of the total Deep Bay population (see Table 4.59).

4.8.48 In the worst-case scenario that all these Great Cormorants are displaced, there will be more than ample capacity in all the other roosts, but specifically, in the northern roost at NSW (based on the peak count of 4044 birds in the northern roost in December 2011 and the recent lower count of less than 1000 bird in 2021-22) for the displaced cormorants. It is thus reasonable to conclude that the potential impact is of low to moderate severity if the consequence of any disturbance is that birds alter their individual locations within the roost area, and between roosts, where there is ample carrying capacity.

Table 4.59: Estimation of the percentage of the cormorant roost potentially disturbed during the construction phase of the project

Date	NSW Total	Deep Bay Total	% at NSW of Deep Bay Total	Within the Reduced Density Zone during the Construction of High-rise Buildings
Date	NSW Total	Deep Bay Total	% at NSW of Deep Bay Total	Total	% of NSW total	% of Deep Bay Total
29 Nov 2021	4851	6041	80%	556	11%	9%
16 Dec 2021	7405	9088	81%	795	11%	9%
19 Jan 2022	5795	8123	71%	984	17%	12%
24 Feb 2022	4748	6500	73%	594	13%	9%

4.8.49 Potential impacts on the cormorant roost are evaluated below.

Table 4.60: Potential impacts on the Great Cormorant roost arising from the project

Criteria	Great Cormorant Roost
Habitat Quality	Cormorant roost tree habitat is not of intrinsically high ecological value but is evaluated as such due to the presence of the roost.
Species	Great Cormorant.
Size/Abundance	Roost is large (up to 7405 individuals). The southern part of the roost (i.e. the part closest to the Development Site) has held up to 6917 individuals; but typically, 500 – 1000 individuals roost in the potential disturbance zone (maximum 984 individuals). Number of birds using flightlines that may be interrupted is very small.
Duration	Most likely wholly or largely temporary as the roost trees will be available to the cormorants throughout the construction and operational phases of the project and it is predicted that they will become habituated to the development. However, the roost trees closest to the high-rise buildings may be permanently abandoned.
Reversibility	Largely or wholly reversible.
Magnitude	On a precautionary basis, magnitude is predicted to be moderate during the construction phase and low to moderate during the operational phase as the trees used by most roosting birds are not within the predicted disturbance zones during either the construction or, especially, the operational phases of the project. However, the response of cormorants to the proximity of high buildings is uncertain, regarding both roost sites and flightlines; thus it is possible that the anticipated reduction in disturbance during the operational phase will be less than is predicted. Under this scenario the magnitude of impact would continue to be moderate during the operational phase. Impacts due to the nature trail in the WEA would be of very low magnitude.
Impact Severity	Whilst the magnitude of the impact is predicted to be moderate during the construction phase and, on a precautionary basis, moderate during the operational phase, the Severity of the impact is considered to be Low to Moderate as the parts of the roost (in particular the northern roost) which will not be impacted have the capacity to accommodate any displaced cormorants under all but the most exceptional circumstances.

Potential Impacts on Bird Flightlines

4.8.50 A flightline can be defined as a route which is regularly followed for local movements of one or more individuals over a period (unlike a flight path, which is the route of a single individual on a single occasion, or a flyway, which is a broad route followed by a large number of migrating individuals over a long distance). A common example relevant to NSW is the flightline between an egretry and the foraging sites used by breeding egrets; birds will follow such a flightline regularly over the course of the day to collect food and then return to the nest to incubate or to feed young. Flightlines also regularly occur between roost sites, especially those of species which roost communally, and foraging areas. This example is also of relevance to NSW as the Project Site includes one of the four roost sites for Great Cormorants in the Inner Deep Bay area of Hong Kong. Starting and finishing points of flightlines may be tightly defined (such as a roost site or a breeding site) or cover a wider area (for example a foraging area covering a wide area of habitat). Geographical features may influence the location of flightlines; for example, waterbirds tend to follow water features such as rivers and may follow valleys even when a route crossing a line of hills would be shorter. Again, this example is relevant to the Project, as a waterbird flightline that follows the SPR has been identified.

4.8.51 Development located on a flightline may result in a decrease in the suitability of a foraging, breeding or roosting site by preventing movement to another site or by reducing the efficiency of movement between sites. In an extreme case, this may lead to the total abandonment of one or both sites. There may also be an increased risk of mortality by collision with structures constructed on or close to a flightline. Studies in Hong Kong have shown that houses of up to three storeys have little effect on flightlines. Conversely, for most species, buildings of more than five or six storeys form a barrier. As a corollary, there is likely to be little or no difference in the barrier effect caused by the lowest (19 storey) and highest (25 storey) apartment blocks proposed in the current development.

4.8.52 The following evaluation addresses potential impacts on the following flightlines:

- Flightlines between the Great Cormorant roost and Deep Bay

- Waterbird flightlines along the SPR

- Egretry flightlines

As is shown In Figure 4.7, surveys demonstrated that all the Great Cormorants follow flightlines directed northwards from the roost, towards Deep Bay and other wetland areas. The proposed development footprint, positioned in the southwest of the Project Site, avoids these Great Cormorant flightlines. As was discussed in para 4.6.31, numbers of waterbirds flying over the Development Site were insignificant in the context and no defined flightlines were present. Throughout the surveys in 2021/2022, fewer than one individual per survey hour was recorded flying across the proposed development footprint (see Table 4.12).

As was discussed in para 4.6.32 and shown in Figure 4.8, a flightline used by a small number of waterbirds (4.7 birds per hour in total) follows the SPR and hence is potentially impacted by the proposed bridge connecting the Development Site to Yuen Long. Of this small number of individuals, most would not have to adjust their flight path or altitude, whilst for the others the adjustment required would be minor.

4.8.53 Flightlines surveys at the SPR egretry (Figure 4.9) showed that 33% of individuals were flying toward LC, 32% were flying along KTMDC, 23% were flying towards FLW in two directions and only 12% were flying southward along the SPR. These flightlines will not be affected by the proposed development, hence are not considered further here.

Table 4.61: Flightline impacts arising from the project

Criteria	Flightlines
Habitat Quality	Not relevant (impact is to existing air-space).
Species	Waterbirds, in particular Great Cormorants flying between their roost at NSW and foraging areas and small numbers of a variety of waterbird species flying north-south along the SPR.
Size/Abundance	The Great Cormorant flightlines are used by several thousand individuals, often comprising the majority of the Hong Kong population; however, almost all cormorants depart northwards towards Deep Bay rather than southwest over the Development Site footprint, so flightlines of only a very small number of individuals from the majority of the roost would be affected and any deviation required to be made by these birds would not be significant. Deviations to flightlines of waterbirds flying along the SPR caused by the proposed bridge would be small as would the number of individuals affected.
Duration	Largely temporary (during the high-rise residential construction phase) but perhaps some permanent avoidance in respect to cormorant flightlines; permanent, but greatest during the construction phase, in respect to flightlines along the river.
Reversibility	Will largely be reversed during the operational phase, some permanent minor adjustments to flightlines of some birds along the SPR, and perhaps some minor adjustments to the cormorant roost locations.
Magnitude	Magnitude will be Low.
Impact Severity	Low severity as the number of individuals affected would be small and, in the case of Great Cormorants, any birds displaced are likely to respond by using an alternative roost site still within NSW

Herpetofauna

4.8.54 During surveys in 2021/2022, four herpetofauna species of conservation importance namely Chinese Stripe-necked Turtle, Reeves’ Turtle, Chinese Soft-shelled Turtle and Common Rat Snake were recorded in NSW and one herpetofauna species of conservation importance, Many-banded Krait was recorded in LC. The Chinese Stripe-necked Turtle observations are considered to be of ex-captive (released) individuals. Reeves’ Turtle is globally Endangered and is wetland-dependent. Chinese Soft-shelled Turtle is globally Vulnerable and is wetland-dependent. Common Rat Snake is of Potential Regional Concern (but is very common and widespread in Hong Kong) and is not wetland-dependent. Many-banded Krait is Potential Regional Concern but it is common and widespread in Hong Kong. Only a single individual for all the species were recorded once, however all three records in NSW were made within the development footprint. Nonetheless, there is no evidence of them being dependent to a particular area and/or habitat within the development footprint. Suitable habitats for these species exist elsewhere in NSW and LC.

4.8.55 Another species, Burmese Python, was recorded in NSW in previous surveys. It is globally Vulnerable but is widespread in a variety of habitats in Hong Kong and was recorded twice in the Project Site previously. Burmese Python has not been recorded within the Development Site footprint, although suitable habitat exists there and the species may be present within the footprint.

Table 4.62: Potential impacts on herpetofauna of conservation importance arising from the project

Criteria	Herpetofauna of Conservation Importance
Habitat Quality	Four of the species were recorded in the Development Site footprint. For Burmese Python, although it was not recorded in the footprint, suitable habitat exists within the footprint and could be impacted by the development.
Species	Reeves’ Turtle, Chinese Soft-shelled Turtle, Burmese Python, Common Rat Snake, Many-banded Krait.
Size/Abundance	All these species were recorded in low numbers and infrequently. A small population of each of these species is expected to be present in the site.
Duration	Impacts of habitat loss would be permanent, but readily compensated.
Reversibility	Readily reversible by compensatory habitat provision, habitat enhancement and improved protection.
Magnitude	Magnitude of impacts is expected to be low.
Impact Severity	Impacts of Low Severity.

Odonates

4.8.56 The only odonate species of conservation importance recorded within the Development Footprint during surveys in 2021/2022 was Ruby Darter. Mangrove Skimmer was recorded in NSW but outside the development footprint. Dingy Dusk-hawker and Scarlet Basker were recorded in the Assessment Area outside the Project Site. Dingy Dusk-hawker, Ruby Darter and Scarlet Basker are included in Fellowes et al. (2002) as being of Local Concern, but all have subsequently increased in Hong Kong and are now Common (Tam et al. 2011). Mangrove Skimmer is included in Fellowes et al. (2002) as being of Global Concern. It was also listed as Vulnerable in the IUCN Red List but the assessment was done in 2009. It is Uncommon in Hong Kong. Dingy Dusk-hawker, Mangrove Skimmer and Scarlet Basker were not recorded in the Development Footprint, while Ruby Darter was recorded in areas that would be directly impacted.

Table 4.63: Potential impacts on odonate species of conservation importance arising from the project

Criteria	Odonates of Conservation Importance
Habitat Quality	Species recorded in Typha Marsh, reedbed and fishpond.
Species	Scarlet Basker, Dingy Dusk-hawker, Ruby Darter and Coastal Glider are Common in Hong Kong. Mangrove Skimmer is Uncommon in Hong Kong.
Size/Abundance	Small numbers of each species recorded.
Duration	Permanent, but readily compensated by habitat provision and enhancement.
Reversibility	Readily reversed by compensatory habitat provision and enhancement.
Magnitude	Impacts would be of Low Magnitude because both species are present in habitats and areas that would not be impacted.
Impact Severity	Low Severity because the species are now common in Hong Kong and some areas used by both species would not be directly impacted and will be enhanced.

Butterflies

4.8.57 During surveys in 2021/22, three butterfly species recorded in NSW are listed by Fellowes et al. (2002) as being of Local Concern (Common Awl, Danaid Egg-fly and Pale Palm Dart) and Small Cabbage White, recorded in LC is considered locally Rare. In the previous surveys, Painted Lady and Grass Demon were also recorded. The former is of Local Concern, while the latter is listed by Chan et al. (2011) as being Rare in Hong Kong. The larval host plants for Danaid Egg-fly, Painted Lady and Grass Demon were not recorded on site during floral surveys, and both species are likely to be wandering individuals from outside the site. However, Pale Palm Dart may breed on site where the host plants (large grasses including Miscanthus sinensis and Pennisetum purpureum) are present. These species are not prone to disturbance, and the only impact would be direct habitat loss.

Table 4.64: Potential impacts on butterfly species of conservation importance arising from the project

Criteria	Butterfly Species of Conservation Importance
Habitat Quality	Suitable habitat for breeding present for Pale Palm Dart, but probably not for the other species.
Species	Pale Palm Dart, Grass Demon, Painted Lady and Small Cabbage White (which is widespread globally as a crop pest) are all listed as Rare in Hong Kong. Danaid Egg-fly is Uncommon. Common Awl is Very Rare.
Size/Abundance	Numbers recorded on surveys are small, involving only one or two individuals.
Duration	Permanent.
Reversibility	Readily reversible by habitat provision and enhancement.
Magnitude	A small number of individuals impacted by habitat loss so magnitude of impacts would be low.
Impact Severity	Low Severity for Pale Palm Dart because a small number of individuals may be impacted by habitat loss but habitat would be retained in WEA. Negligible impact for the other species because none would be impacted by loss of breeding habitat.

MPBWF Pteroptyx maipo

4.8.58 Comparison of the results from the 2019 surveys (see AEC 2020) with those undertaken in 2015 (see AEC 2016) suggests that the population of MPBWF close to the proposed development at NSW has declined, with no recent records of the species in the central part of the previous area of occupancy. Much of the habitat in this area remains suitable for MPBWF, however very bright lights have been installed at the Yuen Long Temporary Driving School on the west side of the former alignment of Kam Tin River and additional lighting has been installed near the Brachiaria marsh - this new lighting may have had an adverse impact on the MPBWF population. The number of MPBWF recorded in this area during 2019 (144 individuals) was about 8% of the total number recorded around Deep Bay (1697 individuals). Most of those recorded were seen in the Brachiaria marsh around the village houses near the jetty, with a smaller population also recorded at a small channel at the northwest. During the 2021 MPBWF surveys, there was a similar distribution to the 2019 surveys (AEC 2020); see Figure 4.12c. Concentrations of MPBWF were recorded in mangrove and Brachiaria marsh to the south of NSW and mangrove around the southern end of MPNR while moderate numbers were recorded in the channel to the south of Lut Chau. However, the number recorded at the channel between Lut Chau and Tai Sang Wai was found to declined further; this may be due to the growth of exotic mangrove stands along the tidal watercourse. The study of habitat preference (AEC (2020), Appendix 4.4) suggested that extensive pure mangrove stands may not be favoured by this species, albeit further study is required. None was recorded within the Development Site.

4.8.59 Potential impacts to this remaining population could arise through direct habitat loss or from indirect impacts due to night-time light pollution arising from the development. Nonetheless, no MPBWF were found within the proposed Development Site, and the project will have no direct impacts on habitats utilised by the species. Furthermore, the development will not alter the hydrology or vegetation within the areas used by MPBWF. Accordingly, no direct impacts on the species due to loss of habitat are predicted.

4.8.60 Both male and female MPBWF use bioluminescent light signals for courtship. Their larvae also use bioluminescence, possibly to deter potential predators. An increase in night-time light levels would potentially limit the effectiveness of these signals, leading to an adjustment of the nature of the signal or to abandon signalling entirely. As no night-time construction work is required, this impact will be largely restricted to the operation phase of the Project.

4.8.61 Nonetheless, there is evidence that MPBWF can tolerate low levels of artificial lighting. For instance, the species has been known to occur in areas close to artificial light sources including near Fairview Park, and along the boundary fences at Tsim Bei Tsui and Mai Po (Yiu 2011). Based on the previous surveys (see AEC 2016), there appears to be no significant effect from light levels on the species up to 0.49 lux, as measurements of ambient light at locations where the species was present in May 2015 recorded flashing at sites with light levels up to 0.49 lux. Conversely, light levels ranged from 0.01 to 0.27 lux where no MPBWF activities were recorded, and ambient light levels at sites where the species was present showed little or no difference from sites where the species was absent (AEC 2016). During the 2019 surveys, adult MPBWF were recorded regularly alongside illuminated roads at KTMDC and NSW, where they were observed at light levels as high as 11.95 lux (see AEC 2020). However, densities of flashing individuals at these sites were lower than at unilluminated sites, and most observations were in darker sites (average 0.41 lux, with 91% below 1 lux), while the average at quadrat locations was 0.11 lux (AEC 2020).

4.8.62 When considering the impacts of night-time lighting, the direction of the light source, the presence of any shielding and the behaviour/location of the species in question should also be considered other than light intensity (see relevant discussions in AEC 2020). MPBWF males typically emit light in courtship flights up to 1.5m above vegetation and females flash while perched on vegetation (Yiu 2012), therefore the impact of lighting would be greatest if unimpeded lights were shining directly onto suitable vegetation or a short distance above this. Lights directed away from this habitat (upwards or in an opposite direction) or screened at a suitable height would have less impact on MPBWF activity. It should be highlighted here that, light intensity drops rapidly away from a source of light, especially if the light is not directed towards the observation point (AEC 2020). Streetlights were separated by 30-35m, and between two successive lights (about 15m from the nearest light source) the measured light intensity fell to 0.05 ± 0.22 lux at KTR, 0.23 ± 1.26 lux at NSW, 0.40 ± 0.68 lux at FLW and 1.49 ± 1.75 lux at Mai Po Nature Reserve. These light levels are dark enough for MPBWF activity and suggest that the light levels were low enough that MPBWF activity would not be affected at 15-20m from the light, provided that the light was not directed towards the habitat.

4.8.63 The wavelength of light may also be important for fireflies (see discussions in AEC 2020, TPAB 2021). Most light level readings (including the ones taken in this study) are measured across wavelengths visible to the human eye, and do not consider variable light intensity at different wavelengths. Streetlights on the roads at KTR and NSW are high-pressure sodium vapour lamps, which emit light at relative long frequency (589nm), appearing orange to the human eye. The lights on the boundary fence roads are not only brighter but also appear to emit in a broader spectrum, appearing white to the human eye. The short wavelength lights along the boundary fence road may have a greater impact on MPBWF flashing activity than the longer wavelength lights at KTR and NSW.

4.8.64 Some insects are attracted towards lights at night, and it may be expected MPBWF would be most attracted towards green lights of a similar wavelength and similar duration to the display flashes phototaxis (see discussions in AEC 2020, TPAB 2021). Street lighting is typically of a different wavelength and is continuous, and indeed no phototaxis of MPBWF towards streetlights has been observed during surveys. Other species of Pteroptyx fireflies that form congregations of flashing males have been found to be attracted towards strings of flashing lights in trees, but this congregational behaviour does not occur in MPBWF (AEC 2020, TPAB 2021).

4.8.65 In addition to disturbance from light, fireflies may also be sensitive to potential fragmentation impacts. These potential impacts are considered in the following section.

Table 4.65a: Potential construction phase impacts on MPBWF arising from the project

Criteria	MPBWF
Habitat Quality	MPBWF occurs in mangrove and marshes along SPR and KTMDC. Habitats would not be directly impacted.
Species	MPBWF was described in Hong Kong in 2011 and was later also found in western Guangdong and Hainan. It is listed as Endangered in IUCN (Yip & Yiu 2023).
Size/Abundance	A relatively high density of the species is present in Brachiaria marsh at the southwest of the Project Site, but none are recorded within the Development Site.
Duration	Permanent in the absence of mitigation measures.
Reversibility	Readily reversible by removing and/or screening existing and any additional light sources.
Magnitude	No direct habitat loss. Impacts from light disturbance low as no night time construction work is required. Impacts of fragmentation of low magnitude.
Impact Severity	No impact from habitat loss and impact of low severity due to fragmentation and of Low Severity as a result of light disturbance in the absence of mitigation.

Table 4.65b: Potential operational phase impacts on MPBWF arising from the project

Criteria	MPBWF
Habitat Quality	MPBWF occurs in mangrove and marshes along SPR and KTMDC. Habitats would not be directly impacted.
Species	MPBWF was described in Hong Kong in 2011 and was later also found in western Guangdong and Hainan. It is listed as Endangered in IUCN (Yip & Yiu 2023).
Size/Abundance	A relatively high density of the species is present in Brachiaria marsh at the southwest of the Project Site, but none are recorded within the Development Site.
Duration	Permanent in the absence of mitigation measures.
Reversibility	Readily reversible by removing and/or screening existing and any additional light sources.
Magnitude	No direct habitat loss. Impacts from light disturbance potentially of moderate to high magnitude. Impacts of fragmentation of low magnitude.
Impact Severity	No impact from habitat loss and impact of low severity due to fragmentation, but overall impacts to this species potentially of Moderate to High Severity as a result of light disturbance in the absence of mitigation.

Fragmentation Impacts

4.8.66 Fragmentation impacts arise where development or other human activities impede or sever ecological linkages between or within habitats and areas. Construction of a development between habitats which show ecological linkage may result in the loss of these links and thus a decrease in the suitability of the habitat for particular species and a reduction in the overall value of the habitat. The isolation of two patches of habitat can prevent the movement of organisms between these habitat patches, resulting in an effective reduction in population size and genetic isolation of the population.

4.8.67 Construction within a large, continuous habitat patch may result in an edge effect reducing the overall value of the habitat. Generally, larger areas of habitat are of higher importance than smaller areas; this is not simply an arithmetical relationship (for example doubling the area of a particular habitat may result in more than double the ecological value). Many species require a minimum area of a habitat and would not utilise two smaller fragments amounting to the same area. The severity of fragmentation impacts will depend upon the extent to which severance occurs (whether this is partial or total), the relative sizes of resulting fragments, the extent and complexity of previous linkages and the baseline species diversity.

4.8.68 At its most extreme, fragmentation impacts may result in the loss of populations of a species if the remaining fragments are too small to support a viable population. Species most affected by fragmentation impacts are habitat specialists, terrestrial species and species with low mobility. Birds and flying insects are generally less affected than mammals, herpetofauna and non-flying invertebrates.

4.8.69 The requirement in the NSW OZP for the residential element of the current project to be situated in the part of the OU(CDWEA1) zone furthest from Deep Bay is (the Explanatory Statement, at para 9.7.2, states more explicitly at ‘the southern most portion of the zone’) in part, a recognition of the need to ensure that fragmentation impacts of the project are minimised. By following this requirement, the Development Site footprint of the project is such that there will be no fragmentation impacts on MPNR and intertidal mudflat habitat, the nearest parts of which will be 2km from the Development Site Boundary, and no significant fragmentation impacts on fishpond and reedbed habitats (indeed, internal fragmentation of these habitats will be reduced during the operational phase). However, the requirement to locate the development in the part of the OU(CDWEA1) zone furthest from Deep Bay, together with the fundamental geography of the river channels, has the consequence that avoidance of all potential fragmentation impacts on the mangrove and Brachiaria marsh and its population of MPBWF to the southwest of NSW and the SPR is not possible. Accordingly, potential fragmentation impacts on these habitats is evaluated in the table below.

Table 4.66: Fragmentation impacts on Mangrove and Brachiaria Marsh habitat located to the southwest of NSW arising from the project

Criteria	Mangrove and Brachiaria Marsh Southwest of NSW
Habitat Quality	Both the mangrove and Brachiaria marsh habitats to the southwest of NSW are assessed as being of Moderate ecological value, largely due to the presence of Pteroptyx maipo.
Species	Movement and dispersal of most fauna species, including mangroves and invertebrates is likely largely by way of the river channels. Birds may, however, move east/west between the mangroves and wetland habitats in NSW and beyond. Dispersal by MPBWF is (presumably) undertaken by flying adults and, in view of its restrictedness to mangroves and nearby brackish water wetland areas, it is perhaps more likely to move along the former alignment of Kam Tin River channel (where there are some mangroves) than over the NSW site. This supposition is supported by the presence of MPBWF in Brachiaria Marsh along the river but not within NSW (Figure 4.12b).
Size/Abundance	The area which might be fragmented from other mangrove habitat is small; there will be no fragmentation of Brachiaria marsh habitat as the areas of this habitat along the KTMDC will not be affected by the project. Numbers of MPBWF are relatively large. Numbers of other species are small.
Duration	Permanent.
Reversibility	Not reversible.
Magnitude	Magnitude of fragmentation impacts for mangrove flora and invertebrate fauna is very low as river channel connection is not affected; magnitude of impacts on wetland birds is low to moderate; magnitude of impacts on MPBWF is low, as the KTMDC connection is considered to be of greater importance in maintaining linkages than NSW where this species is absent.
Impact Severity	Low Severity as impacts on linkages is low (most fauna including MPBWF) or number of individuals is small (wetland birds).

4.8.70 Access is provided by means of a proposed bridge spanning SPR. While this avoids disturbance and fragmentation impacts along the existing NSW Road (alongside KTMDC), it does provide the potential for fragmentation within SPR. The potential fragmentation impact of this proposed bridge is included in the evaluation of impacts to watercourses in the table below.

Table 4.67: Fragmentation impacts on Tidal Watercourse habitat arising from the project (including potential fragmentation caused by the proposed bridge across Shan Pui River

Criteria	Tidal Watercourse
Habitat Quality	The northern section of the SPR is of Very High ecological value due to its importance for large waterbirds, but the southern section, is of Low to Moderate ecological value as it is used by relatively few waterbirds, perhaps because it is somewhat disturbed because of its proximity to Yuen Long. The former alignment of Kam Tin River is of Moderate ecological value but is also somewhat disturbed by adjacent village housing.
Species	Neither the SPR nor the KTMDC will be physically fragmented, and no construction or other works will take place within the river channels; however the proposed bridge over the SPR will potentially form a barrier to waterbird movement; linkages between the watercourses and other wetland habitats will not be broken, but the movement corridor for birds moving between the SPR and wetlands to the north and east will be narrowed.
Size/Abundance	The SPR supports moderate numbers of waterbirds. However, few of these are likely to experience habitat fragmentation due to the bridge construction as numbers using the river to the south of the proposed bridge location are very small.
Duration	Permanent: greatest during the construction phase.
Reversibility	Not reversible but will decrease somewhat once the construction phase ends.
Magnitude	Magnitude will be small as the total area impacted is relatively small, it is used by relatively small numbers of species, of which only wetland birds will experience some fragmentation impacts due to the proposed bridge formation and the narrowing of movement corridors.
Impact Severity	Low Severity.

Pollution Impacts

4.8.71 Pollution of air or water may arise from a large number of different sources and could occur during construction, operation or both. The severity of a pollution event would depend upon the type of pollutant being released, the nature and ecological value of the habitat affected, the size of the habitat affected, the amount of pollutant released, the duration of the event and the sensitivity of species potentially affected. Given the number of factors involved, the potential impacts of pollution are difficult to evaluate with a high degree of accuracy.

4.8.72 Impact evaluation in this report therefore considers a worst-case scenario by assessing the habitats which are of highest ecological importance and assessing the potential impacts arising to the habitats. Impacts of water pollution are given particular importance due to the sensitivity of the species involved and the high ecological value of wetland habitats in Hong Kong. In contrast, air pollution generally disperses over a larger area and the impacts are evident regionally but mostly do not affect local ecology. For the purposes of assessment, water pollution includes the potential impacts of sedimentation or dust deposition. There will be no discharge of sewage or effluent into Deep Bay wetlands during operation of the project.

Table 4.68: Potential pollution impacts on wetland habitats arising from the project

Criteria	Wetland Habitats (Mai Po Nature Reserve, fishpond, open water, reedbed, marshes, intertidal mudflat, mangrove, tidal watercourse)
Habitat Quality	All wetland habitats in the Assessment Area are of at least Low to Moderate ecological value; LC mangroves are of High ecological value; MPNR, intertidal mudflats and the KTMDC and the SPR are of Very High ecological value, due to their supporting internationally important populations of waterbirds.
Species	Many habitat-restricted species of conservation importance, in particular birds, but also several invertebrates.
Size/Abundance	Area is large (potentially the Deep Bay area in a worst-case scenario); faunal community is diverse and abundant.
Duration	Temporary; though some types of chemical pollution can have very long-lasting consequences.
Reversibility	Reversible, though restoration of some habitats, notably mangroves, could take many years.
Magnitude	Magnitude of the more likely pollution events (drift of dust, discharge of sediments, spillage of small quantities of fuel) is small.
Impact Severity	Potentially of Low to Moderate Severity during the construction phase (especially during construction of the proposed bridge over the SPR), but Low Severity during the operational phase.

Table 4.69: Potential pollution impacts on non-wetland habitats arising from the project

Criteria	Non-wetland Habitats (grassland, plantation, developed area)
Habitat Quality	Most non-wetland habitat in the Assessment Area is developed area and grassland of low ecological value; some plantation in the Project Site is of high ecological value due to its use as a cormorant roost.
Species	Most fauna and flora species are common, and not-habitat restricted; the Great Cormorant roost is of significance as it is one of the four main roosts in Inner Deep Bay.
Size/Abundance	Area is large in the Assessment Area (c. 150ha), but 70% of this is developed area. The area of trees forming the cormorant roost is small (3.0ha).
Duration	Temporary; though some types of chemical pollution can have very long-lasting consequences.
Reversibility	Reversible, though restoration of some habitats, could take many years.
Magnitude	Magnitude of the more likely pollution events (drift of dust, discharge of sediments, and spillage of small quantities of fuel) is likely to be small.
Impact Severity	Low Severity during the construction phase, as most of the habitats are not sensitive and the more likely events are both small in scale and reversible; the cormorant roost trees are at least 150m from the developed area boundary and are thus unlikely to be affected by any discharges. Very low during the operational phase, as the most likely events, such as fuel spillage, would primarily affect developed areas.

Hydrological Disruption

4.8.73 Hydrological disruption may have significant impacts to wetland habitats, either by increasing or decreasing water inputs or changing water quality (salinity, temperature, pH and suspended solids). In some circumstances, decrease in water input may result in a loss of a wetland habitat as it dries out. In the current Assessment Area some wetland habitats (open water and reedbed) are largely rain-fed, though some fishponds at LC and at the southwest of NSW have connections to the channels and receive water from this source at higher tides; the river channels are tidal and hence receive the water either as discharge from upstream or from Deep Bay; whilst intertidal mudflats and mangroves largely receive their water from the latter source.

4.8.74 There are two potential causes of hydrological disruption arising from the project: the construction of the proposed bridge over the SPR, and the potential disruption to water circulation at NSW by the development. The proposed bridge design does not require the construction of any piers or footings within the channel, hence no impacts from this source are predicted from permanent habitat loss or from temporary occupation during construction. One connection at NSW to the river will be lost as it will be within the footprint of the Development Site; however the waterbodies which are hydrologically connected will also be within the development footprint; there will be no consequences for other waterbodies at NSW which are not hydrologically connected. The other connection (from the KTMDC) will not be affected by the development and the waterbodies which are hydrologically connected will continue to be tidally influenced.

4.8.75 Accordingly, therefore, there are no predicted hydrological impacts of significance arising from the project.

Direct Mortality

4.8.76 Direct mortality involves the death of organisms because of the development, including individuals killed during the construction process or individuals killed from the structures constructed. Animals may be hit and killed or injured by rapidly-moving vehicles or by collision with stationary objects. Birds and mammals appear to be most susceptible (Van der Grift and Kuijsters 1998), though herpetofauna are also prone. Should roads pass through areas of high animal population density or cut across regular lines of movement, such mortality is likely to be greater. The risk of animal mortality arising from roadkill and collision with buildings is likely to be greater in rural areas than in already developed, urban areas, as wildlife populations are higher in the former.

4.8.77 Numerous studies have documented avian mortality associated with buildings, usually the result of collision with tall buildings with extensive glass in their façades or with windows (e.g. Ogden 1996). Inclement weather during the migration seasons is known to exacerbate nocturnal collision mortality, as birds become disorientated in such weather conditions. The combination of strong night-time lighting emitted from a tall building or structure is a particular source of mortality.

4.8.78 Collisions also occur in daytime, though the causes are generally related to the nature of the building exterior (glass being the prime culprit), and the key risk factors are transparency and reflectivity. Building façades that constitute transparent glass may appear not to present an obstacle to flight and birds may strike windows as they attempt to access potential perches, plants, food and water sources or other lures seen through the glass. Design features such as glass skywalks joining buildings, glass walls around planted atria and windows installed perpendicularly at building corners are dangerous, as birds perceive these as an unobstructed route.

4.8.79 The current project will include tall buildings (up to 25 storeys) and will be relatively close to the Deep Bay area where large numbers of birds are present. However, unlike many commercial buildings, the residential towers will not contain large expanses of glass or reflective surfaces. Similarly, unlike some commercial buildings, residential towers are not normally ‘lit up’ externally to render them conspicuous, as such lighting would be unacceptable to residents. Again, such external lighting is not proposed in the current project.

4.8.80 Terrestrial fauna numbers in the Assessment Area are generally rather small and, as the Development Site and its associated roads are located at the landward fringe of the Project Site close to Yuen Long, it is reasonable to conclude that not many terrestrial fauna movements across the Development Site are likely, hence collision risk is small. Furthermore, since the KTMDC and the SPR are brackish, it is unlikely that amphibians would have reason to move between these channels and the wetland habitats at NSW. Mortality impacts on terrestrial fauna are therefore likely to be low during the operational period. However, solid barriers will be required around any active works areas within NSW during the construction phase to prevent terrestrial fauna wandering into these areas and being killed.

4.8.81 Fauna occupying roosting and breeding/nursery sites is particularly vulnerable to mortality impacts during site clearance works, as even normally mobile species may temporarily be unable to move in response to danger. Sleeping nocturnal animals and breeding animals, are particularly at risk, especially if they are occupying burrows or tree holes; some bat species and birds may also roost or nest in abandoned buildings. However, most such mortality can be avoided by means of diligent site checks prior to any clearance works and, if necessary, rescheduling of works until nests or roosts have been vacated. Site checks by an appropriately qualified ecologist should, therefore, be undertaken prior to the commencement of clearance work.

4.8.82 A particular problem, which has become significant in some locations in Hong Kong in recent years, is that of birds colliding with transparent roadside noise barriers. The current project does not require such barriers as road traffic noise will not be significant. Temporary solid barriers to be constructed around the development during the construction phase will be opaque.

Table 4.70: Potential direct mortality impacts arising from the project

Criteria	Direct Mortality
Habitat Quality	Not relevant.
Species	Mammals and herpetofauna during construction and (from collisions with traffic) during operation; birds during operation; all species, especially roosting/nesting animals during site clearance.
Size/Abundance	Area where impacts could occur is relatively small in the context of the Assessment Area (as it comprises the Development Site footprint); numbers of terrestrial fauna are small but bird numbers are large.
Duration	Death of terrestrial fauna from construction activities would only occur during the construction phase. Other impacts would be permanent.
Reversibility	Not reversible.
Magnitude	Number of individuals likely to be affected is small, hence magnitude of impact is low.
Impact Severity	Low to Moderate Severity during construction phase in absence of mitigation measures. Low Severity during construction so long as site checks are made prior to clearance works and works rescheduled if necessary and solid barriers are in place around works areas, Low Severity during operation.

Potential Impacts on Recognised Sites of Conservation Importance and Ecological Sensitive Resources within and in the Vicinity of the Project

4.8.83 These impacts include direct loss and disturbance to wetland habitats and consequent effects on fauna of conservation importance including, but not limited to, MPNR, Mai Po Inner Deep Bay Ramsar Site, Mai Po Marshes SSSI, Inner Deep Bay SSSI, Deep Bay Wetland Outside Ramsar Site Priority Site for Enhanced Conservation, the Wetland Conservation Area (WCA) and Wetland Buffer Area (WBA). Impacts on those parts of recognised sites of conservation importance which fall partly within the Assessment Area and Project Site (Mai Po Inner Deep Bay Ramsar Site, Mai Po Marshes and Inner Deep Bay SSSIs, Deep Bay Wetland Outside Ramsar Site Priority Site for Enhanced Conservation and the Wetland Conservation Area (WCA)) have been addressed in the preceding sections; in this section any additional impacts are addressed.

4.8.84 MPNR is contiguous with LC and is connected to it by a pedestrian footbridge though this is blocked at present; hence development at LC has the potential to adversely impact the Nature Reserve. However, all works which are proposed under the project at LC are similar in scale and nature to those currently undertaken during the routine management of the fishponds. Accordingly, no additional impacts are predicted.

4.8.85 Outside NSW, part of Yuen Long Bypass Floodway mitigation wetland falls within the southern east of Assessment Area. The proposed Development Site is separated more than 700m away from these wetlands, therefore direct or indirect impact on the mitigation wetlands arising from the project is not expected.

4.8.86 The SPR ardeid night roost is more than 1km away from the proposed Development Site and screened by the extensive mangroves at the intersection of SPR and KTMDC. During flight line surveys the majority of ardeids were observed returning from outer Deep Bay wetlands, rather than NSW. Therefore, direct or indirect impact on the SPR ardeid night roost arising from the project is not expected.

4.8.87 No other impacts on recognised sites of conservation importance outside the Project Site Boundary arising from the project alone (other than those discussed above) are predicted. The potential for cumulative impacts on these habitats arising from the current and other projects is considered below.

Cumulative Impacts

4.8.88 Cumulative impacts are impacts arising because of the combined impact of the project being assessed and of other projects planned or in progress. All the impact categories described above may apply cumulatively. Cumulative impacts may be incremental, synergistic or interactive. The first of these comprises the additional marginal impact of further development; the second involves two or more impacts for which the whole effect is greater than the sum of its parts; whilst the third involves impacts which arise in other areas which, in combination, then have consequent impacts elsewhere or in the original area.

4.8.89 As the current Project Site lies entirely within the Deep Bay area wetland ecosystem, there is a potential for cumulative impacts of the project and any other projects which may impact on that ecosystem. Such cumulative impacts would have the potential to impact the ecological value of the system as a whole; such impacts would potentially have high significance if, for example, they involved the reduction of wetland area to below a threshold point where impacts were not just incremental but synergistic.

4.8.90 A number of major residential projects in the Deep Bay area would have the potential to have such a cumulative effect. These impacts include cumulative direct loss and disturbance to wetland habitats and consequent effects on fauna species of conservation importance including, but not limited to, MPNR, Mai Po Inner Deep Bay Ramsar Site, Mai Po Marshes SSSI, Inner Deep Bay SSSI, Deep Bay Wetland Outside Ramsar Site Priority Site for Enhanced Conservation, the Wetland Conservation Area (WCA) and Wetland Buffer Area (WBA). However, the requirement for developments within the WCA and WBA to be guided by the EIAO process and TPBG No.12C will ensure that significant loss of wetland habitat area and function is avoided. Similarly, pollution impacts on habitats and species will be avoided due to the requirement for projects within the Hong Kong Deep Bay catchment to adhere to need to ensure that there is no net increase in pollution load to Deep Bay.

4.8.91 In addition, whilst there is no approved or planned project relating to the site, it should be noted that immediately to the southeast of the current project is an area which is currently zoned OU(CDWEA2), where development of a similar nature to the current project may take place in the future. Any such project would also be required to meet the requirements of the EIAO process, TPBG No.12C and the need for no net increase pollution load to Deep Bay. Furthermore, the proponent of any such project would be required to ensure that the impacts of that project on the current Project Site, including any mitigation measures, were fully addressed.

4.8.92 Yuen Long Effluent Polishing Plant (DSD 2019) is an upgrade of the existing Yuen Long Sewage Treatment Works (YLSTW) to cater for future needs in the catchment and improve the water quality of Deep Bay. Stage 1 works commenced in November 2022 and are expected to be complete in 2026/2027. The construction phase of the proposed bridge for the NSW and LC and this project would therefore overlap for about two years. Given the separation distance of approximately 1km between the proposed bridge and the YLSTW, cumulative impacts during both construction and operation phase are not expected. Construction of the Development Site itself will commence after the proposed bridge and therefore cumulative impacts are not expected.

4.8.93 The Yuen Long Barrage Scheme (DSD 2021) involves construction and operation of a pumping station, tidal barrier and link bridge near the section of SPR between Yuen Long Industrial Estate and Shan Pui Ho East Road. This project commenced in May 2023 and is expected to be completed in 2030. Due to its small area (when compared to similar habitats within Deep Bay) and nature and the availability of similar habitats nearby, the direct impact of temporary habitat loss due to the construction of the proposed barrage was considered minor to moderate (AEIAR-228/2021). A similar level of disturbance is expected as a consequence of the proposed bridge construction and as it is close to the barrage location the temporarily impacted area will show considerable overlap. Accordingly, cumulative ecological impacts during both construction and operation phases are anticipated to be insignificant.

4.8.94 In conclusion, in view of the comprehensive measures which are in place under the environmental protection and planning systems and ordinances to protect the Deep Bay wetland ecosystem, significant cumulative impacts arising from this project and any other projects are not predicted, as long as any other potential impacts of the current project are fully mitigated.

4.9 Mitigation and Enhancement Measures

4.9.1 Mitigation measures are required to ensure that the potential ecological impacts of significance described in the previous section are avoided, minimised or compensated. The mitigation measures which are proposed follow the principles and guidelines laid out in Annex 16 of the Technical Memorandum of the Environmental Impact Assessment Ordinance. All proposed mitigation measures are feasible to implement within the context of Hong Kong, will be undertaken on-site (i.e., within the Project Site) and, where appropriate, will be undertaken well in advance of the potential impacts to which they relate to ensure that there is no-net-loss or damage caused by the project to ecological habitats or species of conservation importance. Above and beyond the required mitigation measures, however, it is a primary objective of the Project to enhance the habitats in the Project Area such that there is a net ecological benefit to habitats and species of conservation importance both within the project site itself and within the context of the Inner Deep Bay area.

4.9.2 A principle underlying the proposed mitigation measures is that there is a qualitative (as well as quantitative) difference in the ecological gain that can be effected when management is primarily undertaken for wildlife and management is primarily for fish production, where any ecological benefit is secondary to commercial interests. As a corollary, where management measures seek to achieve a balance between ecological objectives and fisheries operations, it is important that fisheries management operations are carefully planned and implemented to ensure that ecological objectives are not compromised. The following section details potential impacts of ecological significance and how these will be fully mitigated and enhanced following principles of active conservation management.

Summary of Potential Impacts on Habitats and Species of Conservation Importance

4.9.3 The following significant impacts which require to be mitigated to reduce impacts to an acceptable level (or, in the case of impacts of Low Severity, where measures to eliminate or minimise the impact would be beneficial), were identified:

● Loss of total wetland area (i.e., wetland habitats including associated bunds and other functionally linked terrestrial habitats) of 10.7ha due to development at NSW and hence loss in overall wetland function.

● Disturbance impact on wetland habitats (including tidal watercourse, open water, reedbed, marshes and mangrove) and associated fauna of conservation importance (especially waterbirds) due to the construction of the proposed bridge over the SPR and the Development Site, and project operation.

● Loss of 8.2ha fishponds at LC due to conversion to reedbed and 5.1ha at LC to shallow tidal pond; for those species that will not utilise the created habitats is required to be compensated by enhancement of remaining fishponds or open water at NSW and LC (conversion of 4.1ha of fishpond at LC to lily pond is not evaluated as habitat loss as all species currently using fishponds at LC would also use this habitat).

● Impacts on populations of faunal species on conservation importance (including waterbirds) because of habitat loss and disturbance.

● Disturbance impacts (from light) on Brachiaria marsh and mangrove habitat of MPBWF to the southwest of the Development Site at NSW.

● Disturbance impacts to Great Cormorant roost and flightlines at NSW during construction and, to a lesser degree, during operation due to noise, human activity and light.

● Potential construction phase pollution impacts on wetland habitats, especially during the construction of the proposed bridge over the SPR.

● Mortality impact on fauna.

4.9.4 In addition, whilst other predicted impacts in the absence of mitigation measures are predicted to be of low severity, measures to avoid, minimise or compensate for these potential ecological impacts should be implemented where feasible.

4.9.5 The mitigation measures proposed to address these predicted impacts are detailed below. Details are given of the proposed mitigation measures, with an explanation of the impacts which will be avoided, minimised or compensated by the adoption of each of these measures. Much of the proposed mitigation has been achieved by modification of the MLP and by protection and enhancement of wetland habitats at the NSW WEA and the LCNR. Full details of the habitat enhancement and management proposed at NSW WEA and LCNR are presented in the Conservation and Management Plan, see Appendix 4.5. The following paragraphs provide the rationale behind the proposed enhancement works and their potential for mitigation of predicted impacts. Given that the Project Proponent intends to follow the “Additional Option” of the Public Private Partnership (PPP) Scheme of the New Nature Conservation Policy (NNCP), while the design, construction and establishment of the Conservation Portion of the Project (i.e. LCNR and NSW WEA) will be implemented by the Project Proponent, the long-term management of the Conservation Portion will be taken up by the Government, subject to (i) the Government’s acceptance of the Project Proponent’s application/proposal for and (ii) the Project Proponent’s compliance with the requirements of the NNCP PPP Scheme. The detailed design, construction, management, and maintenance of the LCNR and NSW WEA may be subject to further review and adjustments as appropriate, subject to the agreement of relevant authorities including EPD, EEB and AFCD.

Mitigation for Loss of Wetland Habitat

4.9.6 Following the completion of the development and the implementation of the CMP the gross wetland area will be 151.8ha. There will be no-net-loss of reedbed habitat, achieved by conversion of Typha marsh and open water into reedbed and the creation of compensatory reedbed at LC. There will be net loss of fishpond habitat area (9.7ha) at LC, by conversion into reedbed, lily pond and shallow tidal pond. There will be total of 9.9ha net loss of marsh habitat, including net loss of 0.1ha Brachiaria marsh and approximately 2.0ha Typha marsh under the Development Footprint and 7.8ha Typha marsh outside the footprint to be converted into reedbed and fishponds. However, following implementation of the CMP there will be no-net-loss in the wetland function. Proposed habitats are shown in Figure 4.14 and detailed in the following table, while the following sections detail how it is proposed that this functional increase will be achieved so that there is no loss of wetland function in the Project Site.

Table 4.71: Current and proposed habitats in the Project Site (ha) and their ecological value

Habitat	Current Area (ha)	Current Ecological Value	Proposed Area (ha)	Predicted Ecological Value	Net Change
Tidal Watercourse	0.2	Low to Moderate	0.2	Low to Moderate	0
Tidal Watercourse	0.3	Moderate	0.3	Moderate	0
Fishpond¹	6.5	Moderate to High (inactive)	12.3	High (actively managed)	+5.8
Fishpond¹	51.6	High (active)	36.1	High (active)	-15.5
Open Water	2.6	Moderate to High	1.6	High	-1.0
Lily Pond	-	N/A	4.1	High to Very High	+4.1
Shallow Tidal Pond	-	N/A	5.1	High to Very High	+5.1
Reedbed	41.0	High	49.4	Very High	+8.4
Brachiaria Marsh	6.1	Low to Moderate elsewhere	6.0	Moderate elsewhere	-0.1
Brachiaria Marsh	0.9	Moderate (south of NSW)	0.9	Moderate to High (south of NSW)	0
Cyclosorus Marsh	4.1	Low to Moderate	4.1	Moderate to High	0
Typha Marsh	9.8	Low	0	N/A²	-9.8
Grassy Bund	16.9	Low	11.1	High	-5.8
Wooded Bund	9.7	Low elsewhere	9.0	Moderate	-0.7
Wooded Bund	0.7	High as cormorant roost	0.7	High as cormorant roost	0
Bare Soil	0.9	Low	0	Low	-0.9
Mangrove	2.6	Moderate (NSW)	2.6	High (NSW)	0
Mangrove	8.4	High (LC)	8.7	High (LC)	+0.3
Plantation	9.6	Low elsewhere	9.4	Low elsewhere	-0.2
Plantation	0.1	High as cormorant roost	0.1	High as cormorant roost	0
Grassland	0.8	Low	0.2	Low	-0.6
Managed Grassland	1.3	Low	1.3	Low	0
Developed Area	3.5	Low	14.5	Low	+11.0
Total	177.3		177.3		0

Notes:

1. Excludes bund and bare soil areas.

2. Remaining Typha Marsh outside of the development footprint will be converted to reedbed and fishpond (approx. 7.8ha)

4.9.7 Where possible, the distribution of wetland habitats has been left unchanged to minimise disturbance; the locations of retained (enhanced and restored) and newly created habitats are shown on in the CMP (Appendix 4.5). Further details of the distribution of the different retained and created habitats are provided in Table 4 of the CMP, while details of the measures to be applied in the different habitats are also provided in the CMP.

Mitigation for Impacts and Compensation in respect of Fishpond and Open Water Habitats

4.9.8 Enhancement of fishpond and open water habitat will be achieved at NSW WEA and LCNR by reduction of disturbance, reprofiling of fishponds, conversion of some fishponds to lily pond and shallow tidal pond, management of vegetation (especially bund vegetation, encouraging roosting and foraging by large waterbirds), management of fish populations of suitable numbers and species to provide food for waterbirds, and management of water levels to include regular drain-down.

4.9.9 At LCNR, existing fishpond structure will largely be retained to minimise construction phase disturbance and to keep it at a similar level to that which routinely takes place in active fishpond areas. However, some bunds will be removed to merge smaller into larger fishponds (the latter being preferred by most large waterbirds), and some will be lowered to create submerged berms on which waterbird species (especially ardeids and spoonbills) will be able to walk whilst foraging thus increasing the effective foraging area). Most fishponds will be internally reprofiled to create a range of water depths, from shallow margins suitable for foraging when fishponds are full, to deep areas where conditions (especially water temperatures) will remain more stable. Six fishponds in the southwest will be converted to a shallow tidal pond of 5.1ha by reprovisioning of a former sluice to create habitat for species which favour shallow brackish water and mudflats (such as those along the SPR) such as Northern Shoveler, Black-winged Stilt, Pied Avocet, Black-headed Gull and shorebirds. Three fishponds, totalling 4.1ha, will be amalgamated and will be planted and managed as a lily pond. This will be utilised by species which also use typical fishponds, but will be especially attractive to smaller ardeids, Little Grebes and other species which favour this habitat.

4.9.10 Enhancement of fishponds at NSW WEA will, broadly, utilise similar methods to those detailed for LCNR. Most of the larger waterbodies at NSW will be retained with their current form. Two patches of reedbed next to the active fishponds at the north of NSW, where reed colonized the abandoned fishponds, will be converted to fishponds. Some smaller waterbodies, largely in the east of the WEA, which currently do not have wooded bunds, will be amalgamated by removal of some intermediate bunds. Edges of larger waterbodies will be deepened to reduce the spread of terrestrial vegetation into fishponds.

4.9.11 Except for the shallow tidal pond, where fish and invertebrate food supplies will be replenished by the tidal cycle, and the lily pond, which will be kept free of larger fish, fishponds will be stocked with fish and drained regularly to encourage use by waterbirds. It is proposed that two fishponds be drained per month throughout the year; however, frequency of drain-down and location and number of fishponds to be drained will be informed by results from monitoring of species and numbers of target waterbird species.

4.9.12 Wildlife using both NSW and LC will also benefit from reduced human disturbance. Existing fisheries operations and associated human activities will cease. Houses will be removed, and no residents will remain, hence there will be no night-time human activity. At other times access will be restricted to those with permission and/or a requirement to be present for ecological and fisheries management or monitoring purposes. All dogs, which are highly disturbing and are known to prey on wildlife (including the locally very rare Eurasian Otter), will be excluded.

4.9.13 Further details of mitigation measures to compensate for impacts on birds using fishpond habitat and explanation as to how ecological and fisheries management will be fine-tuned with reference to monitoring findings to meet performance targets are provided in the accompanying CMP section 1.12.

4.9.14 With the implementation of the proposed measures to increase and diversify ecological function, together with active adaptive management of the stocking and drain-down regime, and, in particular the conversion of habitats which are currently primarily managed for fish production to habitats which will be primarily managed to benefit wildlife, it is predicted that the value of active and inactive fishponds will increase from Moderate to High Value, to High Value, whilst the shallow tidal pond and lily pond are predicted to be of High to Very High value. With these measures in place, it is predicted that there will be no overall loss in function of the fishpond/open water habitat arising from the project.

Mitigation for Impacts and Compensation in respect of Reedbed and Marshes

4.9.15 Currently, the area of reedbed present at NSW (41.0ha) is the largest area of this habitat present in Hong Kong. There is, however, a degree of fragmentation because of the reedbeds’ origin as fishponds, and some reedbeds are only seasonally wet, especially at the fringes, permitting the colonisation of terrestrial vegetation, especially grasses, but also including trees. Of the 41.0ha of reedbed currently present, 6.2ha is under the proposed Development Site footprint. The loss of this area will be compensated by the conversion of 8.2ha of fishponds and associated bund areas to reedbed at the proposed LNCR. Furthermore, the remaining Typha marsh outside of the development footprint (7.8ha) will all be converted in to reedbed and fishpond. These would ensure that there is no-net-loss in reedbed area; following implementation of the project, reedbed area will increase to 49.4ha (net increase of 8.4ha), all of which will be managed as permanently wet reedbed of Very High ecological value. Thus, overall, there will be a net area gain and a net functional gain as permanently wet reedbed is generally of higher ecological value than seasonally wet reedbed (e.g. Poulin et al. 2002, Allcock et al. 2018).

4.9.16 Enhancement of reedbed habitats will be achieved by converting seasonally wet reedbed to permanently wet reedbed through active management of water levels, by providing a continuous block of reedbed in the WEA to reduce habitat fragmentation from bunds or other habitats, by reducing disturbance around existing reedbeds (currently there is no restriction on human access to the reedbeds at NSW), by managing vegetation (especially by controlling the colonisation of exotic Typha angustifolia and terrestrial vegetation), by managing suitable stocks of fish and shrimp for foraging waterbirds (such as Purple Heron, Yellow Bittern and Eurasian Bittern), by creating suitable reedbed/open water interface (which is favoured by some reedbed species) and by managing suitable water levels.

4.9.17 Currently there are 7.0ha of Brachiaria marsh, 4.1ha of Cyclosorus marsh and 9.8ha of Typha marsh in NSW; most of which are embedded in existing reedbed to form a mosaic. The development will result in 0.1ha loss of Brachiaria marsh and 2.0ha loss of Typha marsh, the significance of impacts for both are considered minor due to the small size of habitat affected and the limited ecological value, in particular for the Typha marsh. Typha is an invasive exotic wetland plant which is routinely removed from manged wetlands in Hong Kong.

4.9.18 The existing areas of Typha marsh outside the Development Site footprint will all be converted into reedbed and fishpond habitat, while the remaining Brachiaria marsh and Cyclosorus marsh will be retained. The retained areas of Brachiaria marsh and Cyclosorus Marsh will be managed to prevent the establishment of exotic and non-wetland plant species and existing stands of unwanted exotics, notably Typha, will be removed. Some Brachiaria marsh and Cyclosorus marsh may be stocked with fish and/or shrimps with the aim of establishing self-sustaining populations as food for smaller piscivorous waterbirds. However, stocking of fish and shrimps to this habitat is not planned and other waterbodies will be left unstocked to reduce predation by fish on amphibians and invertebrates.

4.9.19 With the implementation of the proposed measures, it is predicted that the ecological function of all areas will be increased and there will be no loss in function of reedbed and marshes once the enhancement measures are implemented. Further details of mitigation measures to compensate for impacts on birds using reedbed are provided in Section 4.9 respectively as well as in the CMP sections 1.12.13 – 1.12.18.

4.9.20 To avoid significant construction phase loss of reedbed, it is proposed that all areas of compensatory reedbed within LCNR are formed and managed for at least one wet season prior to the commencement of any development works within the proposed Development Site footprint at NSW. The extent and scope of the wetland enhancement work to be completed before commencement of the vehicular bridge construction and Phase I property development will be further reviewed as considered appropriate to ensure that there will not be any net loss of ecological function at any time. The compensatory reedbed within LCNR is not anticipated to be significantly disturbed during the construction of the development, because it is located entirely outside the disturbance exclusion zone and the zone of reduced density (see Figure 16b). Species present in the reedbed are also less likely to be disturbed than those using more open habitats.

Restoration of Mangrove Habitat

4.9.21 No mangrove habitat will be lost because of the project. However, some mangrove habitat at LC has been converted to fishponds in recent years, whilst other mangrove areas have been damaged by deposition of mud removed from fishponds during fishpond reprofiling or by dumping of rubbish. Whilst it is not proposed to reduce fishpond areas in the LCNR, it is proposed to restore 0.3ha of mangrove and mangrove associate habitat at LC by narrowing bunds around the periphery and in areas currently occupied by temporary housing and associated structures. Unwanted structures, rubbish and fill will be removed, as will unwanted plants included exotic mangrove species. Intertidal areas at LC where exotic mangroves will be removed and dumped material will be removed are shown in Figure 4.17 Further details are provided in sections 1.12.23 – 1.12.26 of the CMP.

Mitigation for Impacts to Plantation

4.9.22 Direct impacts on plantations, including all plantation areas utilised by roosting cormorants, have been minimised in the design stage. There will be no loss of trees used by roosting Great Cormorants. However, 0.4ha of plantation of low ecological value will be lost due to the development. Loss of this plantation is of low ecological significance, however, this loss will be compensated by the 0.2ha of tree and shrub planting buffer within the NSW WEA along its northwestern edge where the potential for disturbance by human activities is greatest and 0.2ha of tree and bamboo planting on an island to be formed from current bund area at LC, as well as planting within the development itself.

4.9.23 All plantation areas used by cormorants will be preserved and included in the NSW WEA. Mature trees or tall shrubs along bunds will be retained and an ongoing programme of supplemental planting of suitable tree species will be undertaken to retain and enhance the roost for Great Cormorants. The protection which will be given to the roost trees by their inclusion in the NSW WEA is a considerable potential benefit to Great Cormorant in the longer term by ensuring that the roost trees are protected from human disturbance or from damage (for example from fires). Further measures to minimise impacts to the Great Cormorant roost are detailed below.

Mitigation Measures to Minimise Disturbance Impacts on waterbirds and other disturbance-sensitive species

4.9.24 Potential disturbance impacts to wetland habitats were minimised at the design stage by locating the Development Site footprint in the southwest of NSW, closest to existing developments at Yuen Long. The detailed design of the layout has also considered the presence of the Great Cormorant roost and has avoided any development within 150m of the roost trees, the distance within which it is predicted that roosting birds might be disturbed during the operation phase. Furthermore, to take into account the possibility that roosting cormorants may show greater avoidance of high rise development than is predicted, location of high rise towers has been adjusted such that the great majority of the roost trees will be more than 400m (the maximum predicted extent of construction phase impacts) from these and all of the trees used by roosting cormorants in most recent winters are outside the 200m exclusion zone.

4.9.25 However, despite these measures, it is predicted that some disturbance impacts would occur to birds and other disturbance-sensitive species using wetland habitats, especially during construction, and these potential impacts require further mitigation. Accordingly, the following measures are proposed to ensure that disturbance impacts on bird species of conservation importance and other sensitive species using wetland habitats are minimised and reduced to an acceptable level:

a. All works areas to be surrounded by a 3.5m high solid opaque barrier during the construction phase;

b. Construction of the Development Site to be phased in order that the maximum construction area is restricted to 5.6ha. This phasing would lengthen the duration of construction activities but would reduce the magnitude and area of the disturbance impacts (see Table 4.72 below as compared to Table 4.42) leading to an overall reduction in severity.

c. Use of bored piling with a shroud rather than percussive piling during construction of the high-rise buildings, to minimise the noise produced during construction of these buildings. This will greatly reduce construction noise: percussion piling creates noise levels of 75dB at 200m and 69dB at 400m, bored piling with a shroud would only generate 61dB at 200m and 55dB at 400m;

d. Restriction on the use of piling during winter months (October – April) so that this is conducted only between the hours of 09.00 and 16.00, to avoid disturbance during the hours when Great Cormorants are roosting;

e. The Development Site abutting the wetland to be surrounded by a 2.0m high wall and fence and dense buffer planting of trees and shrubs at least 10m wide of which at least 7m will be in the WEA (see Section 1.14 of the CMP) during the operational phase

f. To minimise potential light impacts to roosting Great Cormorants, low level lighting to be used in public areas facing the roosts, the use of uplighting will be minimised by using full cut-off lighting design where possible, and floodlighting and spotlights will be avoided;

g. Access to LCNR and NSW WEA to be controlled during both construction and operational phases.

Habitat	Phase 1		Phase 2		Phase 3
Habitat	Exclusion Zone	Zone of Reduced Density	Exclusion Zone	Zone of Reduced Density	Exclusion Zone	Zone of Reduced Density
Tidal Watercourse	3.75	5.76	2.56	6.66	4.77	5.67
Fishpond	-	3.47	-	4.47	2.51	4.17
Open Water	0.14	0.99	0.03	1.07	0.06	0.67
Reedbed	6.13	10.89	4.45	8.62	1.56	6.74
Brachiaria Marsh	1.38	3.87	1.15	3.55	2.44	1.07
Cyclosorus Marsh	0.75	1.77	0.10	2.11	0.10	0.37
Typha Marsh	0.82	1.89	0.38	1.14	0.19	0.67
Grassy Bund	1.15	1.55	0.73	2.96	1.75	2.81
Wooded Bund	0.69	3.78	0.33	3.54	1.11	1.12
Mangrove	0.62	1.85	1.24	1.49	2.79	-
Grassland	0.10	-	0.05	0.05	-	0.10
Managed Grassland	-	0.33	-	-	-	-
Plantation	2.04	-	1.21	2.50	2.50	13.98
Total	17.57	36.15	12.23	38.16	19.78	37.37

4.9.26 Mitigation for the disturbance impacts (based on exclusion zones and zones of reduced density for the most disturbance-sensitive species) will be provided by enhancement of the open water habitat at LC during the construction phase, and by the provision of shallow tidal pond designed and managed to provide suitable conditions for the species currently using the tidal habitat.

4.9.27 To ensure that disturbance impacts on waterbirds are not significant during the construction phase of the proposed bridge over the SPR, it is proposed that enhancement of 15.5ha of wetland habitat at LC be completed prior to commencement of construction of the proposed bridge (which will result in disturbance of up to 5.2ha of tidal watercourse and 1.2ha of open water habitat for the most disturbance-sensitive species of waterbirds (Table 4.41). Enhancement works on the shallow tidal pond and fishpond (Phase IA in Figure 4.15) will be completed 12 months prior to commencement of the bridge link to ensure that the enhanced habitats will be fully functional before bridge construction commences.

4.9.28 As noted above, a shallow tidal pond will be formed by reprovisioning a sluice at LC to the adjacent tidal watercourse, providing a tidal connection. This will provide an additional 5.1ha of intertidal habitat. A similar intertidal gei wai at MPNR is effective at providing habitat for shorebirds, ducks, ardeids and spoonbills. Fishponds in the LCNR will have additional stocking with trash fish during the construction phase for the development to provide further mitigation for disturbance impacts. Stocking and draindown regime will be adaptive and will be informed by results from monitoring of the species and number of individuals using the NSW WEA and LCNR and the SPR and KTMDC within the Assessment Area.

4.9.29 Indicative numerical targets for bird species for the utilisation of the NSW WEA and LCNR are detailed in the following table, final targets will be set based on bird numbers using the Project Site and the potentially disturbed section of the SPR prior to the commencement of any construction (see also paragraph 1.18.1 of the CMP). As a corollary, in the event that construction or operation of the project leads to greater than predicted disturbance impacts (for example from a greater than predicted number of cyclists using NSW Road), the numerical targets for the NSW WEA and LCNR will be adjusted and the management regime adapted accordingly, for example by additional fish stocking or fine-tuning habitat conditions to meet the needs of particular species experiencing greater disturbance impacts than predicted.

4.9.30 Further details regarding design, operation and management of the NSW WEA and LCNR are provided in the CMP, together with further details regarding the monitoring methodology prior to commencement of the construction phase and during the construction and operational phases of the project.

4.9.31 Details of measures to minimise the noise impacts during construction are given in the Noise Impact Assessment chapter. In general, noise impacts will be minimised by:

a. Good site practice and noise management, the site practices listed below will be followed during each phase of construction to minimise potential noise impacts:

i. Only well-maintained plant to be operated on-site and plant to be serviced regularly during the construction programme;

ii. Silencers or mufflers on construction equipment to be utilised and to be properly maintained during the construction programme;

iii. Mobile plant, if any, to be sited as far from WEA and other noise-sensitive receivers (NSRs) as possible;

iv. Machines and plant (such as trucks) that may be in intermittent use to be shut down between work periods or to be throttled down to a minimum;

v. Plant known to emit noise strongly in one direction to be, wherever possible, orientated so that the noise is directed away from the WEA and nearby NSRs; and

vi. Material stockpiles and other structures to be effectively utilised, wherever practicable, in screening noise from on-site construction activities.

b. Use of quiet Powered Mechanical Equipment (PME);

c. Phasing of construction activities to minimise concurrent operation of PME;

d. Minimising number of PME working concurrently: owing to the relatively large site area there is a good opportunity to locate the noisy activities towards the central part of the site. The building structures will also provide a self-screening effect to the nearby NSRs. By controlling the number of items of equipment (particularly those relatively noisy ones) working near the existing NSRs, construction noise impact arising from the site can be effectively reduced;

e. Use of temporary movable noise barriers for construction of access road/bridge. In terms of the access road/bridge construction, owing to the proximity to NSRs, the use of temporary mobile noise barriers between noise sources and NSRs will be effective in reducing the potential construction noise impact from road/bridge construction;

f. The minimum effective height of the mobile noise barriers should be as such that no part of the noise sources should be visible from the target NSRs to be protected. The guidelines given in the Booklet entitled “A Practical Guide for the Reduction of Noise from Construction Works” issued by EPD to be referenced in the design of the temporary acoustic barriers. Barriers should have no openings or gaps, and preferably have a superficial surface density of at least 10 kg/m². Where required, temporary barriers of sufficient height with skid footing and a cantilevered upper portion can be erected within a few meters from stationary plant, and at a practicable distance from mobile plant operating over a small area or using a well-defined route, to alleviate potential construction noise impact;

g. In addition to temporary mobile noise barriers, certain types of PME such as generators and compressors can be shielded by machine enclosures, giving a noise reduction of 10dB (A) or more;

h. Significant noise impact during the construction phase can be avoided when noise management is regarded one of the key components of the construction works. The following good site practices are recommended for incorporation into contract documentation.

i. The Contractor should comply with and observe the Noise Control Ordinance (NCO) and its current subsidiary regulations;

j. Before the commencement of any construction works, the Contractor should submit to the Engineer for approval the method of working, equipment and sound-reducing measures intended to be used at the site;

k. The Contractor should devise and execute working methods that will minimise the noise impact on the surrounding NSRs; and should provide experienced personnel with suitable training to ensure that these methods are implemented.

Mitigation for Disturbance (Light) Impacts on MPBWF in Mangrove and Brachiaria Marsh to the southwest of the NSW Development Site

4.9.32 MPBWF is present in the mangrove and Brachiaria Marsh located at the southwestern corner of NSW. Increased light levels because of development may impact the breeding behaviour of this species by acting as a deterrent to flashing used for courtship display (Yiu 2012). However, the species is known to occur in locations close to light sources including the HKSAR Boundary security fence at Tsim Bei Tsui and Mai Po and Fairview Park (Yiu 2011), as well as the location at NSW (which receives light from Yuen Long Industrial Estate). Measurements of ambient light at locations where the species was present in May 2015 recorded flashing behaviour at sites with light levels up to 0.49 lux, and ambient light levels where the species was present showed little difference from those where the species was not recorded (TPAB 2021).

4.9.33 Accordingly, as a precautionary measure, it is proposed to avoid shining bright light directly onto where the MPBWF is present and to minimise the increase in ambient light levels arising from the development; especially during the MPBWF breeding season. To this end, the following mitigation measures are proposed:

a. Residential towers will be ‘single aspect’ with no windows facing the SPR;

b. Solid opaque barriers 3.5m high to be erected along the interface between any project works area and the mangrove/grassland area and maintained for the duration of construction;

c. Bridge to be provided with a high parapet and low-level (in height) lighting;

d. Bridge and any other lighting of roads and public areas of the project to be directional and directed inwards to the development;

e. All lights should be located at least 20m from intertidal habitats and should be directed away from these habitats, with screening provided as much as possible between the two;

f. Lighting to be used in public areas on the western side of the development will be placed at a low level physically in order to reduce spillage;

g. Use of lights emitting a longer wavelength, such as sodium vapour lights or red LED lights. These should provide sufficient light for human vision but would be less visible to MPBWF and should impact less on courtship behaviour. Orange or red lights should be preferentially selected, and white LED lights should be avoided on those parts of the development closer to the areas occupied by MPBWF. Ultraviolet and fluorescent lightings should be avoided throughout;

h. Minimise the use of uplighting by using full cut-off lighting design where possible, and avoid using floodlighting and spotlights;

i. A ‘green wall’ to be formed by a dense hedge of bamboo Bambusa tuldoides (which grows to a height of 3 – 5m) to be planted and maintained permanently on the southwest side of the Development Site this would be established prior to the commencement of residential development;

j. Existing plantation currently forming a strip between the mangrove/grassland area and the Development Site to be maintained and enhanced.

k. No night time construction activities will be permitted, thus removing the requirement to provide lighting to facilitate night time working.

4.9.34 These mitigation measures are considered to address the impacts of lighting from both higher and lower levels, including low-rise buildings and street lighting.

4.9.35 In addition, the mangrove area in LC will be enhanced and restored and will provide additional mangrove and mangrove associate habitats for MPBWF.

Mitigation for Indirect Impacts to Deep Bay Intertidal Area

Mitigation for Disturbance to nearby Habitats

4.9.36 The Master Layout Plan conforms to the concept of development being located in that part of the Project Site which is further from the core Deep Bay wetlands. The north of NSW (the closest area to the important wetlands in Deep Bay, the Ramsar site and the KTMDC) will be used for creation of the NSW WEA, and no development is proposed in this area. Buildings close to the perimeter of the Development Site will be low-intensity Single Family Townhouses, with development intensity increasing with increasing distance from wetlands and towards the south-west of the site (which is located closest to Yuen Long Town and Yuen Long Industrial Estate). Disturbance to nearby habitats will be minimised by the above design.

4.9.37 Disturbance to ecologically sensitive nearby habitats will also be minimised by the creation of the WEA. The WEA provides protection to areas to the north of NSW, especially the lower sections of the KTMDC which are of high ecological value for large waterbirds, as it will buffer these areas from the construction and operation of the Development Site.

4.9.38 Water discharged from existing waterbodies on site during drain-down prior to reprofiling will be transferred to other waterbodies within the site and will not be discharged into Deep Bay to avoid impacts to the nearby habitats.

Mitigation for Pollutant Runoff and Surface Runoff

4.9.39 Pollution of wetland areas, in particular any areas with a hydrological connection to the wider Deep Bay wetland ecosystem has the potential to be a significant adverse ecological impact. Pollution is perhaps most likely to occur during the construction phase, but a potential risk will remain during the operational phase. Measures to minimise pollution risks and reduce these to an acceptable level are described below.

4.9.40 During the construction phase any pollutant runoff and surface water runoff from the residential area will be treated on site. During the operational phase surface water run-off will be connected to the mains drainage system. No direct runoff from the Development Site enters the WEA, which will have its own, distinct, hydrological management regime (see Section 1.16 of the CMP).

4.9.41 Measures proposed to minimise water quality impacts during the construction stage are summarised below. No additional discharge of pollutants into the inland watercourses is proposed in the planned development.

4.9.42 Practice during site construction to minimise surface runoff and the chance of erosion will include:

a. Planning of earthworks (including drains, dykes and embankments) and implementation of erosion and sedimentation control facilities to control water discharge from the site during construction;

b. Provision and maintenance of sand/silt removal facilities such as sand/silt traps and sediment basins;

c. Covering exposed slope/soil surfaces by a tarpaulin or other means, as far as practicable, especially during rainstorms. Exposed soil areas to be minimised to reduce potential for increased siltation and contamination of runoff;

d. Earthwork final surfaces to be well compacted and subsequent permanent work or surface protection to be immediately performed;

e. Provision and maintenance of oil/ grease removal facilities at appropriate locations where chance of contamination of runoff is high;

f. Adequately cover and temporarily seal manholes to prevent silt and debris getting into the drainage system or foul sewers;

g. Provision of onsite sedimentation facility and reuse of treated runoff (e.g. for dust suppression and vehicle wheel washing) to minimise need for offsite discharge;

h. Appropriate storage, treatment and disposal of any chemical waste to prevent spillage into watercourses;

i. Appropriate treatment, neutralisation and disposal of concrete washings.

4.9.43 Chemical spillage during construction will be minimised by the following measures:

j. Onsite storage of chemicals, fuels and oils will be restricted to minimal quantities necessary for daily construction activities;

a. Suitable containers will be used to hold any chemical wastes to avoid leakage or spillage during storage, handling and transport;

b. Chemical waste containers will be suitably labelled, to notify and warn the personnel who are handling the wastes to avoid accidents;

c. Storage area to be selected at a safe location on site and secured by locks, and adequate space to be allocated to the storage area;

d. Designated storage areas will have pollution prevention facilities, e.g. impermeable bunding around the storage area with capacity at least 110% that of the chemicals/chemical wastes stored;

e. Vehicles and equipment involved in activities with potential for leakage or spillage will be correctly maintained;

f. In case of an accidental spillage of chemicals, immediate clean-up / remedial action to be implemented to minimise water quality impacts.

4.9.44 During the construction stage, temporary sanitary facilities, such as portable chemical toilets, will be employed on-site where necessary to handle sewage from the workforce. A licensed contractor will be responsible for appropriate disposal and maintenance of these facilities. These measures will avoid and minimise the pollutant runoff.

4.9.45 In the operational phase, all sewage effluent generated in the Development Site would be conveyed to the Project’s private underground SPS, and then drained into the government sewer for discharge into the YLEPP. The sewage will not enter watercourses. There is sufficient capacity at the existing sewerage system to receive the additional sewage flow from the Project, hence no adverse sewerage impact from the Project would be anticipated.

4.9.46 No discharge of drain down water from the freshwater ponds to the adjacent rivers will be permitted during operation phase (only the existing ponds with intertidal connections to the river will be allowed to naturally drain during ebb tide). Water from drained ponds will be pumped to and temporarily stored at adjacent ponds within the LCNR and the NSW WEA.

4.9.47 Within the Development Site, following the “On-site Detention” approach, the following measures will be implemented:

a. Runoff within the residential area will be minimized by adopting bioswales and permeable paving as part of the design of pedestrian and open space areas;

b. Operation and maintenance of the bioswales shall follow the BMPs to be developed as part of the detailed design of the bioswales;

c. BMPs for stormwater discharge will also be adopted to minimize stormwater pollution arising from the Project. Runoff from roads within the residential area will be drained through a separate drainage system equipped with gullies, oil interceptors and grit traps to remove the pollutants;

d. The internal drainage system will convey runoff to an underground detention tank (with capacity of approx. 31,750m³for temporary storage and settling out of suspended solids;

e. The underground detention tank will be regularly desilted to remove the settled solids and main storage capacity;

f. Outflow from the detention tank will be retained onsite (e.g. pumped to designated ponds at NSW WEA with approval from the Site Ecologist or Wetland Manager) and used for irrigation where practicable. Surplus outflow from the detention tank will be discharged to SPR only when the onsite capacity is exhausted.

Mitigation for Dust Disturbance

4.9.48 Mitigation measures to minimise dust emissions during construction are detailed below. Proposed measures include:

a. Watering of unpaved areas, access roads, construction areas and dusty stockpiles to be undertaken at least eight times daily during dry and windy weather. Watering of the haul road would be undertaken up to once every construction hour during dry or windy weather. Water sprays may be either fixed or mobile to follow individual areas to be wetted as and when required;

b. Effective water sprays to be used during the delivery and handling of all raw sand and aggregate, and other similar materials, activities and areas where dust is likely to be created and to dampen all stored materials during dry and windy weather;

c. Stockpiles of sand, aggregate or any other dusty materials greater than 20m³to be enclosed on three sides, with walls extending above the pile and 1 metre beyond the front of the pile;

d. Areas within the construction site where there is a regular movement of vehicles to have a paved surface and be kept clear of loose surface material;

e. The Contractor to restrict all motorised vehicles within the construction site, excluding those on public roads, to a maximum speed of 20km per hour and confine haulage and delivery vehicles to designated roadways inside the Site;

f. Construction working areas to be restricted to a minimum practicable size;

g. The Contractor to provide a wheel washing facility at the exits from work areas to the satisfaction of the Engineer and to the requirements of the Commissioner of Police. Water in wheel washing facilities and sediment to be changed and removed respectively at least once a month;

h. The Contractor to submit details of the wheel washing facilities, which shall be usable prior to any earthworks excavation activity on the construction site. The Contractor to provide a hard-surfaced road between any washing facility and the public road;

i. In the event of any spoil or debris from construction works being deposited on adjacent land, or steams, or any silt being washed down to any area, then all such spoil, debris or material and silt to be immediately removed and the affected land and areas restored to their natural state by the Contractor to the satisfaction of the Engineer;

j. If spoil cannot be immediately transported out of the Site, stockpiles should be stored in sheltered areas;

k. Plant and vehicles to be inspected annually to ensure that they are operating efficiency and that exhaust emissions are not causing a nuisance. All site vehicle exhausts should be directed vertically upwards or directed away from ground;

l. Path for complaints and handling procedures should be set up and implemented;

m. Waste arising from site clearance during construction will be disposed in accordance with the recommended good waste management practices and EPD’s regulations and requirements.

Mitigation for Light and Visual Disturbance

4.9.49 In addition to the specific measures which will be implemented to avoid light disturbance impacts on MPBWF, such impacts on other fauna will be addressed by surrounding the perimeter of the residential portion of the project with suitable perimeter planting. This will minimise the disturbance to surrounding ecosystems from visual observation of human activity and from domestic light sources. Night-time light sources will be kept to a minimum and will not be directed towards disturbance-sensitive wetland areas, especially the Great Cormorant roosts.

4.9.50 All construction work will be contained within visual barriers to minimise disturbance to wildlife. Construction will not be conducted at night, which will avoid night-time disturbance. Night-time security lighting during construction will be kept to a minimum and will not be directed towards wetland areas, especially the night-time roosts of Great Cormorants in the NSW WEA and in the southeast.

Mitigation for Habitat Fragmentation

4.9.51 The location of the WEA will maintain existing ecological connections with adjacent wetland habitats, including KTMDC, fishponds at Tin Fook Wai and the Inner Deep Bay Ramsar Site. Habitats that will be retained and enhanced within the WEA will be contiguous with wetlands within the immediate environs and will form part of the wider mosaic of the Inner Deep Bay wetland system. Fragmentation of the WEA from other wetland habitats has thus been avoided.

Mitigation for Impacts on Wildlife

4.9.52 The overarching principle of measures to mitigate for potential impacts on wildlife is the requirement to ensure that during both the construction and operational phase of the project the numbers of all species of conservation importance are no lower than numbers observed during baseline surveys. Mitigation measures must be targeted towards the requirements (e.g. habitat preferences) of these species, henceforth referred to as ‘target species’ for ecological mitigation measures. Details of the target species are presented in following sections and Section 1.10 of the CMP.

4.9.53 Target species are thus those species of conservation importance, based upon criteria provided by IUCN Red List of Threatened Species (2022), List of State Key Protected Wild Animals (National Forestry and Grassland Administration and Ministry of Agriculture and Rural Affairs) and/or Fellowes et al. (2002), which were recorded in significant numbers at NSW and LC during the baseline ecological surveys and are predicted to be potentially impacted by the proposed development including those bird species recorded within the section of the SPR potentially affected by disturbance impacts due to the construction and operation of the project (including the new bridge link).

Mitigation for Impacts on Eurasian Otter

4.9.54 Although not recorded during surveys, Eurasian Otter, which is considered to be Near Threatened (Loy et al. 2022) have been recorded at NSW (McMillian et al. 2019); it occurs at low densities and is assumed to have large territories and is generally difficult to survey. Accordingly, a precautionary approach has been taken with this species and it is considered a Target Species.

4.9.55 Any impacts on Eurasian Otter are predicted to be of very low severity. Some disturbance to this species may arise from noise during construction but this will be reduced by avoiding night-time construction and through the noise minimization measures detailed above. Measures will be taken to avoid, minimise and compensate for any impacts in the NSW WEA by providing undisturbed waterbodies with vegetated fringes (especially reedbed) which are suitable habitat for Eurasian Otter. Although Eurasian Otter has not previously been recorded at Lut Chau, the long-term reduction in human disturbance and from dogs and the creation of suitable habitat should allow this species to use the site; this will be encouraged by provision of an artificial otter holt. Holt design will be reviewed taking account recent experience at other Hong Kong sites, to give the best chance of successful occupation. The holt will be located on an island. Two similar artificial otter holts will also be created in the NSW WEA. Proposed locations for artificial holts are shown on Figures 6 and 7 of the CMP).

Mitigation for Impacts on Terrestrial Mammals

4.9.56 Impacts on terrestrial mammals are predicted to be of low severity. Nevertheless, construction and operational stage impacts to terrestrial mammals will be avoided and compensated by the creation of the NSW WEA and LCNR. Terrestrial habitats in NSW WEA and LCNR (especially wooded areas) will provide suitable habitat for terrestrial mammals; however, the greatest benefit which these species will gain is the removal of disturbance by humans and, especially, dogs, reducing both fragmentation and direct impacts. Overall, therefore, the project will be of benefit to larger terrestrial mammals.

4.9.57 There may be some mortality to small mammal species during construction, but this would not continue into the operational phase and the species involved would rapidly recolonise suitable parts of the site from adjacent habitats. Long-term impacts to these small terrestrial mammals (rodents and shrews) are not considered to be significant.

Mitigation for Impacts on Bats

4.9.58 Bats forage around the wetland areas of NSW at night. Moderate numbers were recorded during night-time surveys of other species groups. Although not specifically identified by trapping, all individuals seen or heard with bat detectors appeared to be Japanese Pipistrelle, which is commonly found in the Deep Bay wetlands.

4.9.59 It is likely that some existing structures are utilised by small numbers of bats for roosting. Thus, removal of these structures could result in direct mortality (if structures are demolished when bats are present) and roosting habitat loss. While only of small significance (as the bat species involved are common and widespread), construction stage impacts on bats will be avoided and minimised by checking structures for presence of bat roosts prior to demolition and, if necessary, adjustment to phasing of demolition to avoid destroying roosts used by nursing bats with small young. Loss of roosts will be compensated in the operational stage by provision of bat boxes (see paragraph 1.16.37 of the CMP). Overall, therefore, impacts to bats in the construction stage will be small and reduced to an acceptable level and there will be no adverse impacts in the operational stage.

Mitigation for Impacts on Birds

4.9.60 The overarching requirement in respect to mitigation measures is to ensure that there is no adverse impact on species of conservation importance occurring in significant numbers (the target species) because of the project at any stage (bridge construction, residential construction and operation). Birds of conservation importance potentially displaced by the project were detailed above.

4.9.61 However, while the predicted disturbance impacts on bird species of conservation importance have been detailed, the actual number of birds to be accommodated within the Project Site (i.e. the target to be achieved in order that all adverse impacts on bird numbers are fully mitigated) is the combined total of the birds predicted to be impacted by the project (including those displaced by development and change to wetland habitats and those displaced from the SPR) and the birds present within the Project Site which are not predicted to be impacted/displaced.

4.9.62 As stated previously, the target bird species for mitigation measures are those species of conservation importance recorded in significant numbers at NSW and/or LC during the baseline ecological surveys and are predicted to be potentially impacted by the proposed development, including those bird species recorded within the section of the SPR potentially affected by disturbance impacts due to the construction and operation of the project (including the new bridge link). The following table details those species that have been identified as target bird species (a total of 30 species) . As shown in Table 2 of Appendix 4.2c, a precautionary approach has been taken during the selection of bird target species. The peak counts of each species of conservation importance recorded at NSW, LC and in the affected parts of the SPR during the baseline surveys were identified (columns headed B, C and D respectively). These counts were then summed (column headed B + C + D = E) and compared to the peak count detailed in the most recently available Hong Kong Bird Report 2018 (column headed A) (HKBWS 2021). If the combined total was higher than 1% of the peak count in 2018 (column headed Is E > 1% of A?), then the species was further reviewed for its suitability as a target species, with species that either occur only rarely in Hong Kong, are not typical of the habitats present or are not wetland-dependent not considered suitable targets (as is detailed in the final column of Table 2 of Appendix 4.2c).

4.9.63 Quantitative numerical targets (i.e. the number of individuals to be accommodated in the NSW WEA and LCNR) have been calculated based on the sum of the following:

● Mean of the number of individuals recorded in baseline surveys in the Project Site at NSW and LC.

● Mean of the number of individuals predicted to be disturbed from the Sham Pui River, based on whichever is the higher of high and low tide counts during the baseline surveys.

4.9.64 These combined quantitative targets are detailed in the table below. As noted previously, all targets are indicative at this stage and will be subject to review and revision following pre-construction baseline surveys.

Table 4.73: Target Bird Species and Quantitative Mitigation Targets for bird species potentially impacted by the project

Species	Mean (per survey) within NSW and LC	Mean (per survey) under Zones of Disturbance Impacts of SPR	Proposed Numerical Target¹
Northern Shoveler	6.8	4.1	10.9
Eurasian Teal	3.8	4.3	8.1
Little Grebe	19.9	0	19.9
Black-faced Spoonbill	1.2	0	1.2
Yellow Bittern	0.4	0	0.4
Black-crowned Night Heron	2.9	0.1	3.0
Chinese Pond Heron	30.6	5.8	36.4
Eastern Cattle Egret	13.7	0	13.7
Grey Heron	12.9	2.3	15.2
Purple Heron	1.3	0	1.3
Great Egret	47.5	2.6	50.1
Little Egret	54.3	5.9	60.2
Great Cormorant	50.7	1.3	52.0
Black Kite	7.1	0.1	7.2
Black-winged Stilt	2.3	23.8	26.1
Pied Avocet	4.9	8.4	13.3
Little Ringed Plover	4.3	0.1	4.4
Common Redshank	0.5	4.2	4.7
Marsh Sandpiper	1.6	0.1	1.7
Wood Sandpiper	1.0	0.2	1.2
Common Greenshank	0.8	2.7	3.5
Black-headed Gull	90.3	4.2	94.5
White-throated Kingfisher	2.4	0.3	2.7
Pied Kingfisher	2.8	0.1	2.9
Collared Crow	16.1	0	16.1
Chinese Penduline Tit	9.8	0	9.8
Pallas's Grasshopper Warbler	0.7	0	0.7
Red-billed Starling	25.6	0	25.6
White-cheeked Starling	7.5	0	7.5
White-shouldered Starling	2.6	0	2.6
Note: 1. Numerical target levels are calculated based on mean number of the species recorded from NSW, LC and disturbed sections of tidal watercourse, and apply across the two sites as a whole.

4.9.65 For the target numbers of birds to be achieved, it will be necessary for the NSW WEA and the LCNR to support a larger number of birds than these areas do at present; they will also need to support species displaced from the SPR for which current conditions in the Project Site are not suitable, or at least optimal, currently.

4.9.66 The paragraphs below detail how the proposed habitat design and management measures will accomplish these mitigation objectives (over and above the requirement to meet mitigation requirements for impacts on habitats of ecological importance); and provide evidence to demonstrate that it with the proposed mitigation measures in place, meeting the quantitative targets for bird numbers is feasible and achievable. Further details are provided in the CMP to which cross-reference has been made as appropriate. The table below details the habitats that will be utilised by the target bird species.

Table 4.74: Habitats that will be utilised by target bird species

Common Name	Fishpond#	Grassy Bund	Wooded Bund	Lily Pond	Shallow Tidal Pond	Reedbed	Brachiaria Marsh	Cyclosorus Marsh	Mangrove	Natural Watercourse (Tidal)	Plantation
Northern Shoveler	*			*	*
Eurasian Teal	*			*	*
Little Grebe	*			*		*
Black-faced Spoonbill	*				*
Yellow Bittern				*		*	*	*
Black-crowned Night Heron	*		*		*	*	*	*	*	*	*
Chinese Pond Heron	*	*		*	*	*	*		*	*	*
Eastern Cattle Egret	*	*		*			*
Grey Heron	*				*				*	*	*
Purple Heron						*	*
Great Egret	*				*				*	*	*
Little Egret	*	*		*	*	*	*	*	*	*	*
Great Cormorant	*	*	*							*	*
Black Kite	*	*	*		*						*
Black-winged Stilt	*	*			*
Pied Avocet	*				*
Little Ringed Plover	*	*			*
Common Redshank	*				*
Marsh Sandpiper	*				*
Common Greenshank	*				*
Wood Sandpiper	*				*
Black-headed Gull	*				*
White-throated Kingfisher	*		*	*	*	*	*		*	*	*
Pied Kingfisher	*				*
Collared Crow	*	*	*		*				*		*
Chinese Penduline Tit						*
Pallas’s Grasshopper Warbler						*	*
Red-billed Starling		*	*								*
White-cheeked Starling		*	*								*
White-shouldered Starling		*	*								*

Key: # includes fishponds when drained

*habitat important for breeding, foraging or roosting

Mitigation for Impacts on Bird Species Utilising Reedbed, Brachiaria, Cyclosorus and Typha Marshes

4.9.67 There will be no-net-loss of reedbed, accordingly there will be no-net-loss of the area of habitat potentially available to birds on the Project Site. As shown in previous sections, during the construction of the proposed bridge over the SPR 0.8ha of reedbed will fall in the exclusion zone and 3.0ha in the zone of reduced density for the most disturbance sensitive species respectively. In terms of the Brachiaria, Cyclosorus and Typha marshes, no bird species of conservation importance are dependent of these habitats; and the species diversity and abundance are low.

4.9.68 Thus, the predicted number of birds of conservation importance to be impacted is very small and furthermore birds will be buffered from disturbance by barriers around the construction site, together with the presence of the intervening shelterbelt. Accordingly, compensatory habitat provision is considered unnecessary during the bridge construction phase.

4.9.69 However, during the residential construction phase, the exclusion zone (7.0ha of reedbed and 5.3ha of marshes) and zone of reduced density (10.0ha of reedbed and 6.2ha of marshes) is larger and while the number of birds predicted to be displaced is still relatively small, it is considered that compensatory habitat provision should be undertaken, if only on a precautionary basis.

4.9.70 Accordingly, a net increase of 8.4ha of reedbed will be provided, including the creation of 8.2ha new reedbed within the LCNR and conversion of Typha marsh and open water to reedbed within NSW WEA. The remainder of the Brachiaria and Cyclosorus marshes will also be enhanced in the northeast of NSW WEA. Construction stage impacts on fauna will be minimised by the phasing of the Development Site and the phasing of reprovision and enhancement to allow at least one growing season to elapse before impacts due to construction of residential development on current areas of these habitats commences (see 4.9.20 above).

4.9.71 In the operational phase, 9.6ha of these habitats, largely reedbed, will fall within the exclusion zone and 9.9ha within the zone of reduced density for the most disturbance sensitive species. However, in view of the relatively small number of individuals potentially impacted, and the proposed measures to increase the ecological function of these habitats for the bird species of concern (see below), it is considered that the overall net effect of the project will be to increase the value of NSW for reedbed associated fauna.

Enhancement and Creation Measures to be undertaken on Reedbed Habitat

4.9.72 Existing reed areas in the WEA will be maintained. Recognised measures (Hawke and José 1996) to rehabilitate existing degraded reedbeds will be implemented in the areas of existing reedbed which are degraded from drying out, encroachment by non-reed plant species, lack of management of water levels, and lack of conservation driven management of the reedbeds. Terrestrial grasses which are colonising the dry fringes of waterbodies are of lower ecological value than the reedbed; in such areas the vegetation will be removed (and the area deepened, if necessary) and replaced by new areas of reed. Reedbeds will largely be kept under about 20cm of water but will include deeper areas with a maximum water depth of 2.5m which will be too deep for the growth of reeds and will form small patches of open water within the reedbed which will be important for foraging Great Bitterns and Purple Herons. It is proposed that all reedbeds will be wet reedbeds (i.e., will be under standing water for most of the year), which are generally of higher ecological value to wildlife (Poulin et al. 2002) and especially birds, than reedbeds which are dry for much of the time. Studies have shown that reedbed passerines actively select wet reedbeds as there have higher densities of food items (Chernetsov 1998), and that many passerines, not only those associated with reedbeds, select wet habitats over dry, especially in the autumn (Peach et al. 1996). Wet reedbeds tend to be more homogenous as terrestrial species are unable to colonise and have been shown to display higher bird diversity than dry reedbeds.

4.9.73 Existing bunds will be strengthened and repaired as many of these are in disrepair and no longer properly retain water, where necessary the height of these will also be increased. New water control structures will also be installed (see Figures 6 and 7 of the CMP). These measures will permit the implementation of an appropriate water level management regime and will allow for water levels within the existing reedbed to be raised. This will inhibit the re-colonisation of non-reed plant species, keep the reedbed bed wet throughout the year and maintain small areas of open water within the reedbed for Great Bitterns and Purple Herons. The interface between reedbeds and areas of open water are important habitats for a wide range of waterbirds. Enhancement measures are detailed in the CMP.

4.9.74 In addition, 8.2ha of new reedbed will be created at the proposed LCNR, to the east of the proposed lily pond, in the locations shown in Figure 4b of the CMP to mitigate for the loss of this habitat in the proposed development site . This new reedbed will be created in what are currently active fishponds and will be formed by removal of the existing internal bunds of the fishponds to be modified, the material from which will be used to raise internal levels and to form a profile suitable for the growth of reeds, following which they will be extensively planted with reeds and re-flooded. Creation and maintenance measures are detailed in the CMP.

Enhancement and Creation Measures to be undertaken on Brachiaria and Cyclosorus Marshes

4.9.75 The existing areas of Brachiaria and Cyclosorus marshes within the WEA will be retained, including the area to the southwest of NSW which is used by MPBWF (see Figure 4.11). These areas will be maintained and will be managed to prevent the establishment of unwanted exotic or terrestrial plant species with removal of existing stands of unwanted exotics. Enhancement measures are detailed in the CMP.

Mitigation for Impacts on Bird Species Utilising Fishpond Habitat

4.9.76 There will be a net direct habitat loss of fishpond habitat including 8.2ha fishponds in LC to be converted to reedbeds, and a further 9.2ha of fishponds will be converted to lily pond and shallow tidal pond at LC. Whilst most bird species using the LC fishponds will also utilise shallow tidal ponds, in some cases in larger numbers, and hence will not be adversely impacted by this conversion, some fishpond species do not favour this habitat. Furthermore, many of the species which currently utilise those fishponds to be converted to reedbeds do not use reedbed in significant numbers. Accordingly, there will be a consequential loss of habitat for wetland bird species that utilise fishpond habitat.

4.9.77 To address this quantitative habitat loss, it is proposed to undertake qualitative enhancement of the fishpond to be retained and enhanced at LC to meet the needs of target bird species as described in paragraphs below and further detailed in the CMP. Such enhancement would be conducted at the early stage of Phase IA and Phase IB of the Development and prior to the creation of reedbed and shallow tidal pond as far as possible. At least 15.5ha of enhanced wetland habitat will be formed and will be functional at LC prior to commencement of bridge construction (see programme on Figure 4.15). The extent and scope of the wetland enhancement work to be completed before commencement of the vehicular bridge construction and Phase I property development will be further reviewed as considered appropriate to ensure that there will not be any net loss of ecological function at any time.

Enhancement Measures to be undertaken on Fishpond Habitat

4.9.78 Large fishponds retained or formed by combining smaller fishponds will provide areas of open water up to 2.5m in depth with shallow water margins (0 - 40cm depth) to allow range of depths to provide a variety of conditions for waterbirds. Currently fishponds at both LC and NSW are mostly steep-sided and lacking shallow water suitable for foraging large waterbirds except when drained or being drained. Ponds will be re-profiled to create a mixture of gently sloping bunds and submerged berms to provide extensive areas of suitably shallow water for foraging waterbirds irrespective of water levels. Pond edges will be contoured to allow for structural diversity to fishpond edges both when full and during draindown (see Figures 5a and 5b of the CMP). Ponds will be drained on a monthly cycle (i.e. two fishponds drained per month throughout the year) to encourage use by large waterbirds. Draining fishponds provide an important resource for many target bird species, notably ardeids and spoonbills as fish and shrimp become trapped in shallow water and thus become more accessible.

4.9.79 The fishponds at LCNR and NSW WEA will be maintained as open water habitat by maintenance of water levels at a sufficient depth to limit the spread of vegetation into the fishponds. In addition, maintenance of high-water levels in the wet season will allow the control of unwanted vegetation by herbivorous fish. However, many bunds will become more vegetated with emergent vegetation in the shallow water created by bund reprofiling as this habitat is attractive to some target species.

4.9.80 Three fishponds at LC will be amalgamated and reprofiled so that water depths are suitable for establishment of lilies Nymphaea spp. which provides a habitat that is particularly favoured by some target species, notably Little Grebes and smaller ardeids, though it will also be utilised by other bird species that use fishponds. This lily pond will be kept free of herbivorous and omnivorous fish (which would eat the lilies) and will not be included in the drain-down cycle. Creation and maintenance measures are detailed in the CMP.

Mitigation for Impacts on Bird Species Utilising the Tidal Watercourse on the SPR

4.9.81 In addition to the mitigation measures to address disturbance/displacement to birds currently utilising the Project Site, on-site compensatory habitat provision is required for those birds currently using the SPR that may be disturbed. Some of these species will utilise the enhanced fishpond habitat, especially when fishponds are drained, but others, notably most shorebirds, are less attracted to fishpond habitat. Impacts on these species will be addressed by the conversion of 5.1ha of fishponds at the south of LC to shallow tidal pond with a sluice connection to the KTMDC.

4.9.82 Since shallow tidal pond provision is required to mitigate for bridge construction phase impacts on waterbirds, the work required to convert these fishponds and restore the sluice connection will be completed, and the shallow tidal pond will be operational prior to commencement of bridge construction (see below and programme on Figure 4.15).

Creation and Enhancement Measures to be undertaken on Shallow Tidal Pond Habitat

4.9.83 Six fishponds in the southwest will be converted to a shallow tidal pond of 5.1ha to create shallow water/tidal habitat for species which favour shallow brackish water and mudflats (such as those along the SPR, see above). The shallow tidal pond will be managed in a similar manner to the shallow tidal ponds at MPNR, in particular Gei wai no.22 which is operated as intertidal from 1st November to 31st March each year to provide foraging opportunities for waterbirds (WWF HK 2013).

4.9.84 The shallow tidal pond will be formed by removal of the existing internal bunds of the fishponds to be modified, the material from which will be used to raise internal levels and to form shallow islands. The existing sluice will either be repaired or replaced. In addition, a large sluice will be installed between this shallow tidal pond and the fishponds to the east which will provide the opportunity in the future to also manage this fishpond as intertidal should this be considered desirable. Creation and maintenance measures are detailed in the CMP.

Additional Measures Required to Mitigate for Impacts on Particular Bird Species and Species Groups

4.9.85 In addition to habitat-related mitigation measures to address impacts on birds; species and group-specific mitigation measures are required to address some potential impacts. These are described below.

Mitigation for Impacts on Great Cormorants

4.9.86 As is noted above, the Development Site footprint has been carefully adjusted to ensure that no development will fall within the distance (150m), within which it is predicted that roosting birds might be disturbed during the operation phase. In addition, since there is some uncertainty as to the potential disturbance effect of high-rise development on roosting cormorants, the disposition of high-rise towers has been adjusted so that these are more than 400m from the main group of roost trees in the southern part of the roost. Construction across the Development Site will be split into three distinct phases to reduce the magnitude of disturbance impacts arising from the site at any given time. Based on the disturbance distances given in Table 4.56, disturbance to the more important parts of the Great Cormorant roost would only be anticipated during Phase 2 of the residential construction, as this is the only time when high rise construction is located within 400m of the roost trees. Nevertheless, measures to minimise potential noise disturbance are proposed throughout the construction period because some of the outlying (less frequently used) roost trees are located within this disturbance distance during the other residential construction phases and as a precautionary measure to ensure that the impacts to the roost are minimised. Potential noise disturbance during the construction phase will be minimised by the measures described. Use of bored piling rather than percussive piling will minimise the potential noise created during the construction of the high-rise buildings on site. Bored piling is significantly quieter than percussive piling: decibel levels of percussive piling are 64dB at 200m and 58dB at 400m, while those of bored piles with a shroud are 44dB at 200m and 38dB at 400m. To put these values in context, 50dB is a typical background noise level in a quiet suburb or a normal domestic conversation, while 40dB is a typical background level in a library (IAC-Acoustics 2016). Most cormorants are only present in the roost overnight, so piling will be conducted only between the hours of 09.00 and 16.00 during the period when cormorants are present in Hong Kong (October – April).

4.9.87 The Development Site footprint has been designed so that only a small number of cormorants fly over the area to be occupied by the proposed development, with all flightlines directed away from the development, towards wetlands to the north of the roost. During surveys in 2021, no Great Cormorant departures were directed towards the west, southwest or south; thus none of these was potentially impacted.

4.9.88 Because the Great Cormorant roost will be entirely contained within the WEA, human activity at the roost will be restricted and visual disturbance to roosting birds will be minimised. Visual disturbance from human activity within the development will be avoided by provision of a 3.5m solid opaque barrier during construction and a 2m wall with screening planting during operation. The Nature Trail in the WEA will be closed at night to avoid disturbance to cormorants roosting close to this trail. Any disturbance to the cormorant roost would therefore be limited to birds roosting at the perimeter of the WEA. As there is currently no restriction to human activity at NSW, and there are currently people and dogs living on-site, the overall disturbance from human activity around the cormorant roost would be reduced compared to existing conditions.

4.9.89 These issues were discussed in detail at the TPAB and we note that the decision on TPB Appeal No. 1 of 2017 (1/17) found that ‘…there would be no loss in the carrying capacity of the NSW roosts, and no less number of Cormorants would use the NSW roosts. Instead, there would likely be a significant increase in both carrying capacity and number of Cormorants roosting at NSW. (para. 145 of the decision).

4.9.90 Light disturbance to the Great Cormorant roost is only considered to be a significant problem if lights are directed towards the roosting locations, and the overall increase in light from the development is not expected to have a negative impact on the cormorant roost. Background light levels are already relatively high due to artificial light emanating from Yuen Long Industrial Estate and Yuen Long. Furthermore, both the Lok Ma Chau roost and the new Mai Po San Tsuen roost (which is located adjacent to the Shenzhen River) are subject to even higher levels of night-time lighting due to their proximity to Shenzhen City. Measures taken to minimise light disturbance to MPBWF are also expected to benefit roosting cormorants and residual impacts are expected to be of Low Severity. Great Cormorant will be a target species for both the NSW WEA and the LCNR (see CMP).

Mitigation for Impacts on Wintering Starlings

4.9.91 Flocks of starlings were recorded at NSW in winter of 2021/2022, mostly around the active fishponds. The commonest species was Red-billed Starling, with smaller numbers of White-cheeked Starling and White-shouldered Starling. The number of individuals involved was small relative to the Deep Bay populations of these species described by Carey et al. (2001) and hence potential construction and operational phase impacts are considered to be of Low Ecological Significance. All three species apparently also roosted in reedbeds on site, and observations during evening visits for counts of roosting cormorants indicated that larger numbers of starlings were present during the evening than on morning bird surveys. The reedbed area used by roosting starlings is located in the northern part of NSW, outside the area where disturbance impacts are predicted during both construction and operation, and the proposed development is not expected to have an impact on the starling roost.

4.9.92 Though potential impacts on wintering starlings are considered to be of low significance, all three of Red-billed, White-cheeked and White-shouldered Starlings meet the criteria for recognition as target species for the project. Accordingly, management measures will be undertaken to provide suitable habitat for these species to compensate for the impacts and enhance the habitat. Details of the requirements of starlings and proposed management measures are provided in the CMP paragraphs 1.10.36 – 38 and 1.16.37.

Mitigation for Impacts on Breeding White-shouldered Starlings

4.9.93 In addition to the small number of individuals recorded in winter, White-shouldered Starlings have been recorded (AEC 2016) breeding at NSW around some of the buildings close to the active fishponds. This was formerly considered a rare breeding species in Hong Kong (Carey et al. 2001) but the population has increased considerably in recent years, especially in the Deep Bay area.

4.9.94 Foraging habitat loss to White-shouldered Starlings at NSW has been avoided in the design stage and this species will benefit from habitat enhancement in the NSW WEA and, especially, the LCNR. However, nests in abandoned buildings will be lost when these are demolished. Direct impacts on breeding White-shouldered Starlings will be avoided by checking buildings for nests prior to demolition and phasing accordingly. In addition, as noted above, White-shouldered Starling will be a target species for the NSW WEA and LCNR and loss of nest sites will be compensated by nestbox provision (see CMP paragraphs 1.10.42 and 1.16.36). Experience elsewhere in Hong Kong has shown that availability of nest sites is a limiting factor on the breeding population of this species and that it will breed at high densities if nestboxes are provided (AEC 2010). The project is, therefore, expected to be of net benefit to this species as the availability of nest sites will increase.

Mitigation for Impacts on Collared Crow

4.9.95 The global population of Collared Crow has recently been assessed as Vulnerable due to ongoing declines throughout the range (BirdLife International 2018). The population in Deep Bay is probably of global significance. The species was regularly recorded on surveys with a maximum count of 30 individuals at LC and five individuals at NSW. It is possible (but not confirmed) that the species may breed in trees on site.

4.9.96 Accordingly, Collared Crow is proposed as a target species in the NSW WEA and LCNR. Management measures will be undertaken to provide suitable habitat for the species to compensate the impacts and enhance the habitat. Measures will include retention of large trees as potential nesting habitat and protection from disturbance. In addition, this species will benefit from drain-down of stocked fishponds as this will provide additional foraging opportunities for dead small fish, a preferred food source of this species. Details of the management are provided in the CMP paragraph 1.16.44.

Mitigation for Impacts on Flightlines and Foraging Habitats from Tung Shing Lane and Shan Pui River Egretries

4.9.97 TSL Egretry was last occupied in 2020. It is thus anticipated that the relevant impact would not occur. Nonetheless, the layout of the Development Site has considered potential impacts to the flightlines from TSL egretry. Should the egretry becomes active again, the relevant impact would still be of low severity. The SPR Egretry is located further away from the development footprint as compared to the TSL Egretry. Recent surveys showed that the development footprint will not affect flightlines associated with this egretry.

4.9.98 Ponds and reedbeds at NSW and the SPR provide foraging habitat for both Little Egret and Chinese Pond Heron. Whilst the predicted impact on both species is low, as the number of birds utilising areas which will be developed or disturbed is small, impacts during the construction phase of the project will be minimised by shallow tidal pond creation and fishpond and reedbed enhancement works at NSW and LC which will be undertaken in advance of the commencement of the development, thus ensuring continuity of habitat provision. In addition, as target species for the project, habitat will be actively managed specifically to meet the needs of these species throughout the construction and operational periods (see CMP).

Mitigation for Impacts on Reptiles

4.9.99 There were very few records of reptiles during dedicated surveys, but these were supplemented by observations of four reptile species of conservation significance at NSW: Reeves’ Turtle, Chinese Soft-shelled Turtle, Common Rat Snake and Burmese Python, which were recorded on one and two occasions. The undisturbed bunds around the abandoned fishponds are suitable for reptiles and it is expected that other species should also be present but may be undetected due to their secretive nature.

4.9.100 Direct impacts on reptiles at NSW will be minimised by the retention and enhancement of suitable wetland habitat in the NSW WEA and the two reptile species of conservation importance, Reeves’ Turtle and Chinese Soft-shelled Turtle are proposed to be target species in the NSW WEA and LCNR on a precautionary basis. Management measures will be undertaken to provide suitable habitat for these species. Details of the proposed management are provided in the CMP. In addition, both these species are susceptible to (illegal) human activities, either persecution (Burmese Python) or capture for food (Reeves’ Turtle). The protection afforded by the NSW WEA and the LCNR and the restrictions on human access to these areas will be of direct benefit to these species.

Mitigation for Impacts on Amphibians

4.9.101 All species of amphibians recorded at NSW are common and widespread in lowland habitats throughout Hong Kong and none are of conservation importance. Abundance of these species was generally relatively low considering the large size of the site, especially in comparison with other wetland habitats in Hong Kong. This may relate to the habitats present (active fishponds and reedbeds, for example, are often relatively poor habitat for amphibians) or the salinity, as many waterbodies are slightly brackish.

4.9.102 Impacts of the project on amphibians are therefore predicted to be of Low severity and mitigation of impacts is not required. Nevertheless, habitats to be provided in the NSW WEA and the lily pond at LCNR will provide suitable freshwater wetland habitat for amphibians.

Mitigation for Impacts on Odonates

4.9.103 The odonate surveys at NSW revealed a moderate diversity (24 species) of common species and a high abundance of a few species. Ruby Darter is of Local Concern and Mangrove Skimmer is of Global Concern, Fellowes et al. (2002) and were both recorded during surveys. Impacts of the project on these species are expected to be low and none are considered to meet the criteria for consideration as target species for the project.

4.9.104 Nevertheless, the undisturbed waterbodies with extensive emergent vegetation will provide good conditions for odonate breeding, especially for the more adaptable species, while the reprofiling of fishponds at LCNR to provide more extensive emergent vegetation around the fringes of fishponds can be expected to provide significant benefit to common odonata species.

Mitigation for Impacts on Butterflies

4.9.105 Impacts to butterflies of conservation importance are predicted to be of low severity and no specific mitigation is proposed for these species. Three butterfly species of conservation importance were recorded at NSW, Common Awl, Pale Palm Dart and Danaid Egg-fly; these species are listed as Local Concern according to Fellowes et al. (2002) and Very Rare, Rare and Uncommon respectively by Chan et al. (2011); however, given the low numbers recorded these are not considered to qualify as target species.

Mitigation for Impacts on MPBWF

4.9.106 Concentrations of MPBWF were recorded in Mangrove and Brachiaria marsh to the south of NSW and mangrove around the southern end of MPNR. Moderate numbers were recorded in the channel to the south of LC. Small numbers of MPBWF were observed in the mangroves along KTMDC and SPR. No MPBWF were observed in ponds or other habitats at NSW or LC. The species is currently known largely from the Deep Bay area and is restricted to intertidal grassland and mangrove habitats (Cheng et al. 2010, Ballantyne et al. 2011, Yiu 2011). Given the highly restricted distribution of this species, it is a target species for the project.

4.9.107 Direct impacts on MPBWF at NSW have been avoided at the planning stage by avoiding any loss of Mangrove or Brachiaria marsh where MPBWF is present and at the design stage by designing the residential towers as single aspect buildings with no windows facing MPBWF habitat along the SPR. Further measures to minimise disturbance due to light generated by the development are detailed above and include screening of the site from the mangrove and Brachiaria marsh during both construction and operational phases of the project and use of low-level lighting.

4.9.108 These measures were discussed in detail at the TPAB and we note that the decision on TPB Appeal No. 1 of 2017 (1/17) found that with ‘As a matter of evaluation we find that sufficient practical steps are proposed for avoidance, minimization, and compensation of light impact concerning the adults and larvae [of MPBWF]. (para. 155(2) of the decision). It was also noted at page 117 ‘Having regard to the very dense bamboo planting of 4 to 5 or 6m trees, and shrub Planting of 3 to 5 m, we accept the Appellants’ evidence that there would be no significant residual impact on Fireflies’.

Feasibility of Successful Implementation of Proposed Mitigation and Enhancement Measures

Mitigation of Direct and Indirect Impacts on Habitats and Species of Conservation Importance

4.9.109 The EIAO-TM states that ‘all mitigation measures recommended shall be feasible to implement within the context of Hong Kong’. Whilst this is a large-scale project in a Hong Kong and Deep Bay context in terms of the wetland area which is involved, all proposed mitigation measures are ‘tried and tested’ in a Hong Kong context, either as mitigation measures for developments and/or as enhancement measures to improve the ecological function of protected areas. Specifically, fishpond creation and enhancement of a comparable scale to that proposed in the current project have been undertaken successfully at Hong Kong Wetland Park by Hong Kong Government and at Lok Ma Chau by MTR Corporation; while creation and active management of shallow tidal ponds have been carried out both at Hong Kong Wetland Park and MPNR; and creation/restoration of reedbed and vegetated wetlands has been undertaken at all three of these sites as well as at smaller scale projects in the Deep Bay area. It is considered, that the feasibility of the proposed habitat creation, modification and enhancement works is demonstrated beyond reasonable doubt.

4.9.110 All significant impacts on species have been addressed and impacts have been avoided, minimised or compensated. Where possible, direct and indirect impacts on species due to habitat loss have been avoided or minimised by project design, and where reedbed and marsh loss is unavoidable due to the proposed Development Site, this will be compensated by like-for-like habitat reprovision such that there will be no reduction in the potential carrying capacity of the Project Site for species utilising these habitats.

4.9.111 To demonstrate that such enhancement is feasible, baseline numbers of birds at NSW and LC Commercial Fishponds and in the potentially disturbed section of the SPR have been compared with number of birds recorded at managed pond habitat at Lok Ma Chau and managed shallow tidal pond (gei wai) habitat at MPNR (Table 4.75). Comparable data from one of these sites was available for all proposed target species.

Species	Bird Density (mean number of birds per survey/ha)
Species	LC Commercial Fishpond	NSW Commercial Fishpond	Disturbed SPR Section	LMC EEA¹	*MPNR Gei Wai²*
Northern Shoveler	0.1	0.1	0.5	-	22.5
Eurasian Teal	<0.1	<0.1	0.5	0.3	2.41
Little Grebe	0.4	0.2	-	0.7	0.47
Black-faced Spoonbill	<0.1	<0.1	-	0.4	1.70
Yellow Bittern	-	-	-	<0.1	-
Black-crowned Night Heron	<0.1	0.1	<0.1	0.5	0.36
Chinese Pond Heron	0.6	0.3	0.7	0.8	0.45
Eastern Cattle Egret	0.3	<0.1	-	-	<0.1
Grey Heron	0.2	0.2	0.3	1.7	2.68
Purple Heron	-	-	-	-	<0.1
Great Egret	1.1	0.3	0.3	1.3	5.28
Little Egret	1.1	0.6	0.7	0.7	3.82
Great Cormorant	0.4	1.1	0.2	8.5	7.88
Black Kite	0.1	<0.1	<0.1	-	0.90
Black-winged Stilt	0.1	<0.1	2.8	0.5	0.57
Pied Avocet	0.1	<0.1	1.0	-	7.88
Little Ringed Plover	0.1	<0.1	<0.1	-	<0.1
Common Redshank	<0.1	<0.1	0.5	-	1.17
Marsh Sandpiper	<0.1	<0.1	<0.1	-	1.01
Wood Sandpiper	<0.1	<0.1	<0.1	0.2	0.14
Common Greenshank	<0.1	<0.1	0.3	-	2.41
Black-headed Gull	2.4	0.1	0.5	-	7.51
White-throated Kingfisher	<0.1	<0.1	<0.1	-	0.13
Pied Kingfisher	<0.1	<0.1	<0.1	-	<0.1
Collared Crow	0.4	<0.1	-	-	0.49
Chinese Penduline Tit	-	-	-	-	0.25
Pallas's Grasshopper Warbler	-	-	-	<0.1	-
Red-billed Starling	0.6	0.1	-	2.7	0.12
White-cheeked Starling	0.2	<0.1	-	0.3	-
White-shouldered Starling	<0.1	<0.1	-	1.6	-

Note:

1. Bird density at LMC EEA in 2021. Available from https://www.mtr.com.hk/en/corporate/sustainability/environmental_reporting_lmc.html

2. Mean bird number extracted from SMEC (2021) for data of 2019/2020 of ponds. Area information is extracted from WWF (2021) which includes the area of ponds. Ponds managed to provide shallow pond areas at MPNR are nos. 3, 4, 6, 7, 8a, 11, 16/17, 21, 22a, 22b and 23.

4.9.112 Habitat at LMC is managed to meet the requirements of particular bird species utilising pond, marsh and reedbed habitat that were predicted to be impacted by the Lok Ma Chau Spur Line Project (Binnie, Black & Veatch 2002), while the gei wai at MPNR providing shallow water areas are largely managed for shorebirds in spring and autumn and, variously, for Black-faced Spoonbills, duck and Black-faced Spoonbills and ardeids in winter (WWF Hong Kong 2013). Little Egret numbers at LMC vary every year, the density in 2021 was lower than the control areas referenced for that project and lower than in LC. Thus, whilst target bird species for these sites overlap significantly with those proposed for NSW and LC, they are not identical, nor are the habitats. However, the relatively high numbers at these managed and protected sites, compared with the current commercial fishponds at NSW and LC, provide firm evidence for the enhancement that can be achieved in Deep Bay when wetlands are protected and managed with nature conservation as the primary objective; providing assurance that the proposed enhancement measures at NSW are feasible and realistic.

4.9.113 To further demonstrate the feasibility of the proposed measures meeting the required enhancement levels, the actual target numbers that are required to be achieved utilising pond habitat at NSW and LC and the required level of enhancement relative to baseline numbers (i.e. the enhancement ratios) are compared with the enhancement that has been achieved at LMC in Table 4.76. Species that do not primarily favour fishpond habitat, are not included in Table 4.76.

4.9.114 The minimum level of enhancement (the minimum enhancement ratio) is the proportion by which the number of birds per unit area of habitat must be increased such that there is no reduction in the total number of birds after habitat enhancement. For example, if 50 birds are currently present in fishpond habitat which will be lost either due to direct loss, conversion or disturbance elsewhere on the Project Site, and 100 birds are currently utilising the fishpond area to be enhanced; the total number of birds to be accommodated in the enhanced pond habitat is 150 i.e. 1.5 times the original number. This value (1.5) (or 50% more than existing levels) is the Minimum Enhancement Ratio (MER) required to ensure that there is no net loss in habitat capacity. Any increase beyond the MER would result in a net ecological gain.

4.9.115 All calculations are based on the number of birds to be accommodated during the period when maximum impacts are predicted to occur (i.e. during the residential construction phase).

Table 4.76: Minimum Enhancement Ratios required in remaining pond habitat to mitigate for impacts on species due to reduction in pond area

Species	Total number of birds to be accommodated in pond habitat	Minimum Enhancement Ratio Required	Enhancement Ratio achieved at LMC EEA *
Eurasian Teal	1.83	1.0		9.1
Little Grebe	15.88	1.0		1.0
Black-faced Spoonbill	0.38	1.0		5.6
Black-crowned Night Heron	2.63	1.0		7.6
Chinese Pond Heron	23.71	1.0		1.6
Grey Heron	9.25	1.0		9.8
Great Egret	41.46	1.0		2.9
Little Egret	49.58	1.0		0.8
Great Cormorant	45.08	1.7		38.7
Black-winged Stilt	2.29	1.2		0.7
Pied Avocet	4.71	1.0		-
Little Ringed Plover	4.29	1.5		-
Marsh Sandpiper	1.54	1.0		-
Wood Sandpiper	0.83	1.0		3.3
Black-headed Gull	90.29	1.0		-
White-throated Kingfisher	2.13	1.2		-
Pied Kingfisher	2.42	1.3		-

* Data referring to 2021 Bird Target Performance, retrieved from Environmental Committee Meeting Minutes Jan 2022 for Sheung Shui to Lok Ma Chau Spur Line, enhancement ratios are relative to control areas

4.9.116 If the last two columns in Table 4.76 are compared for the eleven species that are proposed targets for the current project and which are targets at the LMC EEA, it will be seen that the enhancement ratios achieved at LMC for most of the species significantly exceed the MER required at NSW and LC, often by a considerable margin. It is suggested, therefore, that there are considerable grounds to be confident that the design and management measures proposed under the current project will also be highly successful in exceeding the required numerical targets.

4.9.117 Since shallow tidal pond habitat is a newly created habitat designed to cater for the requirements of species for which suitable habitat does not currently exist at NSW or LC, demonstration of the feasibility of supporting the required number of birds in the newly-created habitat with reference to the enhancement which was achieved at LMC is not appropriate. However, the proposed shallow pond habitat will have a similar ecological function to gei wai habitat at MPNR. Accordingly, Table 4.77 details target numbers to be accommodated in the proposed shallow tidal pond habitat and compares these targets with the number of birds utilising comparable habitat at MPNR.

Species	Total number of birds to be accommodated in shallow tidal pond habitat	Number of birds per ha to be accommodated in shallow tidal pond habitat	Number of birds per ha at comparable gei wai habitat at MPNR
Northern Shoveler	4.35	0.85	22.5
Eurasian Teal	4.72	0.93	2.41
Black-faced Spoonbill	0.38	0.07	1.70
Black-crowned Night Heron	0.10	0.02	0.36
Chinese Pond Heron	5.80	1.14	0.45
Grey Heron	2.30	0.45	2.68
Great Egret	2.60	0.51	5.28
Little Egret	5.90	1.16	3.82
Great Cormorant	9.13	1.79	7.88
Black-winged Stilt	24.22	4.75	0.57
Pied Avocet	8.40	1.65	7.88
Little Ringed Plover	1.52	0.30	<0.1
Common Redshank	4.20	0.82	1.17
Marsh Sandpiper	0.10	0.02	1.01
Common Greenshank	2.70	0.53	2.41
Wood Sandpiper	0.20	0.04	0.14
Black-headed Gull	4.41	0.86	7.51
White-throated Kingfisher	0.63	0.12	0.13
Pied Kingfisher	0.39	0.08	0.02

4.9.118 Again, for most of the species, comparison of numbers at Mai Po gei wai per unit area with the numbers required to be accommodated in the created shallow tidal ponds at LCNR per unit area, suggests that the bird density could be achieved, often by a considerable margin.

4.9.119 There are no target levels for non-avifauna target species, due to the low occurrence rates, and often low detection levels. Mitigation measures for mammal and herpetofauna target species have already been discussed. Mammal species will benefit from reduced anthropogenic disturbance including the removal of dogs. Herpetofauna species will also benefit from protected and enhanced wetland habitats.

4.9.120 It is concluded, therefore, by making reference to the enhancement on birds using pond and intertidal wetland habitat in existing successful examples that the proposed habitat enhancement and creation measures would mitigate for all significant habitat loss and disturbance impacts on wildlife and that these measures are fully feasible.

4.9.121 In this context it is of relevance that the TPAB decision on TPB Appeal No. 1 of 2017 (1/17) found that should the CMP be implemented that there would be ‘substantial enhancement of the NSW and LC Sites’ ecological value and functions by several times, up to 5 times over a range over species…’ and ‘There is no good reason practical reason why the extent of enhancement at the NSW and LC Sites should not be at least double.”

4.9.122 The TPAB decision also stated that based on a Site Visit in January 2021 that ‘the sorry state of the LC Site indicates that as a matter of common sense and fact, the Site is functioning well below its ecological potential, and there is significant room for enhancement of ecological value and functions’.

4.9.123 As is shown in Table 4.78, it is notable, that despite its being immediately adjacent to MPNR and hence its lying in the heart of the Ramsar Site, both the Literature Review and the Baseline Surveys undertaken for this EcolIA revealed that fewer bird species and bird species of conservation importance are present at LC than at NSW. Also, whilst the totals shown in Table 4.78 are not directly comparable, the much greater number of species seen at MPNR is a clear indication of the potential that might be realised at LC with appropriate conservation management in place. The Project thus provides a major opportunity to restore and enhance LC and allow it to fulfill its ecological potential after decades of neglect.

Table 4.78: Number of bird species recorded at MPNR, NSW and LC

Avifanua	NSW		LC		MPNR*
Avifanua	Literature Review	Baseline Survey	Literature Review	Baseline Survey	MPNR*
Number of species of conservation importance	57	55	38	37	248
Total number of species	138	122	79	79	420
* The “Mai Po bird species list” (WWFHK, 2022)

Summary of Proposed Mitigation and Enhancement Measures and Evaluation of Residual Impacts

4.9.124 Mitigation measures proposed to avoid, minimise and compensate for significant impacts arising from the project are summarised below and any residual impacts are identified.

Table 4.79: Summary of potential impacts, mitigation/enhancement measures, residual impacts and additional ecological enhancement

Potential Impacts	Significance	Mitigation/ Enhancement Measures	Residual Impact and/or additional ecological enhancement
Direct loss of Habitats
Loss of Habitat due to Development	Development will occupy 11.6ha, of which 10.7ha is wetland. Impact would be of Moderate Severity as the area to be lost is moderate in size in the Project Site and Assessment Area context.	Compensation: loss of 11.6ha to development, of which 10.7ha is wetland habitat, will be compensated by enhancement and active management for nature conservation of 154.4ha. The required enhancement and conservation benefit of active management of wetland for nature conservation has a proven track record in Hong Kong.	No significant residual negative impacts, and a predicted net gain in wetland function.
Loss of Fishpond/ Open Water	Impact of net loss of 9.7ha of fishponds would be of High Severity as this area is moderate in the Project Site and Assessment Area context. Loss of 1.0ha of open water is considered to be of Low to Moderate Severity.	Compensation: net loss of 9.7ha of fishpond and 1.0ha of open water habitat will be compensated by enhancement of remaining 48.4ha of fishpond and 1.6ha of open water, together with creation of 4.1ha of lily pond and 5.1ha of shallow tidal pond. The required enhancement has a proven track record in Hong Kong.	No significant residual negative impacts and a predicted net gain in wetland function.
Loss of Reedbeds	Impact of loss of 6.2ha of reedbed at NSW would be of Moderate to High Severity.	Compensation: loss of 6.2ha to be compensated by conversion of fishpond to reedbed. Net reedbed area will be increased by 8.4ha. In addition, existing reedbed areas will be restored and enhanced.	No residual negative impacts and net gains to reedbed area and predicted ecological function.
Loss of Brachiaria and Typha Marshes	Impact of loss of 0.1ha of Brachiaria marsh and 9.8ha of Typha marsh would both be of Low Severity.	Compensation: Brachiaria marsh to be retained will be enhanced. All remaining Typha marsh will be converted to reedbed and fishpond.	No residual negative impacts. A predicted net gain in ecological function.
Loss of Grassy Bund, Wooded Bund and Bare Soil	Loss of 5.8ha of grassy bund, 0.7ha of wooded bund and 0.9ha of bare soil bund, would have impact of Low Severity.	Compensation: net loss of area to be compensated by enhancement of other wetland habitats and loss of function by conversion to wetland habitat of higher ecological value.	No residual negative impacts.
Loss of Plantation	Loss of 0.4ha of plantation, none of which is utilised by roosting cormorants, would have an impact of Low Severity.	Compensation: loss of 0.4ha of plantation will be compensated by planting 0.2ha of trees and bamboos on an island to be formed at Lut Chau and 0.2ha of buffer plantation at the northwest of NSW. In addition, there will be a programme of replacement planting of eucalypts on bunds at NSW.	No residual negative impacts once trees have matured.
Loss of Grassland and Developed Area	Loss of 0.8ha of grassland would have an impact of Low Severity; loss of 2.3ha of developed area, would have a negligible impact.	Compensation: No compensation required.	No residual negative impacts. Net gains in both wetland area and predicted ecological function.
Disturbance Impacts
Overall Disturbance Impacts arising from the project	Impacts potentially of High Severity during construction of proposed bridge, primarily due to disturbance to waterbirds utilising the SPR. Impacts of High Severity during construction of the residential area due to potential disturbance to waterbirds and the Great Cormorant roost. Operational phase disturbance of Moderate Severity because of its lower magnitude and smaller area of impact.	Avoidance: piling activities to be carried out only during daytime in the dry season to avoid the period when cormorants are roosting. Minimisation: construction phase disturbance to be minimised by good site practice, noise management and use of quiet machines (including bored piling), minimising the number of machines working on site at one time and use of visual and noise barrier around works areas. Phasing of construction works to minimise magnitude and area of disturbance impacts. Compensation: Enhancement measures at NSW WEA and LCNR during construction and operational phases (fish stocking, additional drain-down of fishponds, shallow tidal pond operation) to compensate for disturbance to wetland habitats elsewhere. Adaptive management to intensify enhancement measures should disturbance impacts prove greater than predicted.	No significant residual impacts with multiple measures to mitigate disturbance impacts in place.
Light Disturbance to fauna	Bright lights directed towards bird roosts (in particular, Great Cormorant roost) and mangroves/ Brachiaria marsh utilised by MPBWF potentially of High Severity.	Avoidance/Minimisation: light disturbance to sensitive receivers to be avoided/minimised by appropriate design of external lighting and screening. Green wall to be formed using bamboo along west of Development Site to shield MPBWF habitat. High-rise development along SPR to be single aspect with no windows to face river and area of high MPBWF density. Compensation: compensatory habitat provision will be made at the LCNR for MPBWF, far from sources of light disturbance. A tree planting programme will provide additional roost trees for Great Cormorant.	No significant residual impacts with mitigation measures in place albeit the additional potential cormorant roost trees will take time to mature.
Disturbance Impacts on Tidal Watercourse	Impacts to SPR of High Severity during construction of the proposed bridge, primarily due to disturbance to waterbirds. Impacts to wetland habitats and their fauna of High Severity during construction of the residential area. During operation impacts would be reduced to Moderate Severity because of lower magnitude and smaller area of impact.	Avoidance/Minimisation: measures described with respect to overall disturbance impacts are applicable to impacts on this habitat. Compensation: 5.1ha of shallow tidal pond habitat will be created at Lut Chau in advance of commencement of bridge construction to provide compensatory habitat for waterbirds that may be disturbed.	No significant residual impacts.
Disturbance Impacts on Open Water and Fishpond Areas	Only a small area will be affected at NSW, and this is already disturbed and supports low numbers of disturbance-sensitive waterbirds. Accordingly, impact at NSW will be of Low Severity. At Lut Chau there will be a Moderate Reduction as disturbance will decrease.	Avoidance/Minimisation: measures described with respect to overall disturbance impacts are applicable to impacts on this habitat. Compensation: measures to compensate for loss of open water habitat are also applicable.	No significant residual impacts.
Disturbance Impacts on Reedbed Habitat	The area potentially impacted is small during the bridge construction phase, large during residential construction and moderate during operation. However, the abundance of disturbance-sensitive fauna is low, and the habitat provides cover for most species. Accordingly, the impact Severity is considered to be Low during bridge construction, Moderate during residential construction and Low during operation.	Avoidance/Minimisation: measures described with respect to overall disturbance impacts are applicable to impacts on this habitat. Compensation: measures to compensate for loss of reedbed habitat are also applicable.	No significant residual impacts.
Disturbance Impacts on Marshes (Brachiaria Marsh, Cyclosorus Marsh and Typha Marsh)	The number of disturbance-sensitive species and their abundance are both low in these areas in general, hence the impacts are of Low Severity overall. Numbers of MPBWF in the Brachiaria marsh to the southwest of the development footprint are moderate, however impacts on this species would not be additional to those addressed under Light Disturbance Impacts (see above).	Avoidance, Minimisation and Compensation: measures described with respect to overall disturbance impacts are applicable to disturbance impacts on this habitat.	No significant residual impacts.
Disturbance Impacts on Grassy Bund and Wooded Bund	Magnitude of impacts on Great Cormorants would be of Low Severity as only a small part of the roost would be disturbed; disturbance impacts on other species would be of Very Low Severity as only a few individuals would be disturbed.	Avoidance, Minimisation and Compensation: measures described with respect to overall disturbance impacts are applicable to disturbance impacts on this habitat.	No significant residual impacts.
Disturbance Impacts on Mangrove Habitat	Magnitude of impacts is generally low as only a small area at the southwest of NSW would be disturbed and fauna numbers are low. However, this area is of importance to MPBWF. Accordingly potential impacts are of Low Severity in general, but of Moderate Severity to MPBWF. However, impacts on this species would not be additional to those addressed under Light Disturbance Impacts (see above).	Avoidance/Minimisation: measures described with respect to overall disturbance impacts are applicable to impacts on this habitat. Measures to address light impacts on MPBWF apply in respect to that species. Compensation: an additional area of 0.3ha of mangrove will be formed at Lut Chau. In addition, mangrove areas at NSW and Lut Chau will be enhanced/restored by removing dumped materials and invasive plant species.	No significant residual impact, and there will be a net gain in mangrove area and ecological function.
Disturbance Impacts on Grassland and Managed Grassland	Magnitude of impacts would be of Very Low Severity during construction and insignificant during operation.	No mitigation measures required.	No significant residual impacts.
Disturbance Impacts on Plantation Habitat	Plantation habitat is generally of low ecological value and is used by few species; however, plantation utilised by Great Cormorants is of high importance as a roost. Thus, whilst most disturbance impacts on plantation habitat are of Low Severity, impacts on the part of the cormorant roost utilised by roosting cormorant are of moderate magnitude. Accordingly, overall impact is assessed as of Low to Moderate Severity.	Avoidance/Minimisation: measures described with respect to overall disturbance impacts are applicable to impacts on this habitat. Measures to address impacts on roosting Great Cormorants apply in respect to that species. Compensation: measures to compensate for loss of plantation habitat are applicable. In addition, a programme of eucalypt planting to ensure long term sustainability of the cormorant roost will be undertaken.	While noise and disturbance impact to roosting cormorants will be avoided and minimised as far as possible, some disturbance impacts to a small part of the cormorant roost cannot be avoided. This may result in part of the roost being abandoned. However, in normal circumstances there is adequate capacity in other parts of the roost trees to accommodate any displaced cormorants. The displacement of cormorants from one part of the roost to another is of very limited ecological significance and is not considered to be a significant residual impact in the short to medium term. In the long term, once planted eucalypts mature there will be no impact.
Impacts on mammals
Impacts on mammal species of conservation importance	Loss of wetland habitat and disturbance to remaining wetland habitat; however, mammal populations are small and much of the area is unsuitable due to human activity, hence potential impacts of Low Severity.	Avoidance: Entry restricted to NSW WEA and to LCNR to avoid direct disturbance by people and dogs using these locations. Minimisation: Measures to minimise noise disturbance would minimise these impacts. Compensation: Enhancement of habitats at NSW WEA and LCNR would compensate for habitat loss and residual disturbance from the development footprint.	Impacts at NSW WEA due to loss of wetland area and disturbance would be largely balanced by enhancement and protection of remaining habitats. Furthermore, there would be a net ecological gain at LCNR from habitat enhancement and removal of disturbance. Thus, overall the project is predicted to provide a net gain for mammal species.
Impacts on Birds
Impacts on bird species of conservation importance	A total of 30 bird species of conservation importance are potentially affected by the project. However, the number of individuals that might be affected is small in an Inner Deep Bay context, hence the severity of impacts on most species would be Low, while potentially of Low to Moderate severity for Black-winged Stilt. However, overall the severity of impact would be Moderate due to the relatively large numbers of species which may be affected.	Compensation: loss of fishpond habitat to be compensated by enhancement of fishpond function by management measures including, protection, habitat improvement and stocking and draindown regime; no-net-loss of reedbed; disturbance impacts to be addressed by measures to avoid disturbance to fauna of conservation importance, whilst enhanced and newly-created habitats are predicted to support more individuals than current habitats due to higher quality; off-site disturbance to birds using the SPR to be mitigated by creation of shallow pond habitat at Lut Chau and enhanced function of fishpond habitat.	No residual impacts are predicted and both diversity and numbers of bird species are predicted to be significantly higher than baseline levels, delivering a net ecological gain during operation.
Impacts on Roosting Great Cormorants	The NSW Great Cormorant roost is of high ecological importance. Most of the cormorant roost will not be affected, however impacts on the part of the cormorant roost utilised by roosting cormorant are of moderate magnitude. Accordingly, impact is assessed as of Low to Moderate Severity.	Avoidance/Minimisation: Measures described with respect to overall disturbance impacts are applicable. In addition, design of development footprint is such that no part of the roost is within 150m, and tower disposition has been further adjusted so that only a small part of the roost lies within the 400m maximum disturbance distance. Measures will be taken to reduce noise disturbance by use of quiet machinery. Piling will be conducted using bored piles with a shroud, hence will not be audible above existing background noise levels at the roost and will be restricted to 09.00-16.00 when the roost is largely unoccupied. Human access to NSW WEA will be restricted ensuring no human disturbance (which occurs at present preventing use of otherwise suitable trees); construction works will be phased to minimise works area. Compensation: a programme of Eucalyptus planting to ensure long term sustainability of the cormorant roost will be undertaken.	Noise and disturbance impact to roosting cormorants will be avoided and minimised as far as possible, but some disturbance impacts to a small part of the cormorant roost cannot be avoided. This may result in part of the roost being abandoned. However, in normal circumstances there is adequate capacity in other parts of the roost trees to accommodate any displaced cormorants. The displacement of cormorants from one part of the roost to another is of very limited ecological significance and is not considered to be a significant residual impact in the short to medium term. In the long term, once planted eucalypts mature, there will be no impact.
Impacts on bird flightlines	Low severity as the number of individuals affected would be small. In the case of Great Cormorants, any bird displaced are likely to respond by using an alternative roost site still within NSW.	Avoidance: The proposed development footprint positioned in the southwest of the Project Site, avoids the Great Cormorant flightlines.	No significant residual impacts
Impacts on herpetofauna	Four reptile species of conservation importance may experience loss of habitat. All are present in small numbers and impacts are assessed as of Low Severity. No predicted impacts of significance on amphibians.	Avoidance: Entry restricted to NSW WEA and to LCNR to avoid direct disturbance by people and dogs using these locations. Compensation: Enhancement of habitats at NSW WEA and LCNR would compensate for habitat loss and residual disturbance from the development footprint.	Residual impacts at NSW of Low Severity due to loss of wetland area, but enhancement of remaining habitats would largely balance this. Furthermore, there would be a Net Ecological Gain at LCNR from habitat enhancement and removal of disturbance. Overall, the project is predicted to provide a net gain for reptile (and amphibian) species.
Impacts on odonates and butterflies	Two odonate and one butterfly species of conservation importance may experience loss of habitat. All are present in small numbers and impacts are assessed as of Low Severity.	Compensation: Enhancement of habitats at NSW WEA and LCNR would compensate for habitat loss from the development footprint.	No significant residual impacts.
Impacts on MPBWF	No MPBWF were recorded on the Development Site, but this species was found in the mangrove beside the SPR to the southwest of the site. No Direct Impacts are predicted but impact of light pollution to the nearby habitats potentially of Moderate to High Severity.	Avoidance/Minimisation: light disturbance receivers to be avoided/minimised by appropriate design of external lighting. Green wall to be formed along west of Development Site to shield MPBWF habitat. High-rise development along SPR to be single aspect with no windows to face river and area of high MPBWF density. Compensation: additional MPBWF habitat to be created in the LCNR remote from sources of light disturbance.	No significant residual impacts, and a net ecological gain as a consequence of habitat creation targeted at this species.
Fragmentation impacts on Mangrove and Brachiaria Marsh located to the southwest of NSW and the Tidal Watercourse at the south of the SPR	Fragmentation impacts as a consequence of the bridge and residential development on habitats at the extreme southwest of the Project Area were assessed as of Low Severity as the impact on linkages is low and number of species and potentially impacted would be small.	Avoidance: bridge design will avoid impacts on the most important linkage to these habitats/areas.	No significant residual impacts.
Indirect Impacts to Deep Bay Intertidal areas
Sedimentation	No works will be undertaken in river channels; however, works in the Project Site could affect intertidal areas if silty run-off to channels takes place. As the channels already carry a significant silt load the impact of additional silt load is likely to be of Low Severity.	Site practice during construction will ensure that no silty run-off enters watercourses.	No Residual Impacts.
Pollution Impacts on Wetland Habitats	Certain pollution events (e.g. oil spills) have the potential for a significant ecological impact, depending upon the nature and scale of the pollution event. Potentially of Low to Moderate Severity during the construction phase (especially during bridge construction) and of Low Severity during the operational phase.	Avoidance: measures taken to avoid site runoff, including onsite sedimentation facility for treatment and onsite reuse. Stringent control of construction activities to avoid any accidental discharges.	No significant residual impacts. Potential ecological benefit as site management will ensure that pollution caused by such events as dumping of rubbish will be prevented.
Mortality Impacts on Fauna	Low to Moderate Severity during construction phase in absence of mitigation measures. All non-vagile fauna is potentially vulnerable to mortality during site clearance; terrestrial fauna during construction; and birds (due to collision) during operation.	Avoidance: direct mortality of roosts/nests to be avoided by checking potential roost sites prior to site clearance (including buildings) and rescheduling if necessary. Solid barriers to be placed around development works areas.	No significant residual impacts.

4.10 Conclusions

4.10.1 This ecological impact assessment was conducted in accordance with the requirements of the EIA Study Brief and the guidelines of the EIAO-TM, in particular Annexes 8 and 16. It addresses how the planning and conservation objectives for the Project Site will be achieved, and how as well as fully addressing the requirement to mitigate for all significant adverse ecological impacts of the Project it will deliver a considerable positive contribution to the conservation of the internationally important Inner Deep Bay wetland ecosystem.

4.10.2 Taking as a starting point the previous conservation and development proposals for the Project Area, the constraints posed by existing ecologically important habitats and species were given careful consideration in the planning of the Project. In addition, the existing conservation-related zonings, including the WCA, OU(CDWEA1), and SSSI zone were carefully reviewed to ensure that both the Development Site layout and the habitats to be conserved, restored, enhanced and created upheld the principles of the precautionary approach, no-net-loss in wetland, and avoidance of habitat fragmentation. In particular, the Master Layout Plan (MLP) for the development was adjusted iteratively to avoid and minimise ecological impacts; of importance in this regard was the requirement to avoid all significant impacts on the Great Cormorant roost trees at NSW and Great Cormorant and on ardeid flightlines.

4.10.3 A further important consideration was the need to address the considerable environmental degradation that was ongoing at Lut Chau (LC), not least the continuing destruction of mangroves within the Core Area of the Ramsar Site, and to develop proposals for habitat restoration, enhancement and creation there. Implementation of such proposals would allow LC to fulfil its ecological potential and, in effect, form a complementary extension to the adjacent MPNR.

4.10.4 The Development Site footprint of the project was restricted to 11.6ha, only 6.5% of the Project Site, and there will be no-net-loss of area of reedbed, mangrove or intertidal areas. The net loss of wetland area will be largely restricted to that of fishpond, a habitat in which the enhancement of ecological function will compensate for loss of area. The loss of this area requires a readily achievable increase in wetland function of the remainder of the fishpond area and the newly created area of shallow tidal pond habitat. Some loss of open water, Brachiaria marsh and Typha marsh habitats will also be a consequence, but loss of these areas is considered to be of minor impact severity given the low ecological value of Brachiaria marsh and especially Typha marsh (a habitat dominated by an unwanted non-native plant species), and the small size of the impacted areas.

4.10.5 Disturbance to fauna of conservation importance, using wetland habitats, in particular fishponds and the tidal SPR, will be mitigated by enhancement of fishpond, reedbed and marshes habitat and the creation of reedbed, shallow tidal pond and lily pond habitat thus enabling these habitats to support a larger number of individuals of fauna of conservation importance than is currently the case. Furthermore, there will be a significant overall benefit to wildlife using the Project Site because of the active management of the NSW WEA and LCNR in combination for nature conservation, particularly in the fishpond areas. This will ensure that there is No-Net-Loss in Wetland Function as well because of the development. This complies with the intention of TPB Guideline No. 12C concerning developments in the Deep Bay area.

4.10.6 To address potential impacts on wildlife during the construction and operational phases of the project, mitigation targets have been selected to inform the design of mitigation measures. These ‘target’ are species of conservation importance, based upon criteria provided by IUCN Red List of Threatened Species (2022), List of State Key Protected Wild Animals (National Forestry and Grassland Administration and Ministry of Agriculture and Rural Affairs) and/or Fellowes et al. (2002), and which were recorded in significant numbers at NSW and LC during the baseline ecological surveys and are predicted to be potentially impacted by the proposed development. Mitigation measures will include habitat creation, enhancement and management to meet the requirements of these species and are detailed in the CMP. In total, 34 target species have been selected. Furthermore, species of high conservation importance not recorded during the baseline ecological surveys (and as such not requiring targeted mitigation measures), but which are expected to benefit from the implementation of the proposed mitigation measures were reviewed during preparation of the CMP for inclusion as ‘Conservation Targets’. Two species were selected: Eurasian Otter (an exceptionally rare mammal in Hong Kong) and Pheasant-tailed Jacana (a wetland-dependent bird that formally bred in Hong Kong).

4.10.7 The Great Cormorant roost and reedbed habitats at NSW, which are not currently under any form of legal protection, will benefit from protection and enhancement to protect their long-term future in a financially secure manner.

4.10.8 The NSW WEA of 99.0ha and LCNR of 55.4ha (154.4ha in total), will complement the existing MPNR to increase the area of wetland in Deep Bay which is protected and actively managed for the benefit of wildlife. The location of these sites close to the intertidal mudflats at the core of Deep Bay will provide a larger protected core within the Deep Bay ecosystem and will ensure significant ecological links between these three managed areas.

4.10.9 The following EIA Study Brief requirements in Appendix C paragraph 4(vii)(p), are addressed in the CMP, which is attached as Appendix 4.5 of this application:

(p) propose a conservation and management plan for the proposed Nam Sang Wai Wetland Enhancement Area and Lut Chau Nature Reserve in the Project Site, with particular attention to:

(i) the target species and conservation objectives of the wetland;

(ii) the proposed design, layout, measures/actions for creation, enhancement, maintenance and management of the wetland;

(iii) a practical programme and specification of resources requirement for the implementation of the CMP;

(iv) the management agents and their responsibility;

(v) a contingency plan for the management of the wetland; and

(vi) the ecological monitoring programme for monitoring and evaluating effectiveness of the actions/measures under the CMP.

4.10.10 The area of the NSW WEA and LCNR are demarcated as “Area for Potential Inclusion” in the boundary delineation of the Nam Sang Wai Wetland Conservation Park (NSW WCP), which will be an important element of the WCPs System, which in turn is a crucial component of the Northern Metropolis Development Strategy. The WCPs System is proposed to “create environmental capacity for the development of the Northern Metropolis by enhancing natural habitats and other environmental conditions” (AFCD 2024). It is considered that the ecological mitigation and enhancement measures proposed in this Ecological Impact Assessment are fully in line with the objectives of the WCPs System. Furthermore, the ecological mitigation and enhancement measures proposed here will not only compensate the loss of ecological function arising from the development but may also potentially provide additional environmental capacity for the Northern Metropolis as a whole.

4.10.11 In summary, the MLP and the proposals for habitat enhancement at NSW WEA and LCNR as described in this Ecological Impact Assessment Report and the accompanying CMP are considered to provide mitigation for all the significant potential adverse impacts arising from the construction and operation of the Project, both within the Project Site at NSW and Lut Chau and in adjacent habitats including the SPR and the KTMDC. The increased protection and enhancement proposed for the wetland enhancement areas is considered to provide increased security for wetlands in Deep Bay, preventing the risk of future degradation of these wetland habitats and allowing for future further enhancement by appropriate wetland management.

4.11 References

AEC. 2010. Wetland Compensation Area Ecological Monitoring and Adaptive Management Advice. Annual Report, February 2010. aec Ltd.

AEC. 2016. Ecological Impact Assessment for Proposed Comprehensive Development with Wetland Enhancement at Nam Sang Wai and Lut Chau. aec Ltd.

AEC. 2020. Current population, distribution and habitat preferences of Mai Po Bent-winged Firefly Pteroptyx maipo in the Inner Deep Bay area of Hong Kong. Prepared by AEC as part of a Further Supplementary Witness Statement submitted to the Town Planning Appeal Board in 2020.

AFCD. 2024a. Hong Kong Biodiversity Information Hub. https://bih.gov.hk/en/home/index.html

AFCD. 2024b. Strategic Feasibility Study on the Development of Wetland Conservation Parks System Under the Northern Metropolis Development Strategy. Study Report October 2024. Downloaded from www.afcd.gov.hk/english/conservation/con_wet/wcps_system/wcps_system.html on 29 October 2024

Allcock, J.A., Leader, P.J., Leven, M.R., Stanton, D.J. and Leung, K. 2013. Seasonality of Acrocephalus and Locustella warblers in the reedbeds at Mai Po Nature Reserve. In Allcock, J.A., Carey, G.J., Chow, G. & Welch, G. (Eds) The Hong Kong Bird Report 2011. The Hong Kong Bird Watching Society, Hong Kong.

Allcock, J.A., Leader, P.J., Stanton, D.J., Leven, M.R. & Leung, K.K.S 2018. Permanently inundated Phragmites reedbed supports higher abundance of wetland-dependent bird species than drier reedbed during southward migration through Hong Kong. Forktail 34: 9-13.

Anon. 2007. Summer 2007 Report: Egretry Counts in Hong Kong with particular reference to the Mai Po Inner Deep Bay Ramsar Site. Report by Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon, 2008a. Summer 2008 Report: Egretry Counts in Hong Kong with particular reference to the Mai Po Inner Deep Bay Ramsar Site. Report by Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon. 2008b. Winter 2007-08 Report on Waterbird Monitoring at the Mai Po Inner Deep Bay Ramsar Site. Report by Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon. 2009a. Summer 2009 Report: Egretry Counts in Hong Kong with particular reference to the Mai Po Inner Deep Bay Ramsar Site. Report by Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon. 2009b. Winter 2008-09 Report on Waterbird Monitoring at the Mai Po Inner Deep Bay Ramsar Site. Report by Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon, 2010a. Summer 2010 Report: Egretry Counts in Hong Kong with particular reference to the Mai Po Inner Deep Bay Ramsar Site. Report by Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon. 2010b. Winter 2009-10 Report on Waterbird Monitoring at the Mai Po Inner Deep Bay Ramsar Site. Report by Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon. 2011. Winter 2010-11 Report on Waterbird Monitoring at the Mai Po Inner Deep Bay Ramsar Site. Report by Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon. 2012a. Summer 2011 Report: Egretry Counts in Hong Kong with particular reference to the Mai Po Inner Deep Bat Ramsar Site. Report by the Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon. 2012b. Summer 2012 Report: Egretry Counts in Hong Kong with particular reference to the Mai Po Inner Deep Bat Ramsar Site. Report by the Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon. 2012c. Summer 2012 Report on Waterbird Monitoring at the Mai Po Inner Deep Bay Ramsar Site. Report by Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon. 2012d. Winter 2011-12 Report on Waterbird Monitoring at the Mai Po Inner Deep Bay Ramsar Site. Report by Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon. 2013a. Monthly Waterbird Monitoring Biannual Report 1 (April to September 2013), Mai Po Inner Deep Bay Ramsar Site Waterbird Monitoring Programme 2013-14. Report by the Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon. 2013b. Summer 2013 Report: Egretry Counts in Hong Kong with particular reference to the Mai Po Inner Deep Bat Ramsar Site. Report by the Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon. 2013c. Winter 2012-13 Report on Waterbird Monitoring at the Mai Po Inner Deep Bay Ramsar Site. Report by Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon. 2014a. Monthly Waterbird Monitoring Biannual Report 1 (April to September 2014), Mai Po Inner Deep Bay Ramsar Site Waterbird Monitoring Programme 2014-15. Report by the Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon. 2014b. Summer 2014 Report: Egretry Counts in Hong Kong with particular reference to the Mai Po Inner Deep Bat Ramsar Site. Report by the Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon. 2014c. Winter 2013-14 Report on Waterbird Monitoring at the Mai Po Inner Deep Bay Ramsar Site. Report by Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon. 2015a. Mai Po Inner Deep Bay Ramsar Site Waterbird Monitoring Programme 2014 – 15. Monthly Waterbird Monitoring. Biannual Report 2, October 2014 to March 2015. Report by Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon, 2015b. Summer 2015 Report: Egretry Counts in Hong Kong with particular reference to the Mai Po Inner Deep Bay Ramsar Site. Report by Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon, 2016. Summer 2016 Report: Egretry Counts in Hong Kong with particular reference to the Mai Po Inner Deep Bay Ramsar Site. Report by Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon, 2017. Summer 2017 Report: Egretry Counts in Hong Kong with particular reference to the Mai Po Inner Deep Bay Ramsar Site. Report by Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon, 2018. Summer 2018 Report: Egretry Counts in Hong Kong with particular reference to the Mai Po Inner Deep Bay Ramsar Site. Report by Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon, 2020. Summer 2019 Report: Egretry Counts in Hong Kong with particular reference to the Mai Po Inner Deep Bay Ramsar Site. Report by Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon, 2021a. Summer 2020 Report: Egretry Counts in Hong Kong with particular reference to the Mai Po Inner Deep Bay Ramsar Site. Report by Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon, 2021b. Summer 2021 Report: Egretry Counts in Hong Kong with particular reference to the Mai Po Inner Deep Bay Ramsar Site. Report by Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Anon, 2022a. Summer 2022 Report: Egretry Counts in Hong Kong with particular reference to the Mai Po Inner Deep Bay Ramsar Site. Report by Hong Kong Bird Watching Society to the Agriculture, Fisheries and Conservation Department, Hong Kong Special Administrative Region Government.

Aspinwall 1997. Study on the Ecological Value of Fishponds in the Deep Bay Area. Agreement No. CE72/94. Final Report, Hong Kong. Report by Aspinwall & Co. Hong Kong Ltd. In association with Wetlands International to Planning Department, Hong Kong SAR Government.

Ballantyne, L., Fu, X.H., Shin, Chun-hat, Cheng, Chui-Yu and Yiu Vor. 2011. Pteroptyx maipo Ballantyne, a new species of bent-winged firefly (Coleoptera: Lampyridae) from Hong Kong, and its relevance to firefly biology and conservation. Zootaxa 2931:8-34

Binnie, Black and Veatch. 2002. Sheung Shui to Lok Ma Chau Spur Line Environmental Impact Assessment: Environmental Impact Assessment Report. Binnie Black and Veatch Hong Kong Ltd., Hong Kong.

BirdLife International. 2004. Important Bird Area factsheet: Inner Deep Bay and Shenzhen River catchment area. Downloaded from https://datazone.birdlife.org/site/factsheet/inner-deep-bay-and-shenzhen-river-catchment-area-iba-hong-kong-(china)

BirdLife International. 2017a. Emberiza aureola (amended version of 2016 assessment). The IUCN Red List of Threatened Species 2017: e.T22720966A110690385. https://dx.doi.org/10.2305/IUCN.UK.2017-1.RLTS.T22720966A110690385.en.

BirdLife International. 2017b. Platalea minor. The IUCN Red List of Threatened Species 2017: e.T22697568A119347801. https://dx.doi.org/10.2305/IUCN.UK.2017-3.RLTS.T22697568A119347801.en. Accessed on 14 April 2024.

BirdLife International (2024) Species factsheet: Collared Crow Corvus pectoralis. Downloaded from https://datazone.birdlife.org/species/factsheet/collared-crow-corvus-pectoralis on 14/10/2024.

Carey, G.J., Chalmers, M.L., Diskin, D.A., Kennerley, P.R., Leader, P.J., Lewthwaite, R.W., Leven, M.R., Melville, D.S., Turnbull, M. and Young, L. 2001.The Avifauna of Hong Kong. Hong Kong Bird Watching Society, Hong Kong.

CH2M Hill. 2006. Residential, Golf Course and Nature Reserve Development (Sunnyville Estate Development) at Nam Sang Wai, Yuen Long. Environmental Assessment Study Volume II (Ecology). CH2M Hill Hong Kong Ltd., Hong Kong.

CH2M Hill. 2008. Proposed Development at Fung Lok Wai, Yuen Long. Lot 1457R.P. in D.D.123. Environmental Assessment Study. CH2M Hill Hong Kong Ltd., Hong Kong.

Chan, A., J. Cheung, P. Sze, A. Wong, E. Wong and E. Yau, 2011.A Review of the Local Restrictedness of Hong Kong Butterflies. Hong Kong Biodiversity 21: 1-12

Cheng, C.Y., Shih, C.H. and Cheung, K.H. 2010. First Record in China of the Firefly genus Pteroptyx (齊爍螢屬). Hong Kong Biodiversity 19: 1-3

Chernetsov, N. 1998. Habitat distribution during the post breeding and post-fledging periods in the Reed Warbler Acrocephalus scirpaceus and Sedge Warbler A. schoenobaenus depends on food abundance. Ornis Svecica 8:77-82.

DSD, 2019. Yuen Long Effluent Polishing Plant. Approved Environment Impact Assessment Report (AEIAR-220/2019)

DSD, 2021. Yuen Long Barrage Scheme. Approved Environmental Impact Assessment Report (AEIAR-228/2021)

Fan, Y. and Fu, X. (2017). The complete mitochondrial genome of the firefly, Pteroptyx maipo (Coleoptera: Lampyridae). Mitochondrial DNA Part B 2(2): 795-796.

Fellowes, J. R., Lau, M. W. N., Dudgeon, D., Reels, G. T., Ades, G. W. J., Carey, G. J., Chan, B. P. L., Kendrick, R. C., Lee, K. S., Leven, M. R., Wilson, K. D. P. and Yu, Y. T. 2002. Wild animals to watch: terrestrial and freshwater fauna of conservation concern in Hong Kong. Memoirs of the Hong Kong Natural History Society, 25, 123-159.

Fu, X.H. (2014). An Ecological Pictorial Guide to Fireflies of China. The Commercial Press. 172 pp. (in Chinese).

Hawke, C. J. and José P. V. 1996. Reedbed Management for Commercial and Wildlife Interests. RSBP, The Lodge, Sandy, U.K.

HKBWS. 2021. Hong Kong Bird Report 2018. Hong Kong Bird Watching Society.

IAC-Acoustics. 2016. Comparative Examples of Noise Levels. http://www.industrialnoisecontrol.com/comparative-noise-examples.htm.

IUCN 2024. IUCN Red List of Threatened Species. https://www.iucnredlist.org.

Jiang, Z.G., Jiang, J.P., Wang, Y.Z., Zhang, E., Zhang, Y.Y., Li, L.L., Xie, F., Cai, B., Cao, L., Zheng, G.M., Dong, L., Zhang, Z.W., Ding, P., Luo, Z.H., Ding, C.Q., Ma, Z.J., Tang, S.H., Cao, W.X., Li, C.W., Hu, H.J., Ma, Y., Wu, Y., Wang, Y.X., Zhou, K.Y., Liu, S.Y., Chen, Y.Y., Li, J.T., Feng, Z.J., Wang, Y., Wang, B., Li, C., Song, X.L., Cai, L., Zang, C.X., Zeng, Y., Meng, Z.B., Fang, H.X., and Ping, X.G., 2016. Red List of China’s Vertebrates (RLCV). Biodiversity Science,24 (5), 500-551.

Karsen, S., M.W.N. Lau & A. Bogadek. 1998. HK Amphibians and Reptiles. Provisional Urban Council, HK

Kruuk, H. 1995. Wild Otters. Predation and Populations. Oxford University Press Inc., New York.

Lai, P.C.C., Lee, W.H. and Wong, B.S.F. 2007. Construction of a drainage channel at Inner Deep Bay, Hong Kong: mitigating ecological and landscape impact through an environmentally friendly design. Landscape Ecol. Eng. 3: 179-185.

Lo, P. Y. F. and Hui, W.L. 2010. Hong Kong Butterflies. Cosmos Books Ltd., Hong Kong.

Loy, A., Kranz, A., Oleynikov, A., Roos, A., Savage, M. & Duplaix, N. 2022. Lutra lutra (amended version of 2021 assessment). The IUCN Red List of Threatened Species 2022: e.T12419A218069689. https://dx.doi.org/10.2305/IUCN.UK.2022-2.RLTS.T12419A218069689.en.

Ma, C.K.W. 2014. Post-release monitoring of the northward migration of a Great Cormorant Phalacrocorax carbo sinensis from its wintering site in Hong Kong. Hong Kong Bird Report 2012: 255-277

McMillan, S. E., Wong, T-C., Hau, B. C. H., and Bonebrake, T. C. 2019. Fish farmers highlight opportunities and warnings for urban carnivore conservation. Conservation Science and Practice. 2019;1:e79. wileyonlinelibrary.com/journal/csp2 1 of 12 https://doi.org/10.1111/csp2.79

Minutes of 1051^st Meeting of the Town Planning Board.

MTR Corporation. 2013. MTRC Contract No. M1016-09C Ecological Monitoring and Adaptive Management Advice Services for Lok Ma Chau and West Rail Wetlands. Lok Ma Chau Annual Report – 2012.

MTR Corporation. 2014. MTRC Contract No. M1016-09C Ecological Monitoring and Adaptive Management Advice Services for Lok Ma Chau and West Rail Wetlands. Lok Ma Chau Annual Report – 2013.

MTR Corporation. 2015. MTRC Contract No. M1070-13C Ecological Monitoring and Adaptive Management Advice Services for Lok Ma Chau and West Rail Wetlands. Lok Ma Chau Annual Report – 2014.

O’Connell, T. 2001. Avian window strike mortality at a suburban office park. The Raven 72: 141-149.

Ogden, L. J. E. 1996. Collision course: The hazards of lighted structures and windows to migrating birds. Toronto, Ontario: World Wildlife Fund Canada and the Fatal Light Awareness Program.

Peach, W.J., Buckland, S.T., Baillie, S.R. 1996. The use of constant effort mist-netting to measure between-year changes in the abundance and productivity of common passerines, Bird Study, 43:2, 142-156.

Poulin, B., Lefebvre, G and Mauchamp, A. 2002. Habitat requirements of passerines and reedbed management in southern France. Biological Conservation 107: 315-374.

Roos, A., Loy, A., Savage, M. & Kranz, A. 2021. Lutra lutra. The IUCN Red List of Threatened Species 2021: e.T12419A164578163. https://dx.doi.org/10.2305/IUCN.UK.2021-3.RLTS.T12419A164578163.en.

Shek, C.-t. 2006. A Field Guide to the Terrestrial Mammals of Hong Kong. Friends of the Country Parks, Cosmos Books, Hong Kong.

SMEC. 2021. Mai Po Nature Reserve Infrastructure Upgrade Project EIA Report, prepared for WWF Hong Kong.

Stanton, D.S., Smith, B.R. and Leung, K.K.S. 2014. Status and roosting characteristics of Collared Crow Corvus torquatus at the Mai Po Nature Reserve, Hong Kong. Forktail 30: 79-83.

Tam, T.W., Leung, K.K., Kwan, B.S.P., Wu, K.Y., Tang, S.S.H.So, I.W.Y, Cheng, J.C.Y., Yuen, E.F.M., Tsang, Y., Hui, W. 2011. The Dragonflies of Hong Kong. Cosmos Books Ltd., Hong Kong.

Town Planning Appeal Board (TPAB). 2021. Town Planning Appeal No. 1 of 2017 (1/2017)

https://www.tpab.gov.hk/filemanager/en/content_47/Decision%20TPA%20No.1%20of%202017.pdf

Tscharntke, T. 1992. Fragmentation of Phragmites habitats, minimum viable population size, habitat suitability, and local extinction of moths, midges, flies, aphids, and birds. Conservation Biology 6: 530-536.

Van der Grift, E. A. and Kuijster, R. W. J. 1998. Mitigation measures to reduce habitat fragmentation by railway lines in the Netherlands. ICOWET February 9 – 12: 166 – 170.

Wilson, K.D.P. & Reels, G. 2011. Mortonagrion hirosei. The IUCN Red List of Threatened Species 2011: e.T13891A4362234. https://dx.doi.org/10.2305/IUCN.UK.2011-1.RLTS.T13891A4362234.en.

WWF Hong Kong. 2006. Management Plan for the Mai Po Nature Reserve. WWF Hong Kong, Hong Kong.

WWF Hong Kong and HKBRG. 2008. Study into the Avian Value of Different Aged Stands of Phragmites australis at Mai Po Nature Reserve. Joint report by the World Wide Fund for Nature Hong Kong and Hong Kong Bird Ringing Group.

WWF Hong Kong. 2013. Mai Po Nature Reserve Habitat Management, Monitoring and Research Plan 2013-2018. Volume I Habitat Management. WWF Hong Kong, Hong Kong.

WWF Hong Kong. 2021. Mai Po Nature Reserve Management Plan: 2019-2024. WWF Hong Kong.

Yip, A. & YIU, V. 2023. Pteroptyx maipo. The IUCN Red List of Threatened Species 2023: e.T214613742A214613992. https://dx.doi.org/10.2305/IUCN.UK.2023-1.RLTS.T214613742A214613992.en.

Yiu, V. 2011. A new species of firefly from Hong Kong- Pteroptyx maipo Ballantyne, 2011米埔屈翅螢. Insect News (Hong Kong Entomological Society Newsletter)3: 2-6.

Yiu, V. 2012. Effect of artificial light on firefly flashing activity. Insect News (Hong Kong Entomological Society Newsletter) 4: 5-9.