4. ECOLOGICAL IMPACT ASSESSMENT

4.1 Introduction

4.1.1 The aim of this Ecological Impact Assessment Report (EcoIA) with respect to ecological impacts of the construction and operation of the Spur Line is to:

• Review the ecological resources within the area affected by the project and identify Valuable Ecological Components (VECs), including habitats and Species of Conservation Importance.

• Identify the potential impacts from the construction and operation of the Spur Line.

• Identify potential mitigation measures that may be used to reduce or overcome the potential impacts of the construction and operation of the Spur Line.

• Provide an assessment of the predicted residual impacts of the construction and operation of the Spur Line taking into account the proposed mitigation measures.

4.1.2 This assessment is based on guidelines given in Annex 8 (“Criteria for evaluating an ecological impact”) and Annex 16 (“Guidelines for ecological assessment”) of the Technical Memorandum on Environmental Impact Assessment Process (EPD).

4.2 Methodology

4.2.1 This EcoIA is primarily based on information collated during the production of a Preliminary Project Feasibility Study (PPFS) in 1998 (Binnie 1998) of the proposed Sheung Shui to Lok Ma Chau Spur Line and a recent detailed ecological review of the Spur Line corridor (ERM 1999b).

4.2.2 Following the requirements of Annex 16 of the Technical Memorandum to the EIA Ordinance, the defined Study Areas covered by the PFFS and ERM studies consisted of a one-kilometre corridor (500 m on either side of the proposed alignment). Ecological investigations were carried out by a literature review and a field survey to determine the types and extent of habitats present. A narrower corridor represented by 100 m either side of the alignment was subjected to more detailed survey to establish the importance of these habitats in the context of the study corridor; this 200 m corridor was considered to represent the area that is most likely to be subject to impacts from the construction and operation of the railway.

4.2.3 A preliminary habitat map was produced for the PPFS (Binnie 1998) which was then subsequently updated by the ERM 1999b study. In the ERM study, a 1:5,000 habitat map was prepared by making reference to the latest available aerial photographs taken by the Lands Department, HKSAR (see ERM 1999b Annex A reference list), during which the boundaries of different habitat types within the Study Area were mapped and delineated. Complementary reconnaissance ground-truthing were subsequently undertaken between March 1999 and May 1999 to field check and confirm the nature/latest status of those habitats identified.

4.2.4 It was noted, during the field surveys undertaken as part of the ERM study, (ERM 1999b), that construction activities associated with a range of projects (including the Main Drainage Channel for River Beas and River Sutlej and the kiosk expansion at Lok Ma Chau Boundary Crossing), as well as the management practices of the land owner or tenant of agricultural land, continue to alter the status of the habitats within the Study Area.

4.2.5 The Binnie PPFS and ERM ecological studies also reviewed a number of previous studies carried out in this geographical area, including the following documents:

• Sheung Shui and Fanling Main Drainage Channels EIA (CES 1997);

• West Rail Phases 1 and 2 EIAs (ERM 1998a);

• Sheung Shui to Lok Ma Chau Spur Line Preliminary Project Feasibility Study (Binnie 1998);

• San Tin Drainage Channel EIA (ERM 1999a);

• Lok Ma Chau Boundary Crossing Expansion EIA (Binnie 1999);

• Comprehensive Conservation Strategy and Management Plan for the Deep Bay Ramsar Site; (Aspinwall Clouston and Wetlands International - Asia Pacific (1997);

• various fishpond studies, e.g. Aspinwall & Company Hong Kong Limited (1997), Young (1997 & 1998) and Zhuang and Lau (1997); and

• other sources of information (such as the Annual Reports of the Hong Kong Bird Watching Society).

4.2.6 The ecological importance of habitats with respect to the occurrence of butterflies, dragonflies, fish, amphibians, reptiles, birds and mammals was investigated in the ERM study by a review of existing literature sources as these taxa groups have been relatively well studied in Hong Kong. The identification of Species of Conservation Importance (in accordance with the EPD Technical Memorandum) was therefore restricted in this EIA report to these groups.

4.2.7 The invertebrate groups, Lepidoptera (butterflies and moths) and Odonata (dragonflies and moths) have been relatively well studied in Hong Kong and there are a number of published and unpublished records which provide an indication of their habitat requirements and local status. Unfortunately, other invertebrate groups have been less well studied and, as a consequence, the data required to assess the ecological significance of individually observed specimens within the Study Area (i.e. species lists, spatial distribution, habitat requirements, relative abundance, etc.) are not available.

4.2.8 Therefore, in order to adequately assess the significance of the Study Area for invertebrates, the ERM (1999b) study used Lepidoptera and Odonata as representative invertebrate species and as indicators of the capacity of identified habitats to support invertebrate communities. Whilst the habitat requirements of the two groups would not reflect the specific needs of all invertebrate groups, the presence of the range of habitats required to support their larval and adult stages and overall species diversity is likely to reflect to some degree the relative 'health' of the populations of other, less well studied, groups. The use of butterflies as sensitive indicators of habitat diversity and of the impacts of anthropogenic changes to habitats is supported by Erhardt (1985), whilst Wilson (1995) cites dragonfly larvae as good indicators of both water quality and habitat diversity.

4.2.9 In the Section that follows the findings of a review of relevant field survey reports for Lepidoptera and Odonata are presented and the species of ecological or conservation value are highlighted.

4.2.10 The direct loss of habitats of ecological importance from the Spur Line comprise the land destruction due to column construction and the construction of the station complex at Lok Ma Chau. This was measured by overlaying maps of the alignment onto the current ERM habitat map for 1999. Habitat area calculations were made using GIS software. Non-ecologically valuable habitats (such as urban/residential areas and container storage areas) and areas where works are being carried out under other projects (such as Fanling and Sheung Shui Main Drainage Channels (CES 1997); San Tin Drainage Channels (ERM 1999a) and Lok Ma Chau Border Crossing Expansion (Binnie 1999)) were removed from the area evaluation equation.

4.3 Baseline Identification and Evaluation of Valuable Ecological Components (VECS)

4.3.1 The first objective of an Ecological Impact Assessment (EcIA) should be to review the existing ecological resources and receivers that may be affected by the development and to identify those that are Valuable Ecological Components (VECs) (Treweek 1999). This has therefore been carried out, using the information sources described above, for habitats and then for individual Species of Conservation Importance following the guidelines for the evaluation of ecological importance of habitats and species given in the Technical Memorandum on Environmental Impact Assessment Process (EPD).

Habitats

4.3.2 With respect to the vegetation characteristics and other physical and environmental attributes, 15 different habitat types were categorised by the ERM (1999b study), as follows:

• Fung-shui Forest - forest found around rural villages comprising native species or a mix of native species and fruit trees;

• Lowland Secondary Forest - forest dominated by native tree species and not located behind villages;

• Plantation Forest - landscaping or ornamental plantations mainly found along roads and on cut‑slopes;

• Orchard - patches of fruit trees mainly found scattered among rural village areas;

• Wet Agricultural Land - permanently or seasonally inundated agricultural land on which a range of food crops are cultivated;

• Dry Agricultural Land - agricultural land on which a range of food crops and flowers are cultivated;

• Inactive Agricultural Land - fallow or abandoned agricultural land which is un‑managed and has been invaded by herbaceous or shrubby vegetation;

• Fishponds ‑ fishponds managed for commercial aqua-culture production;

• Inactive Fishpond - fishpond where no management facilities were observed in the immediate vicinity of the pond and edge of the pond bund was found covered by emergent vegetation;

• Pond - water bodies rearing bloodworms for fish feeding;

• Marsh - flooded areas comprising a mix of water‑dependent plants;

• Water-courses - include all riverine habitats such as rivers, drainage channels, nullahs and ditches;

• Grassland-shrubland mosaic - undisturbed areas with grassland comprising 80% of cover;

• Wasteland - secondary vegetative colonisation (mainly grasses and ruderals) on areas opened by human activities (excluding those grown on inactive agricultural land); and

• Developed area ‑ areas highly developed with no visible vegetation cover except road-side weeds.

4.3.3 The distribution of these 15 habitat types within the Study Area is presented in Figure 4.1 and their respective coverage within the 200 m and 1 km corridor are shown in Table 4.1 below. This indicates that, excluding developed areas, fishponds represent the most common habitat type within the 1 km study corridor, followed by grassland-shrubland mosaics and inactive agricultural land. Fishponds also represent the most common habitat within the 200 m corridor, followed by inactive agricultural land and wet agricultural land.

Table 4.1

Coverage of Different Habitat Types within the Study Corridor

Habitat	Coverage within 500 m Zone		Coverage within 100 m Zone
	ha	%	ha	%
Fung-shui Forest	17.81	2.29	0.57	0.39
Lowland Secondary Forest	15.11	1.94	2.58	1.77
Plantation Forest	32.89	4.22	5.55	3.81
Orchard	10.91	1.40	5.43	3.73
Wet Agricultural Land	28.31	3.63	7.52	5.16
Dry Agricultural Land	13.02	1.67	3.83	2.62
Inactive Agricultural Land	38.27	4.91	11.84	8.12
Fishpond	73.74	9.45	21.04	14.39
Inactive Fishpond	12.04	1.55	2.93	2.01
Pond	2.31	0.30	0.61	0.42
Marsh	5.17	0.66	2.30	1.58
Grassland-shrubland mosaic	54.27	6.96	3.10	2.13
Water-courses	8.05	1.03	2.18	1.50
Wasteland	35.52	4.56	5.33	3.66
Developed Area	432.03	55.43	70.95	48.67
TOTAL	779.46	100	145.77	100

Note: Due to rounding up, the percentage total may not equal 100.

4.3.4 A description of the key habitat areas within the Study Area is provided in Appendix F.

Sites of conservation importance

4.3.5 The location of the proposed route of the railway in relation to existing protected areas of high ecological value is indicated in Figure 4.2. This indicates that the only protected area near to the study area is the Inner Deep Bay Ramsar site (a Wetland of International Importance under the Ramsar Convention) and SSSI. The Deep Bay area comprises natural and man-made wetlands (rivers, freshwater marshes, fish ponds, gei wais (tidal shrimp ponds), mangal, inter-tidal mudflats and the bay) which provide a wide range of habitats that support a high diversity of biota (plants, insects, reptiles, amphibians, birds and mammals).

4.3.6 The Ramsar site is particularly important for waterbirds, including a number of globally threatened species (Collar et al. 1994), species which regularly have high proportions of their global or biogeographical wintering or passage populations within the site and species that are of regional or local conservation importance (Aspinwall Clouston & Wetlands International - Asia Pacific, 1997). Many of these species have been recorded within the study area, or are likely to occur (Table 4.2).

4.3.7 The fish ponds in the San Tin / Lok Ma Chau area are close to and contiguous with the Deep Bay Ramsar site. Formerly included in the Deep Bay Buffer Zones, the importance of these fish ponds was demonstrated in the recent Fish Pond Study (Aspinwall & Co. 1999) and, following the completion of this study, they have been included in the redelineated Wetland Conservation Area (WCA) and Wetland Buffer Area (WBA) by the Town Planning Board (TPB PG-No 12B) (Figure 4.2).

4.3.8 The Town Planning Board has adopted a “precautionary approach” to development in the Deep Bay area in view of the known intrinsic value of fish ponds in ecological terms, and the complex response of birds to future landuse changes which has not been fully understood. The intention is to protect and conserve the existing ecological function of fish ponds in order to maintain the ecological integrity of the Deep Bay wetland ecosystem as a whole. This “precautionary approach” is formulated with the support of scientific surveys and analysis as provided in the Study on the Ecological Value of Fish Ponds in the Deep Bay Area.

4.3.9 In considering development proposals in the Deep Bay Area, the Board adopts the Fish Pond Study’s recommended principle of “no-net-loss in wetland” which provides for the conservation of continuous and adjoining fish ponds. The no-net-loss can refer to both loss in “area” and “function”.

4.3.10 The planning intention of the WCA is to conserve the ecological value of the fish ponds which form an integral part of the wetland system ecosystem in the Deep Bay Area. It is noted that TPB PG-No 12B states that new development within the WCA should not be allowed unless it is required to support the conservation of the ecological value of the area or if the development is an essential infrastructural project with overriding public interest. In the latter case, an ecological assessment for any such development is required to demonstrate that it will not result in a net loss of wetland function or negative disturbance impacts. Wetland compensation is required for any development requiring pond filling and mitigation measures against disturbance are also required.

4.3.11 It is recognised that the Spur Line, and especially the Lok Ma Chau Station, is the first major intrusion into the Wetland Conservation Area which forms the core wetland area of Inner Deep Bay between Tin Shui Wai in the west and the Lok Ma Chau Boundary Crossing in the east. Therefore, the long-term impacts to ecology and a practical ecological scheme have been considered at the outset. However, it is not intended to set a precedent as ecological impacts and mitigation should consider the particular situation of individual projects. The Spur Line also encroaches into the Wetland Buffer Area (WBA) which requires an ecological assessment to ensure that negative impacts can be mitigated.

Species of conservation importance

4.3.12 Following guidance given in Annex 16 (Note 3) of the EPD's Technical Memorandum on Environmental Impact Assessment, species have been identified that are considered to be Species of Conservation Importance. These are defined as species that regularly occur in the Study Area in significant numbers and are:

• Globally threatened (Groombridge 1993; Collar et al. 1994) or threatened in South China (i.e. Collar et al. 1994)

• Internationally Protected Species, i.e. protected by international conventions: (1) The Convention on the Conservation of Migratory Species of Wild Animals (the Bonn Convention*), (2) listed under an agreement between the Government of Australia and the Government of the People’s Republic of China for the Protection of Migratory Birds and their Environment, (3) listed under an agreement on the Protection of Migratory Birds and their Habitats between the Government of Japan and the Government of the People’s Republic of China.

Note that as all migratory birds are listed on Annex 1 of the Bonn Convention only those that are also occur in internationally or regionally important numbers (see below) are considered to be of Species of Conservation Importance.

• Protected by Chinese laws and regulations: (1) The Wildlife Protection Law of the PRC, (2) The Wildlife Protection Implementation Regulation of the PRC, (3) The Guideline for Nature Reserves and Wildlife Species in the PRC, (4) The Nature Reserve Regulations (Provision 32), (5) The National Protection List of Important Wild Animals, (6) The Management Measures for Forests of Guangdong Province, (7) Guangdong Provincial Implementary Detailed Regulations for Nature Reserves of Forest and Wildlife Species (Section 12).

• Protected by Hong Kong legislation and guidelines: (1) The Forests and Countryside Ordinance (Cap 96), (2) The Forestry Regulations (subsidiary legislation of Cap 96), (3) The Wild Animals Protection Ordinance (Cap 170), (4) The Animals and Plants (Protection of Endangered Species) Ordinance (Cap 187), (5) The Fisheries Protection Ordinance (Cap 171), (6) The Fisheries Protection Regulations (subsidiary legislation of Cap 171).

Note: All birds and bats are listed in the Wild Animals Protection Ordinance primarily as protection from hunting, persecution and trade etc. Therefore, only those that meet other criteria listed here or are Locally Important Species according to the ERM (1999b) criteria listed below are considered here to be Species of Conservation Importance.

• Endemic to Hong Kong or south China; if they are rare in the territory or have special conservation importance according to other scientific studies, e.g. for birds if they are of international importance (i.e. >1% of the biogeographical population) or regional importance (as listed in the Deep Bay Ramsar Conservation Strategy (Aspinwall Clouston & Wetlands International - Asia Pacific, 1997).

Species that occur within the Study Area in significant numbers according to the ERM study (1999b) that are also considered to be Species of Conservation Importance are listed in Table 4.2. Birds have also been divided into the following five categories of Species of Conservation Importance

• Globally threatened species

• Regionally Important Species

• Restricted range species in Hong Kong, i.e. species (including non-breeding species) which are found at fewer than three sites in Hong Kong.

• Declining species in Hong Kong, i.e. species (including non-breeding and locally extinct species) for which there is certain or probable long-term decline in Hong Kong.

• Locally important species that occur regularly in the Study Area.

4.3.13 The assessment of the importance of populations is based on Rose and Scott's (1997) estimates of waterfowl populations world-wide and the 1% population levels for identifying wetlands of international importance under the Ramsar Convention criteria. Using this 1% threshold, Carey and Young (in prep.) have identified 32 species occurring within the Mai Po Marshes and Inner Deep Bay Ramsar Site that have wintering or passage populations that meet the Ramsar criteria for global or regional importance.

4.3.14 The latter three criteria relate to species that have locally important populations and are regularly found within the Study Area. These were identified as species for which the highest annual count was within the Study Area, in at least three years for either of the two five-year periods (1988-92 and 1993-7) under review. Although these species include some that are reasonably widespread in Hong Kong they are considered important in sustaining the long-term biodiversity of the territory.

Table 4.2

Species of Conservation Importance that are known or likely to occur in significant numbers within 500 m of the proposed route between

Sheung Shui station and Lok Ma Chau station

Species	Protection status	Distribution	Rarity and threat status
Butterflies
Halpo porus	-	-	HK
Baoris farri	-	-	HK
Horaga albimacula	-	-	HK
Castalius rosimon	-	-	HK
Taraka hamada	-	-	HK
Deudorix epijarbas	-	-	HK
Vagrans egista	-	-	HK
Parnara ganga	-	-	HK
Caltoris bromus	-	-	HK
Catochrysops strabo	-	-	HK
Horaga onyx	-	-	HK
Pachliopta aristolochiae	-	-	HK
Parnara ganga	-	-	HK
Potanthus trachala	-	-	HK
Euthalia lubentina	-	-	HK
Amphibians
Chinese Bull Frog Rana rugulosa	-	-	HK
Narrow-mouthed Frog Kalophrynus interlineatus	-	R	HK
Reptiles
Chinese Water Snake Enhydris chinensis	-	-	HK
Buff-striped Keelback Amphiesma stolata	-	-	HK
Burmese Python Python molurus	HK	-	HK?
Chinese Soft-shelled Turtle Pelodiscus sinensis	HK	-	HK
Birds
Globally Threatened Species
Greater Spotted Eagle	C, HK	-	G
Imperial Eagle	C, HK	-	G
Japanese Yellow Bunting	HK	-	G
Red-billed Starling	HK	I/R?	NT
Schrenck's Bittern	HK	-	NT
Regionally Important Species
Black-winged Stilt	HK	-	-
Common Teal	HK	R	-
Eurasian Coot	HK	R	-
Great Cormorant	HK	I	-
Chinese Pond Heron	HK	I	-
Great Egret	A,J,HK	I	-
Grey Heron	HK	R	-
Little Egret	HK	I	-
Restricted Range In Hong Kong
Bonelli's Eagle		-	HK
Eurasian Eagle Owl	HK	-	HK
Greater Painted-snipe	HK	-	HK
Northern Hobby	HK	-	HK
Declining in Hong Kong
Black-naped Oriole	HK	-	HK
Pheasant-tailed Jacana	HK	-	HK
Watercock	HK	-	HK
Locally Important Species
Bluethroat	HK	L	-
Chestnut Bittern	HK	L	-
Common Snipe	HK	L	-
Common Stonechat	HK	L	-
Japanese Quail	HK	L	-
Pallas's Grasshopper Warbler	HK	L	-
Pintail Snipe	HK	L	-
Richard's Pipit	HK	L	-
Swinhoe's Snipe	HK	L	-
Zitting Cisticola	HK	L	-

Protection Status: B = listed on Appendix II of the Bonn Convention; A = listed under an agreement between the Government of Australia and the Government of the People’s Republic of China for the Protection of Migratory Birds and their Environment; J = listed under an agreement on the Protection of Migratory Birds and their Habitats between the Government of Japan and the Government of the People’s Republic of China; C = protected in the People’s Republic of China; HK = protected in the Hong Kong SAR.

Distribution: E = endemic. For birds only: I = population of international importance, i.e. more than 1% of the biogeographical population occurs, at some point in their annual cycle, in Hong Kong (in the Deep Bay area); R = Hong Kong population of regional conservation importance (i.e. south China), (Source: Aspinwall Clouston & Wetlands International - Asia Pacific, 1997); L = Study Area population of local importance.

Rarity (and threat status): G = Globally threatened species, NT = Near Threatened globally (Source: Groombridge 1993; Collar et al. 1994); C = rare and threatened in China (Birds to Watch 2: The World List of Threatened Birds); HK = threatened in Hong Kong (e.g. uncommon and declining or rare).

Dragonflies and damselflies (Odonata)

4.3.15 The ERM (1999b) report found that although wetland habitats dominated the whole Study Area, only about 20% of the recorded Odonata in Hong Kong were reported and none of them could be considered to be rare or otherwise Species of Conservation Importance. This result suggested that valuable Odonata habitats such as high quality freshwater stream or undisturbed marshy area are absence within the Study Area.

Butterflies

4.3.16 Extensive butterfly surveys have been undertaken in Long Valley, primarily by M. Bascombe along Beas River at Ho Sheung Heung and at Tsung Pak Long. The total number of species reported from these surveys were 48 from the Beas River, 73 from Ho Sheung Heung and 83 at Tsung Pak Long (Maunsell, 1997). Among the 97 species recorded as a whole from Long Valley, 5 very rare species and 10 rare species can be considered to be Species of Conservation Importance (Table 4.2) and their distribution among the three sites and the respective larval food plants are shown in Table 4.3 below.

Table 4.3

The location, larval food plant and status in Hong Kong

of butterfly Species of Conservation Importance

Species	Location	Larval Food Plant	HK Status
Halpo porus	Tsung Pak Long	Bambusa multiplex*	very rare
	Ho Sheung Heung
	Tsung Pak Long
Baoris farri	Beas River	Bambusa multiplex*	very rare
	Ho Sheung Heung
Horaga albimacula	Ho Sheung Heung	unknown	very rare
	Tsung Pak Long
Castalius rosimon	Ho Sheung Heung	Paliurus ramosissimus	very rare
	Tsung Pak Long
Taraka hamada	Tsung Pak Long	unknown	very rare
Deudorix epijarbas	Beas River	unknown	rare
	Tsung Pak Long
Vagrans egista	Beas River	Homalium cochinchinensis	rare
Parnara ganga	Ho Sheung Heung	Leersia hexandra*	rare
Caltoris bromus	Ho Sheung Heung	unknown	rare
	Tsung Pak Long
Catochrysops strabo	Ho Sheung Heung	Desmodium heterocarpon	rare
	Tsung Pak Long	Dunbaria villosa
Horaga onyx	Ho Sheung Heung	unknown	rare
	Tsung Pak Long
Pachliopta aristolochiae	Ho Sheung Heung	Aristolochia tagala(*)	rare
Parnara ganga	Ho Sheung Heung	unknown	rare
Potanthus trachala	Ho Sheung Heung	unknown	rare
Euthalia lubentina	Tsung Pak Long	Elytranthe cochinchinensis	rare
		Loranthus parasiticus

4.3.17 A number of EIA studies have reported on the butterfly fauna of the San Tin, Chau Tau and Lok Ma Chau areas and a summary of the information reported is provided in ERM (1999b). In total, 20 butterfly species were reported and although 5 of them (Common Jay Graphium doson, Graphium sarpedon, Purple Sapphire Heliophorus epicles, Junonia atlites and Common Redeye Matapa aria) are considered uncommon, none can be considered to be Species of Conservation Importance.

Fish

4.3.18 As described above, the Study Area contains extensive areas of fishponds and other open water wetland habitats including watercourses and ponds. The fishponds are mostly used for poly-culture of freshwater fish and contain high densities of various commercially stocked species (see Chapter 10 for a detailed assessment).

4.3.19 Other aquatic habitats tend to be grossly polluted, small in size, or isolated from any natural water courses of good water quality and are therefore unlikely to be able to support a diverse or abundant fish community. Consequently, the ERM (1999b) study found that only thirteen species of fish have been recorded within the Study Area and these are mainly commercially stocked species of fishpond habitats. Although none of these are considered uncommon or rare in Hong Kong (Chong & Dudgeon, 1992) and are consequently not Species of Conservation Importance, some are important prey for piscivorous birds, especially when the pond is drained as part of the management practice.

Amphibians and reptiles

4.3.20 A review of the importance of the Study Area for amphibians and reptiles was carried out by ERM (1999b) through a literature review, which also included unpublished personal records of local herpetologist, Dr Michael Lau. Although there are significant herpetofaunal records for Long Valley, as the site is frequently visited by local herpetologists, little is known about the herpetofauna in Chau Tau and Lok Ma Chau and the sparse records indicate that they have seldom been surveyed by herpetologists (Lau, pers. comm.).

4.3.21 This analysis revealed that a total of nine species of amphibian, four species of lizard, two species of snake and one terrapin have been reported from Long Valley. Most of these are common and widespread species typical of lowland areas in Hong Kong, such as the Changeable Lizard Calotes versicolor and Chinese Skink Eumeces chinensis. However, two uncommon amphibians, the Chinese Bull Frog Rana rugulosa and the Narrow-mouthed Frog Kalophrynus interlineatus are present. These two uncommon frogs breed in marshes within the Study Area and are also found in inactive or abandoned agricultural fields and wet agricultural fields. The Narrow-mouthed Frog was once considered to be rare (Lai & Ng, 1972) and was recorded from a few localities only (Romer, 1979b). However, more recent studies indicate that this species is widely distributed in the lowlands, valleys and hills of the north-western, north-eastern and central New Territories (Lau, 1998). Rana rugulosa was once considered to be the most common amphibian in Hong Kong (Boring, 1934) but has since declined considerably (Romer, 1979b; Karsen et al., 1986).

4.3.22 The major habitat in the San Tin, Lok Ma Chau and Chau Tau areas are fishponds and agricultural land. Only a few published records of amphibians and reptiles are available for these areas, probably due to the lack of field studies carried out in the area. The amphibian and reptile community in these areas appears to be less diverse than in Long Valley and the majority of recorded species are widespread. However, three species are considered to be of Conservation Importance. The Chinese Water snake, Enhydris chinensis, is localised and uncommon in Hong Kong whilst the Buff-striped Keelback Amphiesma stolata is a widespread but uncommon species because its optimum habitat (rice fields) has been entirely replaced by vegetable cultivation (Karsen et al., 1998). The agricultural land located at Lok Ma Chau is considered to be a nesting and foraging habitat and also serves as a movement corridor for this reptile (Binnie, 1999). The Burmese Python Python molurus is also listed by ERM (1999b) as occurring in the San Tin, Lok Ma Chau and Chau Tau areas and, although its status is uncertain, it is of Conservation Importance as it is protected in Hong Kong and thought to be uncommon.

4.3.23 In addition, although not specifically recorded from the Study Area, the regionally uncommon and protected Chinese Soft-shelled Turtle Pelodiscus sinensis may well occur as this is more or less restricted to fishponds (Lau 1995). Indeed, fish ponds in the Deep Bay area may currently be the stronghold for this species in Hong Kong (Aspinwall Clouston & Wetlands International - Asia Pacific, 1997).

4.3.24 Thus, in conclusion, despite the paucity of information on amphibians and reptiles it is apparent that wetland habitats along the Spur Line route are of ecological importance for these species. In particular, the wide variety of lowland wetland habitats in Long Valley supports a diverse lowland amphibian fauna including two locally uncommon frog species. Due to rapid development in lowland areas in Hong Kong, sizeable agricultural fields with a mixture of cultivation (both active and abandoned) have become increasingly scarce. Hence, the remaining habitats in Long Valley, albeit influenced by humans to a varying extent, are still of ecological significance for amphibians and reptiles. The herpetofauna in the remaining areas are less diverse, but include four Species of Conservation Importance.

Birds

4.3.25 In total 227 species have been reported within the Study Area during the period 1988 to 1998 (see ERM 1999b Annex B for full details). This figure represents 50% of the species recorded in Hong Kong since 1958, when the Hong Kong Bird Watching Society (HKBWS) started collecting records, and probably reflects the high habitat diversity and quality of some of the habitat types present within the Study Area. Among the 227 species, 179 (79%) were found within the 100 m corridor of the alignment and this represents 40% of the species recorded in Hong Kong. In Long Valley alone, over 200 species have been recorded.

4.3.26 Of these 227 species, 30 that are regularly recorded in the Study Area can be considered to be Species of Conservation Importance, of which five are Globally Threatened Species (Table 4.2). It is these Species of Conservation Importance that are the focus of this study in terms of assessing the major ecological impacts of the project. It should be noted, in passing, that the Black-faced Spoonbill was listed as “known or likely to occur within 500 m of the proposed route the Sheung Shui to Lok Ma Chau Spurline PPFS” (Binnie 1999). However, the ERM (1999b) study found no records of this species within the Study Area. Although this species is known to forage in habitats similar to the fishponds in the San Tin area, the lack of field records implies that the Study Area does not provide resources that are exploited by the species.

4.3.27 Of the other Species of Conservation Importance eight are Regionally Important Species, four are Species with a Restricted Range in Hong Kong, three are species Declining in Abundance in Hong Kong, and ten are Locally Important species.

4.3.28 The distribution of Species of Conservation Importance within the Study Area is indicated in Table 4.4. below.

Table 4.4

Species of Conservation Importance occurring within 500 m of the Spur Line

and their distribution within the study area (Source: adapted from ERM, 1999b).

Common name	Long Valley	Kwu Tong	Chau Tau	Lok Ma Chau	San Tin	No. of sites
Globally Threatened Species
Greater Spotted Eagle	Ö			Ö	Ö	3
Imperial Eagle	Ö	Ö		Ö	Ö	4
Japanese Yellow Bunting	Ö			Ö		2
Red-billed Starling	Ö	Ö		Ö	Ö	4
Schrenck’s Bittern	Ö					1
Black-winged Stilt	Ö				Ö	2
Common Teal	Ö			Ö	Ö	3
Eurasian Coot	Ö				Ö	2
Great Cormorant	Ö				Ö	2
Chinese Pond Heron	Ö	Ö	Ö	Ö	Ö	5
Great Egret	Ö				Ö	2
Grey Heron	Ö				Ö	2
Little Egret	Ö	Ö		Ö	Ö	4
Bonelli’s Eagle	Ö		Ö			2
Eurasian Eagle Owl			Ö			1
Greater Painted-snipe	Ö			Ö		2
Northern Hobby	Ö		Ö	Ö	Ö	3
Black-naped Oriole	Ö			Ö	Ö	3
Pheasant-tailed Jacana	Ö			Ö		2
Watercock	Ö					1
Bluethroat	Ö			Ö		2
Chestnut Bittern	Ö					1
Common Snipe	Ö	Ö		Ö	Ö	4
Common Stonechat	Ö	Ö	Ö	Ö	Ö	5
Japanese Quail	Ö			Ö		2
Pallas’s Grasshopper Warbler	Ö			Ö	Ö	3
Pintail Snipe	Ö			Ö	Ö	3
Richard’s Pipit	Ö		Ö	Ö	Ö	4
Swinhoe’s Snipe	Ö			Ö		2
Zitting Cisticola	Ö	Ö		Ö	Ö	4
Total number of species recorded	213	49	55	118	110
Total number of Species of Conservation Importance regularly recorded	29	7	7	20	18

Note: No Species of Conservation Importance were regularly recorded in the Shueng Shui area.

4.3.29 Despite some biases from the use of bird records (which inevitably partly reflect the distribution and effort of observers), the results of the analysis of the distribution of birds clearly shows the high ecological importance of the Lok Ma Chau, San Tin and in particular, the Long Valley areas.

4.3.30 The reason for this is primarily because the habitats containing the highest number of species and particularly Species of Conservation Importance are fishponds, agricultural habitats and marshland (Table 4.5). The fish pond habitats are almost entirely restricted to the areas of Lok Ma Chau and San Tin whilst the small remaining areas of marshland and agricultural habitat are mostly restricted to the Long Valley area (Figure 4.1). The San Tin and Long Valley areas are also likely to be particularly valuable because of their large unfragmented area which has the capacity to support relatively large populations of many of the Species of Conservation Importance in question.

4.3.31 The fishponds in the San Tin / Lok Ma Chau area are of additional importance because these are close to and contiguous with the Deep Bay Ramsar Site and as described above are within the Wetland Conservation Areas (WCA) and Wetland Buffer Areas (WBA) adopted by the Town Planning Board (Figure 4.2). Counts carried out by the Hong Kong Bird Watching Society in the San Tin / Lok Ma Chau area of some waterbirds that are Species of Conservation Importance have shown that these ponds are frequently used by significant numbers of these birds (Appendix G).

4.3.32 It should be pointed out that the analysis of the use of the Study Area through a literature review does not reflect recent changes in habitats. In particular, the historical records suggest that the Lok Ma Chau area supported similar numbers of Species of Conservation Importance to the San Tin area. However, the Lok Ma Chau area has lost a large proportion of its wet agricultural land and its ecological importance (e.g. for Greater Painted-snipe) has declined over the period under review, and no longer breeds at this site. This decline in importance is illustrated by the annual waterfowl counts undertaken by the HKBWS. Consequently, although some areas of marshland and agricultural habitat remain in the Lok Ma Chau area these are now small and highly fragmented, and therefore probably now of only moderate ecological value.

4.3.33 In conclusion, the available data clearly show that Study Area is of international, regional and local importance in terms of the avian community it supports. Within the Study Area the large area of fishponds around San Tin and Lok Ma Chau and the extensive agricultural areas with remnant marshland habitats in Long Valley are of particular importance.

Table 4.5

Species of Conservation Importance occurring within 500 m of the Spur Line and their distribution with respect to habitat type (Source: adapted from ERM, 1999b).

Common name	LSF	G-S	W	WA	DA	IA	M	FP	IFP	WC	P	N
Globally Threatened Species
Greater Spotted Eagle		Ö						Ö	Ö			3
Imperial Eagle		Ö						Ö	Ö			3
Japanese Yellow Bunting				Ö		Ö	Ö					3
Red-billed Starling				Ö	Ö			Ö	Ö			4
Schrenck’s Bittern						Ö	Ö					2
Regionally Important Species
Great Cormorant								Ö	Ö			2
Chinese Pond Heron				Ö	Ö	Ö	Ö	Ö	Ö	Ö	Ö	7
Great Egret				Ö		Ö		Ö	Ö	Ö		4
Little Egret				Ö	Ö	Ö	Ö	Ö	Ö	Ö	Ö	7
Common Teal				Ö			Ö	Ö	Ö		Ö	5
Eurasian Coot								Ö	Ö			2
Black-winged Stilt				Ö				Ö			Ö	3
Grey Heron								Ö	Ö			2
Restricted Range in Hong Kong
Bonelli’s Eagle		Ö						Ö				2
Northern Hobby	Ö			Ö	Ö	Ö						4
Greater Painted-snipe				Ö	Ö	Ö	Ö					4
Eurasian Eagle Owl		Ö										1
Declining in Hong Kong
Pheasant-tailed Jacana				Ö		Ö	Ö					3
Watercock						Ö	Ö				Ö	3
Black-naped Oriole	Ö							Ö				2
Locally Important Species
Japanese Quail			Ö		Ö	Ö						3
Common Snipe				Ö	Ö	Ö	Ö	Ö	Ö			6
Pintail Snipe				Ö	Ö	Ö	Ö	Ö	Ö			6
Swinhoe’s Snipe				Ö	Ö	Ö	Ö					4
Richard’s Pipit		Ö		Ö	Ö	Ö		Ö	Ö			6
Bluethroat				Ö	Ö	Ö	Ö					4
Common Stonechat				Ö	Ö	Ö		Ö	Ö			5
Zitting Cisticola				Ö	Ö	Ö	Ö	Ö	Ö			6
Chestnut Bittern						Ö	Ö					2
Pallas’s Grasshopper Warbler				Ö		Ö	Ö		Ö			4
Total number of species recorded	45	40	12	89	66	92	40	89	84	6	23
Total number of Species of Conservation Importance regularly recorded	2	5	1	18	13	19	15	18	16	3	5

Habitat types: LSF – Lowland secondary forest; S-G – Grassland-shrubland mosaic, W-Wasteland, WA – Wet agricultural land; DA – Dry agricultural land; IA – Inactive agricultural land; M – Marsh; FP – active fishpond; IAF – Inactive fishpond; WC- watercourses; P- pond.

Note: No Species of Conservation Importance occurred in fung-shui forest, plantation forest, orchard or developed areas.

Mammals

4.3.34 Being nocturnal and secretive the mammal fauna of Hong Kong is poorly known and no detailed studies have been conducted within the Study Area. However, sightings of mammal signs have been reported and the probability of their presence has been discussed in a number of EIA studies and these have been reviewed in the ERM (1999b) ecological study.

4.3.35 The ERM (1999b) review concluded that although a range of mammal species can be expected to occur within the common agricultural and fishpond habitats within the Study Area, the presence of only a few species could be verified by sightings or proxy records. However, based on the limited information collected, the Study Area does appear to support a relatively rich mammal fauna. In fact, wetland habitats such as fishponds and wet agricultural land probably support large populations of bats due to the associated abundance of flying insects. No species that are definitely known to occur within the Study Area are Species of Conservation Importance. However, a number of species that are likely to occur, given their habitat requirements and range in Hong Kong, would be Species of Conservation Importance, including Chinese Otter Lutra lutra chinensis, Leopard Cat Prionailurus bengalensis, and Crab-eating Mongoose Herpestes urva.

4.4 Habitat evaluations

4.4.1

4.4.1 On the basis of the information on habitats and species reviewed by ERM (1999b), as summarised above, an evaluation of the overall ecological value of each of the main habitat types was carried out by the ERM study. This is reported in full with only minor editorial amendments below and summarised in Section 4.4.13. The habitat review was carried out in accordance with the criteria listed in Annex 8 of the EIA Ordinance Technical Memorandum.

Fung-shui Forest

4.4.2 Four fung-shui forests have been identified within the Study Area - Ho Sheung Heung, Tsung Pak Long, Chau Tau Village and Pun Uk Tsuen. Given the similarity in habitat attributes between the latter two forests, these have been evaluated together (Table 4.6). It is noted that part of the Pun Uk Tsuen fung-shui forest is within the 200 m study corridor.

Table 4.6

Ecological Evaluation of Fung-shui Forests within the Study Area

Criteria	Ho Sheung Heung	Tsung Pak Long	Chau Tau/Pun Uk Tsuen
Naturalness	Natural habitat with very little modification	Natural habitat but heavily modified by the villagers	Natural habitat but with some modification as evidenced by the presence of fruit trees and graves
Size	Of moderate size – extending to~6 ha.	Of relatively small size - extending to ~1.6 ha.	Both are of moderate size – extending to over 4 ha.
Diversity	Structurally complex and species diverse	Structurally simple and species poor	Both are structurally complex with moderate species diversity
Rarity	No rare species known of	No rare species known of	No rare species known of
Re-creatability	The habitat characteristics are difficult to recreate	The habitat characteristics are difficult to recreate	The habitat characteristics are difficult to recreate
Fragmentation	Not fragmented	Seriously fragmented in the under-storey	The Chau Tau forest is not fragmented; while the Pun Uk Tsuen forest is slightly fragmented
Ecological Linkage	Functionally linked with Long Valley, providing a movement corridor and shelter for wildlife inhabiting or utilising the area	No direct ecological linkage, although the large Ficus microcarpa could provide foraging opportunities to insectivorous and frugivorousa animals inhabiting the surrounding area	Functionally linked with the agricultural habitat around Chau Tau and Lok Ma Chau, providing shelter and a movement corridor for wildlife utilising the adjacent habitats
Potential Value	Ecological potential is high if the forest remains undisturbed	Limited due to village development in the surrounding area	Ecological potential would be high if disturbance factors are removed and succession is allowed to occur
Nursery/breeding ground	Not reported, but the sparse forest cover in the surrounding area makes it likely that it is utilised by localised species of birds and mammals	Not reported and not expected	Not reported, but the sparse forest cover in the surrounding area makes it likely that it is utilised by localised species of birds and mammals
Age	The size of the trees indicates an age of over 50 years	The size of the trees indicates an age of over 50 years	The size of the trees indicates an age of over 50 years
Abundance/Richness of Wildlife	High	Low	High
Conclusion	Moderate to High Ecological Value	Low to Moderate Ecological Value	Moderate Ecological Value

Lowland Secondary Forest

4.4.3 The lowland secondary forests within the Study Area are highly fragmented and are mainly found in Lok Ma Chau, Pak Shek Au and the foothills of Ki Lun Shan. With the exception of those patches located in Pak Shek Au and near Ha Wan Tsuen in Lok Ma Chau, each forest area is located beyond the 200 m corridor of the proposed alignment. The ecological value of these forests is evaluated in Table 4.7. For the purposes of the evaluation the patches are combined into two groups according to their habitat characteristics.

Table 4.7

Ecological Evaluation of Lowland Secondary Forest

Criteria	Lok Ma Chau	Pak Shek Au/Ki Lun Shan
Naturalness	Natural habitat, moderately modified by human intervention	Natural habitat, moderately modified by human intervention
Size	The forest next to Ha Wan Tsuen is quite small (~1.2 ha) whereas the other is moderate in size (~4ha)	All are about 2 ha in size and are considered small
Diversity	Both are considered moderate in habitat and species diversity	All of them are moderate in habitat and species diversity
Rarity	Neither the habitat nor the species reported are rare	Neither the habitat nor the species reported are rare
Re-creatability	Re-creatable because of the openness of the forest	Easy to re-create due to the immaturity of the forest
Fragmentation	Both forests are intact	The forests on the foothill of Ki Lun Shan are slightly fragmented; the remaining forest areas are intact
Ecological Linkage	Both are functionally linked with agricultural and fishpond habitats around Chau Tau and Lok Ma Chau, providing shelter and a movement corridor for wildlife utilising adjacent habitats	None of the forest areas are functionally linked to any high value habitat
Potential Value	The ecological potential would be high if disturbance factors are removed and succession is allowed to occur	The ecological potential would be high if disturbance factors are removed and succession is allowed to occur
Nursery/breeding ground	Not reported, but the sparse forest cover in the surrounding area makes it likely that it is utilised by species of birds and mammals	Not reported, but the sparse forest cover in the surrounding area makes it likely that it is utilised by species of birds and mammals
Age	Tree size indicates that both are relatively old when compared with the other lowland secondary forest in Hong Kong	The size and type of tree species present indicates a relatively young age
Abundance/Richness of Wildlife	Moderate	Moderate
Conclusion	Moderate Ecological Value	Low to Moderate Ecological value

Plantation Forest

4.4.4 Plantation forests within the Study Area are small and fragmented, and are mainly found along roadsides or on cut-slopes; the only sizable area of this habitat is located in the Kwu Tung area, close to Pak Shek Au. The ecological value of this habitat type is evaluated in Table 4.8. For the purposes of the evaluation roadside and cut slope plantations are grouped together. It should be noted that only roadside plantation forests are within the 200m study corridor.

Table 4.8

Ecological Evaluation of Plantation Forest

Criteria	Roadside Plantations	Kwu Tung/Pak Shek Au
Naturalness	Plantation	Plantation - semi-natural due to the establishment of native tree species
Size	They are all small in size	Considered moderate in size (~2ha)
Diversity	Both habitat and species diversity are poor	Low in habitat diversity, moderate in species diversity
Rarity	Neither the habitat nor assocaited species are rare	Neither the habitat nor associated species are rare
Re-creatability	This habitat type is readily re-creatable	This habitat type is readily re-creatable
Fragmentation	Heavily fragmented	Not fragmented
Ecological Linkage	None	May provide movement corridors for wildlife
Potential Value	Negligible because of the type of species planted and the dimension and location of the habitat	With time it could develop into secondary forest, of higher ecological value than present
Nursery/breeding ground	None	No wildlife of conservation importance
Age	Young	Young to moderate
Abundance/Richness of Wildlife	Poor	Low to moderate
Conclusion	Low Ecological Value	Low Ecological Value

Orchard

4.4.5 Orchards are a common feature in Kwu Tung and Pak Shek Au and are mostly small in size and highly fragmented. Table 4.9 evaluates the ecological value of this habitat type.

Table 4.9

Ecological Evaluation of Orchard

Criteria	Orchards
Naturalness	Man-made habitats
Size	They are small in size
Diversity	Both the habitat diversity and species diversity are low
Rarity	Neither the habitat nor the species found are rare
Re-creatability	This habitat type is readily re-creatable
Fragmentation	Heavily fragmented
Ecological Linkage	May provide movement corridor for wildlife using the surrounding area
Potential Value	Poor in potential value as they are close to settlements and subject to high levels of disturbance
Nursery/breeding ground	No wildlife of conservation importance is expected
Age	Varies from relatively recent plantings to mature fruit tree of considerable age
Abundance/Richness of Wildlife	Low
Conclusion	Low Ecological Value

Agricultural land

4.4.6 Agricultural land is found mainly in Long Valley and around Chau Tau Village. For the purposes of this evaluation, agricultural land has been classified according to the type and status of current land management. Tables 4.10 - 4.12 evaluate the ecological value of each type of agricultural habitat. It is noted that two large patches of wet agricultural and inactive agricultural land are within the 200m study corridor in Long Valley.

Table 4.10

Ecological Evaluation of Wet Agricultural Land

Criteria	Long Valley	Chau Tau	Lok Ma Chau
Naturalness	Man-made habitat actively managed for crop production	Man-made habitat actively managed for crop production	Man-made habitat actively managed for crop production
Size	Although individual units are small, the overall area of wet agriculture is large	Moderate	Small
Diversity	Habitat diversity is low but diverse invertebrate and vertebrate fauna are supported	Habitat diversity is low but diverse invertebrate and vertebrate fauna are supported	Habitat diversity is low but diverse invertebrate and vertebrate fauna are supported
Rarity	The habitat type is not rare in Hong Kong, but is declining rapidly, with few large areas remaining. A number of rare avifauna species utilise this habitat as feeding/roosting site, including Japanese Yellow Bunting and Painted-Snipe	The habitat type is not rare in Hong Kong, but is declining rapidly, with few large areas remaining. A number of rare avifauna species utilise this habitat as feeding/roosting site, including Northern Hobby and Common Stonechat	The habitat type is not rare in Hong Kong, but is declining rapidly, with few large areas remaining . It may be utilised by some rare/uncommon avifauna using the adjacent ponds as feeding/roosting site, including Greater Painted-snipe and Bluethroat
Re-creatability	It is readily re-creatable provided that land is available	It is readily re-creatable provided that land is available	It is readily re-creatable provided that land is available
Fragmentation	It is moderately fragmented	It is slightly fragmented	It is slightly fragmented
Ecological Linkage	It is part of a well-known site that supports a number of local and international important avifauna	Ecological linkage is low due to surrounding development, but it could still provide habitats for a range of wildlife	Linked to adjacent fishponds and provides feeding habitat for insectivorous animal
Potential Value	High potential if ecologically sensitive management practices are retained or adopted	High potential if ecologically sensitive management practices are retained or adopted	High potential if ecologically sensitive management practices are retained or adopted
Nursery/breeding ground	A wide range of resident and migratory wildlife and birds roost and/or feed in Long Valley, including ardeids from the nearby egretry	A wide range of resident and migratory wildlife and birds are known to breed and/or feed in this habitat.	Not confirmed, although the area is likely to serve as a nursery ground for wildlife known to breed and/or feed in the surrounding area
Age	No information is available and this criteria is considered irrelevant to the ecological value of active wet agricultural land	No information is available and this criteria is considered irrelevant to the ecological value of active wet agricultural land	No information is available and this criteria is considered irrelevant to the ecological value of active wet agricultural land
Abundance/Richness of Wildlife	High	High	High
Conclusion	High Ecological Value	Moderate Ecological Value	Low to Moderate Ecological Value

Table 4.11

Ecological Evaluation of Dry Agricultural Land

Criteria	Long Valley
Naturalness	Man-made habitat actively management for crop production
Size	Small
Diversity	Moderate in habitat diversity and species diversity due to the diversity of food crops planted.
Rarity	The habitat type is not rare in Hong Kong, but a number of rare avifauna species utilise this habitat as feeding and/or roosting sites.
Re-creatability	Readily re-creatable provided that land is available
Fragmentation	Heavily fragmented, but part of a large unfragmented mosiac of habitats
Ecological Linkage	Functionally linked with the surrounding habitat, providing movement corridors and foraging habitat, in particular when flooded in the wet season serves similar functions as the surrounding wet agricultural land
Potential Value	Dry agricultural land could be managed to promote wildlife use by adopting some management practice to further enhance the habitat heterogeneity; however, this may conflict with commercially driven crop production
Nursery/breeding ground	A number of bird species are known to breed in the surrounding areas and this habitat provides nursery grounds to those species, including Greater Painted-snipe, which is known to breed and forage with chicks in this habitat
Age	No information is available and this criteria is considered irrelevant to the ecological value of active dry agricultural land
Abundance/Richness of Wildlife	Moderate
Conclusion	Low to Moderate Ecological Value

Table 4.12

Ecological Evaluation of Inactive Agricultural Land

Criteria	Long Valley	Chau Tau Village
Naturalness	Man-made habitat	Man-made habitat
Size	Large in size	Moderate in size
Diversity	Habitat diversity is moderate but supports diverse species	Habitat diversity is moderate but supports diverse species
Rarity	The habitat type is not rare in Hong Kong, but a number of rare avifauna species have been reported from these areas	The habitat type is not rare in Hong Kong, but a number of rare avifauna species have been reported from these areas
Re-creatability	Readily re-creatable	Readily re-creatable
Fragmentation	Heavily fragmented within the Long Valley	Slightly fragmented
Ecological Linkage	Functionally linked with the surrounding habitats, providing movement corridor, shelter and roosting sites and foraging habitat. It’s significance in avifaunal terms indicate ecological linkages of international significance	Ecological linkage is lower than previous years because of the surrounding development. Provides important reservoir habitat for a range of wildlife that forage in the surrounding area
Potential Value	The potential ecological value of this habitat type could be enhanced by active management as a wildlife habitat	The potential ecological value of this habitat type could be enhanced by active management as a wildlife habitat
Nursery/breeding ground	A wide array of wildlife including rare resident birds utilise this area as a breeding and nursery habitat, including Greater-Painted-snipe which has been noted breeding in this habitat	A wide array of wildlife including rare resident or migrant birds utilise this area as a breeding and nursery habitat
Age	Varies according to period of inactivity, however areas that are abandoned in the long term will either dry out and revert to grass or shrubland or remain wet and revert to marsh.	Varies according to period of inactivity, however areas that are abandoned in the long term will either dry out and revert to grass or shrubland or remain wet and revert to marsh.
Abundance/Richness of Wildlife	High	Moderate
Conclusion	High Ecological Value	Moderate Ecological Value

Fishpond/pond

4.4.7 Fishpond and other pond habitats are mainly located in the areas surrounding San Tin, Lok Ma Chau and Ho Sheung Heung. For the purposes of this evaluation, these pond habitats have been categorised into three different types: active fishpond, inactive fishpond and ponds. Tables 4.13 - 4.15 provide an evaluation of these three habitat types; those located within the 200 m study corridor are primarily the active fishponds in San Tin and Lok Ma Chau, a few inactive fishponds in Chau Tau and Ho Sheung Heung, and a small pond in Ho Sheung Heung near Yin Kong.

Table 4.13

Ecological Evaluation of Active Fishponds

Criteria	San Tin	Lok Ma Chau	Ho Sheung Heung
Naturalness	Man-made habitat, actively managed for fish production	Man-made habitat, actively managed for fish production	Man-made habitat, actively managed for fish production
Size	Active fishponds cover an extensive area within the Study Area	The overall size of active fishpond in Lok Ma Chau is moderate	Small
Diversity	Low habitat diversity but very high in terms of the diversity of wildlife supported	Low habitat diversity but very high in terms of the diversity of wildlife supported	Low habitat diversity but high in terms of the diversity of wildlife supported
Rarity	The habitat itself is not rare but some of the species supported are rare locally or globally	The habitat itself is not rare but some of the species supported are rare locally or globally	The habitat itself is not rare but some of the species supported are rare locally or globally
Re-creatability	Readily re-creatable	Readily re-creatable	Readily re-creatable
Fragmentation	Not fragmented	Not fragmented	Not fragmented
Ecological Linkage	Part of the large contiguous area of fishpond habitat in the Deep Bay area, serving a similar function to those at Mai Po Nature Reserve	Part of the large contiguous area of fishpond habitat in the Deep Bay area, serving a similar function to those at Mai Po Nature Reserve	Linked to the habitats of Long Valley which is known to support a diverse array of wildlife
Potential Value	High ecological potential if managed to enhance wildlife use	High ecological potential if managed to enhance wildlife use	High ecological potential if managed to enhance wildlife use
Nursery/breeding ground	Highly productive systemthat is particularly important for the high densities of important food resources provided for birds during draw-down for harvest periods	Highly productive systemthat is particularly important for the high densities of important food resources provided for birds during draw-down for harvest periods	Highly productive system that is particularly important for the high densities of important food resources provided for birds during draw-down for harvest periods. But fairly isolated from similar habitats
Age	No information is available and is considered irrelevant to the ecological value of this habitat type	No information is available and is considered irrelevant to the ecological value of this habitat type	No information is available and is considered irrelevant to the ecological value of this habitat type
Abundance/Richness of Wildlife	High	High	High
Conclusion	High Ecological Value	High Ecological Value	High Ecological Value

Table 4.14

Ecological Evaluation of Inactive Fishponds

Criteria	San Tin	Lok Ma Chau	Ho Sheung Heung
Naturalness	Man-made habitat, currently subject to limited human disturbance	Man-made habitat, currently subject to limited human disturbance	Man-made habitat, currently subject to limited human disturbance
Size	Small	Small	Small
Diversity	Low habitat diversity but high in terms of wildlife recorded	Low habitat diversity but high in terms of wildlife recorded	Moderate habitat diversity but high in terms of wildlife recorded
Rarity	The habitat is not rare but some of the species supported are rare locally or globally, notably some avifauna	The habitat is not rare but some of the species supported are rare locally or globally, notably some avifauna	The habitat is not rare but some of the species supported are rare locally or globally, notably some avifauna
Re-creatability	Readily-re-creatable	Readily-re-creatable	Readily-re-creatable
Fragmentation	Heavily fragmented	Slightly fragmented	Slightly fragmented
Ecological Linkage	Part of the wider fishpond habitat of the Deep Bay area and serves similar function to those located at Mai Po Nature Reserve	Part of the wider fishpond habitat of the Deep Bay area and serves similar function to those located at Mai Po Nature Reserve	Part of the Long Valley known to support a diverse array of wildlife, notably avifauna
Potential Value	High ecological potential if managed properly to enhance wildlife use	High ecological potential if managed properly to enhance wildlife use	High ecological potential if managed properly to enhance wildlife use
Nursery/breeding ground	Part of the wider Deep Bay fishponds habitats that provide an important breeding/nursery ground for the prey of bird and other mammals, notably the Tilipia and chironomids	Part of the wider Deep Bay fishponds habitats that provide an important breeding/nursery ground for the prey of bird and other mammals, notably the Tilipia and chironomids	Part of Long Valley providing an important nursery ground for the prey of bird and other mammal, notably the Tilipia and chironomids
Age	No information is available	No information is available	No information is available
Abundance/Richness of Wildlife	High	High	High
Conclusion	Moderate to High Ecological Value	Moderate to High Ecological Value	Moderate Ecological Value

Table 4.15

Ecological Evaluation of Ponds

Criteria	Ho Sheung Heung
Naturalness	Man-made habitat actively managed for aqua-culture produce (bloodworms)
Size	Small in size
Diversity	Low habitat diversity but high in terms of wildlife species supported
Rarity	The habitat is not rare but some of the species found utilising this habitat are rare locally, notably some avifauna
Re-creatability	Readily-re-creatable
Fragmentation	Slightly fragmented
Ecological Linkage	Part of the wetland habitat in Long Valley and provides important feeding and/or breeding habitat for a wide array of wildlife
Potential Value	High ecological potential if managed to enhance wildlife use
Nursery/breeding ground	Part of the wetland habitat of Long Valley and the crop of invertebrate fauna (bloodworms and chironomid) provides an important prey base for insectivorous animals, notably ardeids of the nearby egretries as well as bat species. It also provides breeding sites for amphibian and dragonflys.
Age	No information is available and considered irrelevant
Abundance/Richness of Wildlife	High
Conclusion	Moderate to High Ecological Value

Marsh

4.4.8 Marshland is mainly found in Long Valley/Ho Sheung Heung and San Tin. The ecological importance of the main areas of marsh is evaluated in Table 4.16. It should be noted that much of the marshland identified within the study area is within the 200 m study corridor.

Table 4.16

Ecological Evaluation of Marsh

Criteria	Ho Sheung Heung	San Tin
Naturalness	The marshes are natural but may have originated from inactive agricultural fields or ponds	The marsh is natural but may have originated from inactive agricultural fields or ponds
Size	Relatively small compared with similar habitat elsewhere in Hong Kong	Considered as moderate in size
Diversity	Both species and habitat diversity are considered high	Species and habitat diversity are considered moderate and low respectively
Rarity	Not a rare habitat but the site is a known breeding site for the locally rare Greater Painted-snipe	Neither the habitat nor the species found are considered rare
Re-creatability	This habitat is readily re-creatable provided land and appropriate environmental condition is available	This habitat is readily re-creatable provided land is available
Fragmentation	Slightly fragmented	Not fragmented
Ecological Linkage	Part of the wetland system within Long Valley which supports a wide array of wildlife and serves as a shelter, roosting, breeding and feeding site	Part of the wetland systems of San Tin and the Deep Bay area as a whole; invertebrate form an important prey base for insectivorous animal
Potential Value	Moderate potential value - would be higher if actively managed as wildlife habitat	Moderate potential value - would be higher if actively managed as wildlife habitat
Nursery/breeding ground	Part of the wetland system within Long Valley, providing shelter, breeding and nursery sites for a diverse wildlife, including the rare Painted-Snipe	Part of the wetland habitat in the Deep Bay Area and is very important in providing a nursery/breeding ground for the invertebrates which serve as an important food item to avifauna that frequent the Deep Bay area
Age	Evidence from land use divisions and aerial photographs suggests an age of no more than 5-10 years	No information is available, but species diversity and plant age would suggest that this marsh is over 5 years old.
Abundance/Richness of Wildlife	High	Moderate
Conclusion	High Ecological Value	Moderate Ecological value

Water Courses

4.4.9 Water courses found within the Study Area are categorised as river (including the nullah), drainage channels and ditches. Given that the physical conditions of the three rivers (River Beas, River Indus and Sutlej) are quite similar, they are discussed together in Table 4.17.

Table 4.17

Ecological Evaluation of Water Course

Criteria	Rivers	Drainage Channel	Ditches
Naturalness	All rivers are natural with little modification	Man-made habitat	Man-made habitat
Size	Considered inapplicable but the rivers are long and relatively narrow	Considered inapplicable but the channels are long and relatively wide	Considered inapplicable but the ditches are long and narrow
Diversity	Considered low in both habitat and species diversity	Considered low in both habitat and species diversity	Considered low in both habitat and species diversity
Rarity	Neither the habitat nor species found supported is considered rare	Neither the habitat nor species found supported is considered rare	Neither the habitat nor species found supported is considered rare
Re-creatability	Readily re-creatable	Readily re-creatable	Readily re-creatable
Fragmentation	Not fragmented	Not fragmented	Not fragmented
Ecological Linkage	Ecological linkage to surrounding habitat is considered low because of the poor condition of the water and the Main Drainage Channel work.	Ecological linkage to surrounding habitat is considered low because of the poor condition of the water, but the grassy riparian vegetation provides invertebrate prey for insectivorous birds in the surrounding area	Ecological linkage to surrounding habitat is considered low because of the poor condition of the ditches, but the grassy riparian vegetation provides invertebrate prey for insectivorous birds in the surrounding area
Potential Value	The potential of the three rivers could be higher if the water quality improved, but the naturalness of the rivers will be altered by the construction of the MDC	The potential of the channel could be higher if the water quality improved, but the naturalness of the channel will be altered by the construction of the MDC	The potential of the ditches could be higher if the water quality improved
Nursery/breeding ground	Not considered significant	Not considered significant	Not considered significant
Age	No information but considered irrelevant	No information but considered irrelevant	No information but considered irrelevant
Abundance/Richness of Wildlife	Poor	Poor	Poor
Conclusion	Low Ecological Value	Low Ecological Value	Low Ecological Value

Grassland-Shrubland Mosaic

4.4.10 This habitat type is found mainly on the hill-slopes behind Chau Tau, Pak Shek Au and on the foothills of Ki Lun Shan. Given the similarity of the habitat attributes of these areas, they are evaluated as a whole in Table 4.18.

Table 4.18

Grassland-Shrubland Mosaic

Criteria	Grassland-Shrubland Mosaic
Naturalness	Natural habitat but subject to frequent hill-fire
Size	Moderate to large in size compared to other habitats within the Study Area
Diversity	Low habitat and species diversity is found in this habitat
Rarity	Neither the habitat nor the species known to be present are considered rare
Re-creatability	Readily re-creatability
Fragmentation	Not fragmented as a whole over the hill-slope
Ecological Linkage	May function as a movement corridor to wildlife foraging over the agricultural habitat within the Study Area, and the fruit of some shrubby plants could provide food sources to fruitivorous wildlife such as civet.
Potential Value	The potential is moderate provided that hill-fire disturbance ceases and succession is allowed
Nursery/breeding ground	None
Age	In its present form this habitat is best described as being either in the early stages of the succession gradient or in a “pseudo-climax” imposed and maintained by hill-fires
Abundance/Richness of Wildlife	Poor
Conclusion	Low Ecological value

Wasteland

4.4.11 Wasteland is mostly associated with developed areas and is scatter over the Study Area. The ecological value of this habitat is evaluated in Table 4.19.

Table 4.19

Ecological Evaluation of Wasteland

Criteria	Wasteland
Naturalness	Heavily disturbed and disrupted by past and current human activities
Size	Usually small in size
Diversity	Poor in both species and habitat diversity
Rarity	Neither the species present nor the habitat is rare
Re-creatability	Readily re-creatable
Fragmentation	Heavily fragmented within the Study Area
Ecological Linkage	Not functionally linked with any important habitat, although may provide an opportunistic movement corridor for some species
Potential Value	Ecological potential would be moderate if colonisation and succession were allowed to occur
Nursery/breeding ground	None recorded
Age	Usually young
Abundance/Richness of Wildlife	Poor
Conclusion	Low Ecological Value

Developed Area

4.4.12 Developed Area includes densely populated areas, rural villages, open storage areas, construction sites and other light industrial areas. This habitat covers most of the Study Area in Sheung Shui, Kwu Tung, and part of San Tin, and is also the most common habitat type within the 200m study corridor. The ecological value of this habitat is evaluated in Table 4.20.

Table 4.20

Ecological Evaluation of Developed Area

Criteria	Developed Area
Naturalness	Heavily disrupted and disturbed by human activities
Size	Largest habitat type within the Study Area
Diversity	Poor in both species and habitat diversity
Rarity	Neither the species nor the habitat is rare
Re-creatability	Readily re-creatable
Fragmentation	Heavily fragmented within the Study Area
Ecological Linkage	Not functionally linked with any important habitat
Potential Value	The ecological potential is moderate developed areas pass into wasteland and succession is allowed to proceed
Nursery/breeding ground	None for Species of Conservation Importance
Age	No information and considered irrelevant
Abundance/Richness of Wildlife	Poor
Conclusion	Low Ecological Value

Conclusion

4.4.13 Most of the habitats occurring within the Study Area are man-made or highly modified and relatively widespread and common in Hong Kong. Of these, plantation forest, orchard, watercourses, grassland / shrubland mosaics, wasteland and developed areas are of low overall ecological value (Table 4.21).

4.4.14 However, several other habitats are of moderate or high ecological value, primarily because they regularly support Species of Conservation Importance as listed in Table 4.2. Of these the low-lying wet agricultural habitats of Long Valley and the fishponds around San Tin / Lok Ma Chau are of particular ecological value because they support a diverse range of flora and fauna, including large numbers of Species of Conservation Importance.

4.4.15 The fishponds in the San Tin / Lok Ma Chau area are of major importance as this is a large area of habitat that is contiguous with the wetlands of the Deep Bay Ramsar site and because it holds substantial populations of many waterbirds that are Species of Conservation Importance. The conservation importance of the area has been recognised by the Town Planning Board through its designation as a Wetland Conservation Area.

4.4.16 The Long Valley area contains a diverse range of wetland habitats that also support a rich variety of fauna, including a significant number of butterflies, amphibians, reptiles and particularly birds that are Species of Conservation Importance. Such wetland habitats have declined rapidly in Hong Kong in recent years and are under continuing development pressure. Long Valley also contains remnant areas of freshwater marsh, which are now especially threatened in Hong Kong.

Table 4.21

Summary of habitat evaluations.

Habitat / location	Overall evaluation
Fung-shui forest
· Ho Sheung Heung	Moderate – High
· Tsung Pak Long	Low – Moderate
· Chau Tau / Pun Uk Tsuen	Moderate
Lowland secondary forest
· Lok Ma Chau	Moderate
· Pak Shek Au / Ki Lun Shan	Low – Moderate
Plantation forest
· Roadside plantations	Low
· Kwu Tung / Pak Shek Au	Low
Orchard	Low
Wet agricultural land
· Long Valley	High
· Chau Tau	Moderate
· Lok Ma Chau	Low – Moderate
Dry agricultural land
· Long Valley	Low – Moderate
Inactive agricultural land
· Long Valley	High
· Chau Tau	Moderate
Active fishpond
· San Tin	High
· Lok Ma Chau	High
· Ho Sheung Hung	High
Inactive fishpond
· San Tin	Moderate – high
· Lok Ma Chau	Moderate – high
· Ho Sheung Hung	Moderate
Ponds	Moderate – high
Marsh
· Ho Shueng Heung	High
· San Tin	Moderate
Watercourses
· Rivers	Low
· Drainage channel	Low
· Ditches	Low
Grassland-shrubland mosaic	Low
Wasteland	Low
Developed area	Low

4.5 Identification of Potential Impacts

4.5.1 The main operations involved in the construction of the railway, stations and associated structures are described in Chapter 1 of this report. These are likely to have the impacts described below on habitats and their associated species.

Habitat loss

4.5.2 Direct temporary on-site loss of habitats during the construction phase are likely to have significant impacts on all habitats other than those which are already highly degraded or artificial. Habitat loss will occur as a result of the land used for the construction of the track, supporting structures, stations, access roads and storage areas for materials and equipment etc. However, the actual operation phase footprint (i.e. land directly and permanently altered by the project) will be relatively small in most cases, limited to the width of the rail track and the station footprint.

4.5.3 The greatest impact on the ecologically valuable habitats in the WCA around San Tin / Lok Ma Chau is the loss of fishponds, primarily as a result of the construction of the proposed station and border crossing and the impact from the presence of the viaduct across fishponds down to Chau Tau, the total impact in the Lok Ma Chau area amounting to the loss of 9.6 ha of fishponds (Table 4.22). A small, but important, amount (0.85 ha) of agricultural land (active and inactive areas) will be lost within the Long Valley area. Direct losses of other ecologically valuable habitats are negligible.

Table 4.22

Long-term estimated direct habitat loss as a result of the construction of the Spur Line

Section	Sheung Shui	Long Valley	Kwu TongTung/Pak Shek Au	Chau Tau	Lok Ma Chau	Total loss	Total present within 500 m^*1	% Loss within 500 m
Fung Shui wood	0.00	0.00	0.00	0.00	0.01	0.01	17.81	0.0%
Lowland secondary forest	0.00	0.00	0.07	0.00	0.00	0.07	15.11	0.5%
Plant. Forest	0.60	0.00	0.07	0.01	0.01	0.69	32.89	2.1%
Orchard	0.00	0.00	2.46	0.00	0.00	2.46	10.91	22.5%
Dry Agric. land	0.01	0.40	0.00	0.00	0.00	0.41	13.02	3.1%
Wet Agric. land	0.00	0.83	0.00	0.02	0.00	0.85	28.31	0.2%
Inactive Agric. land	0.00	0.00	0.03	0.58	0.00	0.61	38.27	4.3%
Active Fishpond	0.00	0.01	0.00	0.00	9.5	9.5	73.74	12.9%
Inactive Fishpond	0.00	0.00	0.00	0.62	0.04	0.66	12.04	5.5%
Pond	0.00	0.00	0.00	0.00	0.00	0	2.31	0.0%
Marsh	0.00	0.00	0.00	0.00	0.01	0.01	5.17	0.2%
Water courses	0.00	0.00	0.00	0.01	0.00	0.01	8.05	0.1%
Grass/shrub mosaic	0.00	0.00	0.06	0.00	0.01	0.07	54.27	0.1%
Wasteland	0.90	0.00	1.00	0.01	0.00	1.91	35.52	5.4%
Developed areas	8.50	0.01	6.80	2.40	0.07	17.78	432.03	4.1%

Note: All valves are in ha.

^*1 See Table 4.1.

Direct mortality of animals from collisions with trains and associated railway buildings and structures

4.5.4 It is well known, though poorly documented and quantified, that animals may be hit and killed or injured by rapidly moving vehicles and may collide with stationary objects. Some studies in the Netherlands suggest that birds and mammals are the most susceptible to collisions (Van der Grift and Kuijsters, 1998). Such mortality may be substantial where roads or railways pass through areas of high population density or cut-across regular lines of movement e.g. to or from breeding sites for amphibians, and migration routes, roosting flight lines or foraging areas for birds.

4.5.5 The movement of trains along the railway is therefore likely to cause some mortality of animals. However, this is likely to be relatively small as the frequency of train movements is low compared to roads, much of the line is elevated and no areas of particular high animal abundance or regular movement lines are known to occur along the alignment. Nevertheless, some species are particularly susceptible to such hazards, such the Eagle Owl, which is a Species of Conservation Importance. This and other owls and raptors often feed along railway tracks where tall grass and other vegetation can provide suitable habitat for small mammals, reptiles, invertebrates and other prey species. For example, in the Netherlands it was found that collisions with railway traffic was the second most common cause of death for the Buzzard (Buteo buteo) and Kestrel (Falco tinnunculus) (Van den Tempel, 1993). For this study, the area where this is of most concern is along the at grade section of the alignment at Chau Tau. Avifauna such as Common Buzzard, Common Kestrel and Eurasion Eagle Owl can be deterred from feeding along the track by growing up close canopy shrubland, instead of grass, along the embankments.

4.5.6 The proposed cable stay footbridge across the Shenzhen River has a potential impact on larger waterbirds utilising the airspace above the river as a flyway; either directly if birds collide with the structure, or indirectly in causing avoidance. Avoidance impacts within the SAR are subsumed within the 100m impact zone from the station structure.

4.5.7 Collisions are most frequent where buildings or structures are transparent or reflective (i.e. glass) or otherwise difficult to see (e.g. wires). If such structures are avoided then collisions are likely to be infrequent and of minor ecological significance. This issue will be taken into consideration in the mitigation of landscape and visual impacts. As the footbridge will be designed and constructed by mainland engineers, it is not subject to the requirements of this EIA. However, it is proposed that detailed design of the station and footbridge will be required to take this mitigation measure on board.

Habitat fragmentation

4.5.8 Ecologically valuable habitats surrounding the site may be fragmented by the direct effects of habitat loss (e.g. where the track cuts through a block of habitat, such as a woodland) and the physical barriers formed by the track and associated structures (Van der Grift and Kuijsters, 1998). This can have detrimental effects on the remaining habitats through vegetation composition and structural changes at habitat edges, inhibition of dispersal of species and increased susceptibility to disturbance and predation.

4.5.9 Fragmentation impacts are likely to be most severe for woodland and wetlands as many important species inhabiting these require large tracts of suitable habitat (e.g. predatory mammals and birds).

4.5.10 The physical structures of linear structures such as roads and railways can be major barriers to dispersal and migration for many terrestrial species. This can be particularly severe for small animals, such as amphibians, that may return annually to specific traditional breeding sites. In Europe, observations of the Common Toad (Bufo bufo) have shown that railway lines may be an insurmountable barrier when there are no openings between the rail and ballast (Ingelmann, 1994; Müller and Berthold, 1996). For mammals the degree to which railway lines form barriers is generally considered to be inversely proportional to their body size and hence their mobility (Bergers, 1997). However, some mammals that are sensitive to disturbance and require large areas for foraging may also be affected.

4.5.11 At Lok Ma Chau, a combination of direct habitat loss (to open storage/vehicle parking), abandonment of wet agriculture, lowered water levels and vegetational succession, compounded by the fragmentation of the remaining areas of suitable habitat are considered the most likely reasons for the site being abandoned by Greater Painted-snipe, which bred there in the past.

4.5.12 Of the freshwater wetlands in Hong Kong that are of importance in terms of the avian community they support, most are of considerable size (>20 ha), indicating the intolerance of such communities to fragmentation

Disturbance

4.5.13 In addition to direct habitat loss, the Spur Line will cause disturbance impacts to Species of Conservation Importance in adjoining wetland areas, even though these will not be physically affected by development. These impacts include active disturbance (arising from noise or other human activities) and passive disturbance resulting from birds avoidance of structures (principally the station and the railway viaduct). However, for the purposes of the present impact assessment they do not include impacts from other existing sources which have already rendered habitats disturbed (for example the Lok Ma Chau Boundary Crossing) and the structures on the opposite side of the boundary. Therefore, the assessment will be on the conservative side.

4.5.14 The constructed wetlands which will be constructed as compensation for wetland losses under the Main Drainage Channel project for San Tin (ERM 1999a) and Lok Ma Chau Boundary Crossing Expansion (Binnie 1999) are very close to the existing Lok Ma Chau Boundary Crossing. It is considered that tThe additional potential impact on these constructed wetlands from the railway Spur Line involves loss of habitat and disturbance. This impact has been included on in the calculation of habitat loss and disturbance under Spur Line., although it will only occur over a very short distance of the track, and can be mitigated though the implementation of mitigation measures proposed for the Spur Line schemehas considered the impact on the compensation wetland areas, and the impact from the viaduct piers has been included in the calculation of habitat loss and disturbance. The implementation of mitigation measures to be provided under Spur Line include compensation for potential impacts on these constructed wetlands. has therefore provided for compensation of habitats which may be impacted in its construction and operation.

4.5.15 General disturbance effects are widely recognized and have been documented in the past (e.g. Hockin et al. 1992). Such effects may include the complete avoidance of an area (which is therefore comparable to habitat loss) and reduced densities (e.g. where only certain less-sensitive or accustomed individuals use the area). Other less-obvious disturbance effects may include reduced habitat quality through reduced feeding efficiency (e.g. because of a need for increased vigilance), which may in turn lower survival rates or breeding output. Noise may also interfere with breeding behaviour, through for example obscuring bird song which may then reduce breeding densities (Reijnen et al. 1995, 1996). Care must therefore be taken in interpreting the observation of a species close to a disturbance source as being evidence that there is no disturbance effect. On the other hand, the observation that there is a disturbance effect, may not necessarily mean that this results in a population impact, as animals may, for example, move to alternative areas if suitable habitat is in excess (i.e. populations are not at carrying capacity).

4.5.16 Species affected by disturbance are likely to be most birds and mammals, particularly the larger species. Such species are likely to be disturbed by loud noises, moving objects and the presence of people. Stationery objects such as buildings and trees, are also sources of disturbance as these may obscure flight lines and views of potentially approaching predators. Many species will therefore avoid approaching such structures.

4.5.17 According to Hill et al. (1997) the severity and overall impact of disturbance (at least on birds) is likely to depend on the following factors:

i) intensity of disturbance;

ii) duration and frequency of disturbance (continuous, infrequent, regular, variable);

iii) proximity of source;

iv) seasonal variation in sensitivity of affected species;

v) presence of people associated with the source;

vi) whether animals move away, but return after disturbance ceases;

vii) whether important numbers are affected;

viii) whether there are alternative habitats available nearby;

ix) whether rare, scarce or especially shy species are affected.

4.5.18 In particular, disturbance impacts are normally greatest when they are close and frequent. Although, some species can adapt to regular disturbances from noise and moving machinery, the close proximity of people is rarely adapted to. Thus, species tend to show increasing responses across a gradient of severity, from passive-low-level and continuous to active-high-level continuous (Hockin et al. 1992).

4.5.19 Although no information on the specific disturbance effects of railways is known to be available there are numerous international references documenting cases of impacts from road disturbance. Such impacts tend to be fairly low when compared to industrial, urban and residential sources where people are highly visible or where disturbance sources are unpredictable. In the case of the Spur Line, the low frequency of trains (compared with road traffic) and the noise mitigation measures which are included in the design, will reduce the disturbance impacts within the Long Valley area. In addition the design and alignment of the railway has specifically taken into account disturbance to wildlife and aimed to reduce ecological impacts, as described in Chapter 2 of this report.

4.5.20 Although there are no scientific studies of the impacts of railway disturbance in Hong Kong, anecdotal observations suggest that some wetland Species of Conservation Importance occurring along the Spur Line route are not highly susceptible to railway disturbance. Observations by team members of birds alongside the East Rail line south of Tai Po Market and at Lo Wu indicate that at least some egrets adapt to the presence of trains passing by. Breeding numbers at an egretry alongside the KCRC at Tai Po Market do not appear to be affected by normal operational use of the railway and station, and although this egretry was established after the station came into use, the effects of construction activity on an active egretery are unknown (Young and Cha 1995).

4.5.21 In spite of the anecdotal evidence presented above and the relatively few studies that have been carried out in Hong Kong, there is some evidence of disturbance impacts. The Town Planning Board Guidelines for Application for Developments within the Deep Bay Area (TPB PG-NO. 12B) acknowledges that, on the basis of confidential scientific studies undertaken during the Fish Pond Study, reduced bird usage occurs on fish ponds that are adjacent to or in the vicinity of open storage, industrial uses, dispersed village developments and roads. Consequently the Board recommends that mitigation for such disturbance impacts should be undertaken.

4.5.22 The use by birds of habitats close to major highways was also shown to be limited due to noise disturbance in the EIA for the Expansion of the Lok Ma Chau Boundary Crossing (Binnie 1999).

4.5.23 In conclusion, given the amount of circumstantial evidence it must be assumed that some degree of disturbance impacts will occur as a result of the operation of the railway. The main species groups that will be affected will be sensitive species, such as flock-feeding waterbirds, larger herons and egrets and birds of prey. These, are some of the most threatened and important species using the wetland habitats within the vicinity of the Spur Line (ERM 1999b). Mitigation to be implemented along the Spur Line (for noise reduction and ecological habitat compensation) will therefore be designed to minimize the disturbance impacts and provide suitable compensation habitat for sensitive species.

Predicted effects on species

4.5.24 As described above, the sensitivity of different species to disturbance will vary. An assessment of the predicted impacts of disturbance on each of the bird Species of Conservation Importance that regularly occur in significant numbers has therefore been carried out. This assessment interprets the findings of the above literature review in relation to the context of the present Spur Line development by additionally taking account of:

• Observations by the ecological study team along Hong Kong's existing rail network,

• Disturbance impacts observed during the course of development projects within the Deep Bay area, in particular monitoring of the Shenzhen River Regulation Works between Lok Ma Chau and Mai Po, during construction,

• Disturbance impacts observed from the operation of the Lok Ma Chau (road) Border Crossing during ecological survey in connection with the proposed improvement of the Border Crossing Plaza, and during ecological surveys for the San Tin main Drainage Channel EIA

• Field experience of the ecological study team, who are highly familiar with the distribution of birds in the Deep Bay area and have undertaken studies of waterbird ecology in fishponds (including fieldwork for the Planning Department Fish Pond Study).

• Existing noise levels in the vicinity of the Spur Line and anticipated noise levels from the construction and operation of the Spur Line.

4.5.25 The Spur Line is designed to achieve noise levels of 50dB(A) at night-time in rural areas of Long Valley and Lok Ma Chau at the nearest sensitive receivers to the track (approximately 15m away). Permitted noise levels increase to 60dB(A) during the day-time. This noise level of 50 dB(A) is considerably lower than the noise levels expected at a similar distance from a major highway such as Fanling Highway. The potential impacts on wildlife in Long Valley depend on the relative distance from the future Spur Line and the existing highway.

4.5.26 Noise level measurements carried out as part of site surveys for the present study indicate that Leq (30 minute) levels range from 49.8 dB(A) at Tsung Pak Long (Southwest of NSR 14) to 64.9 dB(A) alongside the River Sutlej. The major source of noise at the River Sutlej was the equipment used for river training construction. Trains on the East Rail line also contributed to this noise level. Just north of the marsh area in the centre of Long Valley, a Leq (30 minute) of 52.1 dB(A) was measured. L₉₀ levels which indicate the noise level of 90% of the time varied from 46.-dB(A) southwest of NSR 14, to 47.5dB(A) at the Long Valley marsh and 54.0dB(A) alongside the River Sutlej. The noise levels close to the Spur Line will clearly be influenced by the railway more than by the Fanling highway. However, comparing the required night-time Leq (30 minute) background levels for Spur Line of 50dB(A), with the measured background levels, the noise impact on wildlife is not expected to be great. The type of noise, its source, frequency and variation may have different effects on the wildlife of the area. For example, the presence of a human form has an impact on the behaviour of certain species, even though noise levels are negligible. Noise levels alone, therefore, can only give an incomplete indication of potential disturbance impacts.

4.5.27 In the Lok Ma Chau area, noise levels have been predicted during the construction and operation of the Lok Ma Chau Boundary Crossing. During construction, levels of 57 to 74dB(A) are expected at noise sensitive receivers (NSRs) (mainly residences in the area) from various construction plant. Mitigation measures are being implemented to reduce the level to 60dB(A) in accordance with Deep Bay Guidelines for the Special Measures Zone. During operation of the road, noise levels are predicted to be 60 to 70dB(A) at various sensitive receivers (Binnie 1999). These levels fall within the acceptable limits for traffic within the area.

4.5.28 During operation of the Spur Line noise levels will exceed Acceptable Noise Levels (ANLs) at some NSRs which are very close to the railway line, in the absence of mitigation measures. A variety of mitigation measures is being considered to reduce noise levels to ANLs, which are described in the Noise Chapter of this report.

4.5.29 Although the additional Hong Kong information is not truly quantitative or amenable to rigorous statistical analysis, or adequate for application in respect to other development proposals, it is sufficient to provide a basis for predictive assessment with respect to this particular development.

4.5.30 Thus, disturbance impacts have been calculated by defining distances from the edge of the disturbance source to the furthest point of:

• an avoidance zone, i.e. where a particular species is precluded from using the area; and

• a zone of reduced density, i.e. where the numbers of a species are lower than they would be in the absence of disturbance either because it occurs in lower numbers (more tolerant individuals) and/or for a shorter period of time (for example during periods of reduced human activity). In this analysis, it is assumed that the overall utilisation of the zone of reduced density is 50% of that in undisturbed areas (0% at the border with the avoidance zone and 100% of density at the border with the undisturbed areas).

4.5.31 These distances have been estimated for construction and operation phase impacts and for disturbance impacts in the fishpond area around San Tin / Lok Ma Chau (which take into account the predicted avoidance effects of the large station structure) and for the viaduct sections of track crossing Long Valley. An overall qualitative assessment of disturbance impacts has then been attributed to each species.

4.5.32 Disturbance impacts in other habitats are not considered because they are unlikely to be significant due to existing disturbance sources and the low ecological value of these habitats.

4.5.33 These predicted disturbance impacts have been calculated on the assumption that there will be minimal visual human disturbance from the station (i.e. people within the station will not be highly visible from outside) and take into account basic mitigation measures, including the creation of wetland habitats with reedbeds and the planting of trees and bamboo as screening.

4.5.34 The results of this assessment are presented in Table 4.23 below. Further information on the basis of each species’ assessment is provided in Appendix G.

Table 4.23

Predicted disturbance impacts from the construction and operation of the railway and station on

regularly occurring Species of Conservation Importance

Species	Construction phase		Operation phase				Overall disturbance sensitivity
			Lok Ma Chau station complex and viaduct section		Viaduct track sections only
	Exclusion distance (m)	Max distance of reduced density (m)	Exclusion distance (m)	Max distance of reduced density (m)	Exclusion distance (m)	Max distance of reduced density (m)
Globally Threatened Species
Greater Spotted Eagle	200	500	200	500	#	#	Very high
Imperial Eagle	200	500	200	500	#	#	Very high
Japanese Yellow Bunting	40	100	20	30	0	30	Low
Schrenck's Bittern	50	100	20	20	50	75	Low
Red-billed Starling	100	200	50	100	40	75	Moderate
Regionally Important Species
Great Cormorant	200	400	100	150	#	#	High
Chinese Pond Heron	100	300	20	30	0	100	Moderate
Great Egret	200	400	100	100	100	150	High
Little Egret	100	400	20	100	30	100	Moderate - High
Common Teal	100	300	50	100	50	100	Moderate - High
Eurasian Coot	50	200	20	50	25	50	Moderate
Black-winged Stilt	100	200	50	50	50	75	Moderate - High
Grey Heron	200	400	100	200	100	150	High
Restricted Range In Hong Kong
Greater Painted Snipe	75	150	-	-	0	50	Moderate
Northern Hobby	100	200	100	100	50	100	High
Declining in Hong Kong
Pheasant-tailed Jacana	100	200	30	50	50	75	Moderate - High
Watercock	50	100	30	50	30	75	Moderate
Black-naped Oriole	50	100	50	50	-	-	Moderate
Locally Important Species
Chestnut Bittern	50	100	20	20	50	75	Low
Common Snipe*	50	200	30	30	0	60	Moderate
Pintail Snipe*	50	200	30	30	0	60	Moderate
Swinhoe's Snipe*	50	200	30	30	0	60	Moderate
Japanese Quail	40	75	-	-	20	75	Moderate
Richard's Pipit	50	100	20	30	20	50	Low
Bluethroat*	50	100	20	30	0	50	Low
Common Stonechat*	40	100	20	30	0	50	Low
Zitting Cisticola*	40	60	20	30	0	50	Low
Pallas's Grasshopper Warbler*	40	60	20	30	0	30	Very Low

Notes. * The predicted impacts for these species are for the station site and the at-grade section of the line. Since these species will also use freshwater wetlands, there is no exclusion distance for the elevated sections of the line where these species are expected to use the area under the line to some extent.

# Species not predicted to be affected. - Species absent.

Predicted impacts take into account basic mitigation measures, including the creation of wetland habitats with reedbeds and the planting of trees and bamboo as screening.

Dust deposition on vegetation

4.5.35 Unmitigated construction operations, if allowed to take place, would be likely to create significant levels of dust, e.g. due to the use of haul roads, and wind blown dust in works areas which can be deposited on nearby habitats. This can cause vegetation damage, which can, in turn, have secondary affects on associated fauna (such as insects and birds). In severe cases dust deposition can also affect animals, such as nesting birds directly. Impacts from dust deposition as a result of construction operations are, however, likely to be temporary and reversible and standard mitigation measures will be implemented which will negate harmful dust emissions.

Increased sediment load

4.5.36 Dust and exposed earth from construction operations may also enter watercourses through run-off, particularly during heavy rainfall periods. This can lead to high turbidity from soil particles and eutrophication as a result of nutrient enrichment (as phosphates, which are normally the limiting nutrient in freshwater systems, are bound to soil particles). Aquatic macrophytes may be reduced or lost completely as a result of reduced light penetration due to the increased turbidity from soil particles and increased free-floating algae populations following eutrophication. Severe eutrophication can also lead to oxygen depletion and the impoverishment of aquatic animal communities and, in turn, other animals, such as waterbirds, that feed on them. In addition, construction of the footbridge across the Shenzhen River will lead to sediment in the river being disturbed during cofferdam construction for the bridge piers. Such disturbed sediment will have a similar effect to that of increasing turbidity in the river.

4.5.37 However, as reported under the habitat evaluation section, the water quality in existing watercourses is poor and their ecological quality is low. Ecological degradation of these habitats may not, therefore, occur if pollutant levels in site run-off are not significantly above those of the water-courses, or if the run-off volume is relatively low. Additional sediment inputs including those arising from construction in within the Shenzhen River, and potential associated additional nutrients loads may, however, have off-site impacts on the Deep Bay mudflats and coastal waters. Mitigation measures must be put in place to prevent adverse impacts on the ecological resources of Deep Bay.

4.5.38 Mitigation measures are designed to strive towards minimal pollution of water-courses from the construction and ongoing operation of the railway, station and footbridge across the Shenzhen River, following the precautionary principle. These issues are dealt with in more detail in the Water Chapter of this report.

Toxic pollutants from construction operations

4.5.39 During construction, there is potential for some areas which are contaminated with chemicals and toxic pollutants from storage activities to be disturbed during excavation. It is essential that measures are taken to minimise run-off from these sites during construction and avoid contaminated material entering water courses.

4.5.40 Spills and run-off from construction sites can contain high levels of toxic pollutants (such as oil) which can cause direct mortality of plants and animals (in severe cases), sublethal impacts (e.g. by reduced breeding success) or indirect effects through impacts on food resources etc. Watercourses and soils alongside or under the viaduct sections of the track may therefore become contaminated. Where soils and sediments are contaminated long-term effects may occur. For example, soil contamination under and adjacent to the viaducts sections may inhibit vegetation reestablishment and planned habitat compensation restoration and creation measures. Contamination of watercourses may have off-site effects on ditches, rivers and the valuable estuarine waters and mudflats of Deep Bay. During the construction of the Footbridge across Shenzhen River, there is a potential for concrete workingswashings to enter the river, which would have an adverse effect on ecology downstream. Measures which should be put in place to prevent thisn impact are described in the underwWater Chapter of this report.

4.5.41 Bioaccumulation may also occur where toxic substances are passed up the food chain in increasing concentrations. As a result top level predators such as some reptiles, mammals and birds of prey can be particularly susceptible.

Soil compaction

4.5.42 During construction, areas of land will be cleared for use as Works Areas. During this period, areas of ground, which are currently undisturbed, will be compacted for use as storage areas, vehicle movement or other purposes. This is particularly important beneath the viaduct within Long Valley. Wetland areas into which rain quickly infiltrates, will becomes hardened and rainfall will wash off, taking sediment with it, to be deposited on land adjacent to the compacted site, or in watercourses. The wetland function of these area may therefore be lost temporarily. Although the impact is temporary, extensive compaction over a long period will make habitat restoration more difficult because of the need to re-establish the soil base for a wetland area.

Hydrological disruption

4.5.43 Channelisation of several tributaries within the River Indus basin is being carried out for flood control of the Sheung Shui and Fanling Hinterland. Potential impacts resulting from this channelisation on the ecology of the Long Valley area have been studied by Maunsell (1997). Reduction in the frequency of flooding in Long Valley will change the wetland nature of the area, possibly leading to a decline in the numbers of wetland dependent birds in this area. Channelisation is also likely to lower the groundwater levels in the area, if not mitigated.

4.5.44 Mitigation measures to minimize impacts are focused on retention of the groundwater level through the incorporation of weirs along the channel length and creation of wetland areas in the meanders which will be cut off by the channelisation project. The frequency of flooding will be reduced from approximately every 2 years to approximately once in 50 years.

4.5.45 Hydrological impacts from the Spur Line will be minimal compared with the channelisation project. The presence of an impermeable structure will increase the run-off, however, the area would receive rainfall even without the railway present. The main effect will be a small increase in the height of the peak on the hydrograph for the area.

4.5.46 The potential impact from the preservepresence of piers for the footbridge areasacross Shenzhen River will be minimized through good design of the pier footings. Reduction in sediment turbulence during intermediate storems will minimize effects on the Deep Bay mudflats downstream.

4.5.47 In conclusion, the reduction in the frequency of flooding in the area may reduce the extent of wetlands in the area, or change permanently flooded areas to become seasonally flooded. The design of, and water source for, wetlands proposed as part of ecological compensation for the Spur Line project should take these possible drainage changes into account. Appropriate footbridge design will result in minimal impacts on ecological resources downstream.

Summary evaluation of impacts

4.5.48 The severity of general impacts described above will vary according to the habitats and species present and the actual operations taking place in various locations along the railway line. An evaluation of the severity of the predicted impacts is therefore presented for each habitat type of moderate or high ecological value, firstly with respect to construction phase impacts and then long-term impacts from the operation of the railway and station. This evaluation is based on the “criteria for evaluating ecological impacts” given in Annex 8 of the EPD “Technical Memorandum on Environmental Impact Assessment Process”.

4.5.49 Along much of the alignment, current or future works (by others) will occur before the construction of the Spur Line. These include the expansion of the Lok Ma Chau Boundary Crossing and the construction of the Fanling, Sheung Shui and Hinterland Drainage Channels. The San Tin Drainage Channel construction will start during the construction of Spur Line. These will result in ecological impacts including habitat loss and degradation and disturbance to wildlife. Mitigation measures to address ecological impacts have been incorporated in the design of each of these projects. As the present ecological assessment is concerned with impacts under the Spur Line Project, changes to the existing value of the habitats and the presence of Species of Conservation Importance from these current and future works are taken into account in the baseline assessment. Cumulative impacts from these projects are therefore taken into account through their inclusion in the baseline ecological conditions for the Spur Line EIA.

4.5.50 A summary of the likely implementation programme for the works to be carried out in the area is shown below:

Project	Period of Works
Lok Ma Chau Boundary Crossing	May 1999 - 2003
San Tin Main Drainage Channel	Construction Start Date 2002 - 2003
Fanling and Sheung Shui Main Drainage Channels (Long Valley area)	June 1999 - May 2001

4.5.51 This indicates that there will be some overlap between the Spur Line and other projects. This will be primarily in the area of San Tin Drainage Channel, during construction works. Impacts should be minimised through adherence to good site management practices which will minimise disturbance impacts and other adverse ecological effects.

Construction phase

4.5.52 The predicted short-term impacts of the construction phase of the Spur Line project are likely to be:

• Permanent and temporary habitat loss (i.e. for works access and storage areas)

• Habitat fragmentation

• Disturbance

• Dust deposition

• Increased sediment load (including in Shenzhen River and downstream in Deep Bay)

• Pollution from the construction operations (e.g. oil leaks) concrete washings during construction, or accidents involving toxic chemicals.

• Soil compaction

• Hydrological disruption

4.5.53 A preliminary evaluation of the severity of each predicted impact is presented for each habitat of moderate or high ecological value in Tables 4.24 – 4.27. More detailed assessments for individual bird Species of Conservation Importance are provided in Appendix H.

Table 4.24

Initial assessment of potential ecological impacts on forest habitats

from the construction of the Spur Line

Potential impact	Location	Type	Evaluation of impact
			Species of Conservation Importance	Size / Abundance	Duration	Reversibility	Magnitude
Habitat loss from construction operations	On-site	Direct	Few affected	Not quantified but low	Mostly temporary (track mainly on viaduct)	Reversible but regrowth is very slow	Considerable ecological change
Habitat fragmentation	Off-site	Indirect	Few affected	Not quantified but low	Mostly temporary (track mainly on viaduct)	Reversible but regrowth is very slow	Considerable ecological change
Disturbance of wildlife by noise and visual movement	Off-site	Indirect	Few affected	Not quantified but very low	Mostly temporary (track mainly on viaduct)	Reversible	Low impact
Dust deposition on surrounding habitats	Off-site	Indirect	None directly affected	Potentially a large area could be affected	Short and temporary	Reversible	Low impact
Increased sediment load in watercourses	Off-site	Indirect	None directly affected	Potentially a large area could be affected	Short and temporary	Reversible	Moderate impact
Pollution from toxic chemicals	On-site & Off-site	Direct & indirect	Few likely to be affected	Potentially a large area could be affected	Potentially long-lasting	Dependent on chemicals involved	Potentially high impact
Soil compaction	On-site	Direct	None directly affected	Not quantified but probably moderate	Temporary	Mostly reversible	Low impact
Hydrological disruption	On-site	Direct	None directly affected	Not quantified but probably low	Mostly temporary	Mostly reversible	Low impact

Table 4.25

Initial assessment of potential ecological impacts on agricultural habitats

from the construction of the Spur Line

Potential impact	Location	Type	Evaluation of impact
			Species of Conservation Importance	Size / Abundance	Duration	Reversibility	Magnitude
Habitat loss from construction operations	On-site	Direct	Many wet and dry agricultural land species affected	Not quantified but low	Mostly temporary (track mainly on viaduct)	Reversible	Considerable ecological change
Habitat fragmentation	Off-site	Indirect	Many wet and dry agricultural land species affected	Not quantified but low	Mostly temporary (track mainly on viaduct)	Reversible	Considerable ecological change
Disturbance of wildlife by noise and visual movement	Off-site	Indirect	Many wet and dry agricultural land species affected	Moderate area affected	Mostly temporary (track mainly on viaduct)	Reversible	Moderate impacts
Dust deposition on surrounding habitats	Off-site	Indirect	Uncertain, amphibians and invertebrates most likely to be affected	Potentially a large area could be affected	Short and temporary	Reversible	Low impact
Increased sediment load in watercourses	Off-site	Indirect	Uncertain, amphibians and invertebrates most likely to be affected	Potentially a large area could be affected	Short and temporary	Reversible	Moderate impact
Pollution from toxic chemicals	On-site & Off-site	Direct & indirect	Potentially many species could be affected	Potentially a large area could be affected	Potentially long-lasting	Dependent on chemicals involved	Potentially high impact
Soil compaction	On-site	Direct	Many wet agricultural land species affected	Not quantified but probably moderate	Temporary	Mostly reversible	Moderate impact
Hydrological disruption	On-site & off-site	Direct	Many wet agricultural land species affected	Not quantified but probably low	Mostly temporary	Mostly reversible	Moderate impact

Table 4.26

Initial assessment of potential ecological impacts on fishpond and pond habitats

from the construction of the Spur Line

Potential impact	Location	Type	Evaluation of impact
			Species of Conservation Importance	Size / Abundance	Duration	Reversibility	Magnitude
Habitat loss from construction operations	On-site	Direct	Many wetland waterbirds affected	High	Mostly permanent due to station building	Reversible	Considerable ecological change
Habitat fragmentation	Off-site	Indirect	Many wetland waterbirds affected	Moderate	Mostly permanent due to station building	Reversible	Considerable ecological change
Disturbance of wildlife by noise and visual movement	Off-site	Indirect	Many wetland waterbirds affected	Large area affected	Mostly permanent due to station building	Reversible	High impacts
Dust deposition on surrounding habitats	Off-site	Indirect	None likely to be affected	Potentially a large area could be affected	Short and temporary	Reversible	Low impact
Increased sediment load in watercourses	Off-site	Indirect	None likely to be affected	Potentially a large area could be affected	Short and temporary	Reversible	Moderate impact
Pollution from toxic chemicals	On-site & Off-site	Direct & indirect	Potentially many species could be affected	Potentially a large area could be affected	Potentially long-lasting	Dependent on chemicals involved	Potentially high impact
Soil compaction	On-site	Direct	None likely to be affected	Not quantified but probably moderate	Temporary	Mostly reversible	Low impact
Hydrological disruption	On-site & off-site	Direct	None likely to be affected	Not quantified but probably low	Mostly temporary	Mostly reversible	Low impact

Table 4.27

Initial assessment of potential ecological impacts on marsh habitats

from the construction of the Spur Line

Potential impact	Location	Type	Evaluation of impact
			Species of Conservation Importance	Size / Abundance	Duration	Reversibility	Magnitude
Habitat loss from construction operations	On-site	Direct	Many wetland species affected	Not quantified but low	Mostly temporary (track mainly on viaduct)	Reversible	Considerable ecological change
Habitat fragmentation	Off-site	Indirect	Many wetland species affected	Not quantified but low	Mostly temporary (track mainly on viaduct)	Reversible	Considerable ecological change
Disturbance of wildlife by noise and visual movement	Off-site	Indirect	Many wetland species affected	Moderate area affected	Mostly temporary (track mainly on viaduct)	Reversible	Moderate impacts
Dust deposition on surrounding habitats	Off-site	Indirect	Uncertain, amphibians and invertebrates may be affected	Potentially a large area could be affected	Short and temporary	Reversible	Low impact
Increased sediment load in watercourses	Off-site	Indirect	Uncertain, amphibians and invertebrates may be affected	Potentially a large area could be affected	Short and temporary	Reversible	Moderate impact
Pollution from toxic chemicals	On-site & Off-site	Direct & indirect	Potentially many species could be affected	Potentially a large area could be affected	Potentially long-lasting	Dependent on chemicals involved	Potentially high impact
Soil compaction	On-site	Direct	Many wet agricultural land species affected	Not quantified but probably moderate	Temporary	Mostly reversible	Moderate impact
Hydrological disruption	On-site & off-site	Direct	Many wet agricultural land species affected	Not quantified but probably low	Mostly temporary	Mostly reversible	Moderate impact

Operation Phase

4.5.54 The predicted impacts of the ongoing operation of the railway line from Sheung Shui station and Lok Ma Chau station are likely to be:

• Permanent habitat loss (i.e. after completion of construction operations and re-creation / restoration of habitats)

• Habitat fragmentation

• Direct mortality from collisions with trains and associated railway buildings and structures

• Disturbance

• Pollution from the railway operations (e.g. oil leaks) or accidents involving toxic chemicals.

• Hydrological disruption

4.5.55 As with the construction impacts, the severity of operation impacts will vary according to the habitats and species present and the actual operations taking place along the railway line. A preliminary evaluation of the severity of the predicted impacts is therefore present for each section of the line in Tables 4.29 – 4.32.

4.6 Concurrent Projects with Potential for Cumulative Impacts

4.6.1 Table 4.28 lists major projects that will be conducted concurrently with the construction of the Spur Line. Each of these projects has the potential to impact the ecology within the Study Area. The locations of these projects in relation to the Spur Line Alignment are shown on Figure 6.1. Projects of particular significance, due to their proximity to the proposed Lok Ma Chau station, are: construction of the San Tin Eastern Main Drainage Channel, and expansion of the Lok Ma Chau Boundary Crossing. The NENT Study which includes the recently publicised Kwu Tung SGA may also have a considerable impact upon the ecological resources of the Long Valley area, primarily through direct habitat loss.

4.6.2 The San Tin Eastern mMain Drainage Channel which will be constructed to the west of the Lok Ma Chau Boundary Crossing road will be crossed by the elevated section of the Spur Line. The section of the channel which will be impacted will have grasscrete banks and will include an adjoining area of managed wetland to be provided in mitigation for fish pond loss due to the channel’s construction. Direct impact due to the Spur Line at its crossing point will be the formation of one or possibly two pillars for the viaduct, resulting in direct loss of grasscrete bank and/or managed wetland of 25-50 square metres. The habitat to be lost is stated in ERM (1999a) to be of moderate to high value for wetland dependant bird species, though the targeted species are not specified. Since this habitat is to be provided on approximately a 1:1 basis in compensation for fish pond loss (ERM 1999a), this direct loss is catered for within the requirement to mitigate for fish pond loss or disturbance described later in paragraph 4.7.12 and in the wetland enhancement requirements specified inparagraphs 4.7.15 to 4.7.18. In addition, the Spur Line will create disturbance impacts upon a short section of the channel and constructed wetland area in the same fashion as it will impact on surrounding fish ponds and othe rother wetlands. This impact is likewise included within the land requirement calculated to mitigate for disturbance impacts detailed in Paragraph 4.7.15 to 4.7.18 and Table 4.33.

4.6.3 Potential cumulative impacts of this project had been addressed as far as is possible through further enhancement of mitigation areas for the Fanling, Sheung Shui & Hinterland Main Drainage Channels.

4.6.4 Subsequent projects should avoid direct negative impacts on mitigation areas, temporary or permanent, that are already designated through this or any other project. Such negative impacts include direct loss, disturbance, or any lowering of the ecological value of such areas.

Table 4.28

Summary of Major Projects Adjacent to the Spur Line

Project description and potential impacts	Commencement Date	Completion Date
Shenzhen River Training Phase III involves the realignment of the Shenzhen River to improve drainage efficiency and prevent flooding. Impacts include direct habitat loss and construction phase disturbance	Mid 2001	Late 2004
San Tin Eastern Main Drainage Channels – drainage improvement works to alleviate flooding in the San Tin area and provide flood storage ponds for Chau Tau and Pun Uk Tsuen. Impacts include direct habitat loss, fragmentation, and construction phase disturbance	Undecided	Undecided
Fanling, Sheung Shui & Hinterland Main Drainage Channels – drainage improvement works to alleviate flooding in the Fanling, Sheung Shui & Hinterland areas. Impacts include direct habitat loss, fragmentation, construction disturbance, and possible indirect loss of wetland habitat through lowering of water tables.	1999	2001
Northwest New Territories Planning & Development Study on Northwest New Territories involves examination of the scope and feasibility of accommodating strategic growth development needs in the NWNT. Potential impacts include habitat loss, fragmentation, construction and operation phase disturbance.	After 2000	-
Northeast New Territories Planning & Development Study on Northeast New Territories involves examination of the scope and feasibility of accommodating strategic growth development needs in the NENT. Potential impacts include habitat loss, fragmentation, construction and operation phase disturbance.	After 2000	-
Lok Ma Chau Control Point Expansion Project – kiosk expansion and works to improve vehicle and passenger throughput and circulation. Impacts include direct habitat loss, construction phase disturbance	Mid 1999	2003

Table 4.29

Initial assessment of potential ecological impacts on forest habitats

from the operation of the Spur Line

Potential impact	Location	Type	Evaluation of impact
			Species of Conservation Importance	Size / Abundance	Duration	Reversibility	Magnitude
Habitat loss	On-site	Direct	Few species affected	Low	Permanent	Irreversible	Considerable ecological change
Mortality from collisions with trains, buildings and other structures	On-site	Direct	Few species affected	Whole alignment and associated buildings potential hazard	Permanent	Irreversible	Very low impact
Habitat fragmentation	Off-site	Indirect	Few species affected	Low	Permanent	Reversible	Low impact
Disturbance of wildlife by noise and visual movement	Off-site	Indirect	Few affected	Not quantified but very low	Permanent	Reversible	Low impact
Pollution from toxic chemicals	On-site & Off-site	Direct & indirect	Potentially many species could be affected	Potentially a large area could be affected	Potentially long-lasting	Dependent on chemicals involved	Potentially high impact
Hydrological disruption	On-site	Direct	None directly affected	Not quantified but probably very low	Permanent	Mostly reversible	Very low impact

Table 4.30

Initial assessment of potential ecological impacts on agricultural habitats

from the operation of the Spur Line

Potential impact	Location	Type	Evaluation of impact
			Species of Conservation Importance	Size / Abundance	Duration	Reversibility	Magnitude
Habitat loss	On-site	Direct	Many wet and dry agricultural land species affected	Low	Permanent	Irreversible	Considerable ecological change
Mortality from collisions with trains, buildings and other structures	On-site	Direct	Potentially many Species of Conservation Importance	Whole alignment and associated buildings	Permanent	Irreversible	Very low impact
Habitat fragmentation	Off-site	Indirect	Many wet and dry agricultural land species affected	Low	Permanent	Reversible	Low impact (track on viaduct)
Disturbance of wildlife by noise and visual movement	Off-site	Indirect	Many wet and dry agricultural land species affected	Moderate	Permanent	Reversible	Moderate impact
Pollution from toxic chemicals	On-site & Off-site	Direct & indirect	Potentially many species could be affected	Potentially a large area could be affected	Potentially long-lasting	Dependent on chemicals involved	Potentially high impact
Hydrological disruption	On-site & off-site	Direct	None directly affected	Not quantified but probably very low	Permanent	Mostly reversible	Very low impact

Table 4.31

Initial assessment of potential ecological impacts on fishpond and pond habitats

from the operation of the Spur Line

Potential impact	Location	Type	Evaluation of impact
			Species of Conservation Importance	Size / Abundance	Duration	Reversibility	Magnitude
Habitat loss	On-site	Direct	Many waterbird species affected	High (9.6 ha)	Permanent	Irreversible	Considerable ecological change
Mortality from collisions with trains, buildings and other structures	On-site	Direct	Potentially many Species of Conservation Importance	Track and station buildings potential hazard	Permanent	Irreversible	Low impact
Habitat fragmentation	Off-site	Indirect	Many waterbird species affected	Moderate	Permanent	Reversible	Moderate impact
Disturbance of wildlife by noise and visual movement	Off-site	Indirect	Many waterbird species affected	Large area affected	Permanent	Reversible	High impact
Pollution from toxic chemicals	On-site & Off-site	Direct & indirect	Potentially many species could be affected	Potentially a large area could be affected	Potentially long-lasting	Dependent on chemicals involved	Potentially high impact
Hydrological disruption	On-site & off-site	Direct	None directly affected	Not quantified but probably very low	Permanent	Mostly reversible	Very low impact

Table 4.32

Initial assessment of potential ecological impacts on marsh habitats

from the operation of the Spur Line

Potential impact	Location	Type	Evaluation of impact
			Species of Conservation Importance	Size / abundance	Duration	Reversibility	Magnitude
Habitat loss	On-site	Direct	Many species affected including species almost entirely dependent upon this habitat e.g. Greater Painted-snipe	Low	Permanent	Irreversible	Considerable ecological change
Mortality from collisions with trains, buildings and other structures	On-site	Direct	Potentially many Species of Conservation Importance	Whole alignment and associated buildings potential hazard	Permanent	Irreversible	Very low impact
Habitat fragmentation	Off-site	Indirect	Many marshland species affected	Low	Permanent	Reversible	Low impact (track on viaduct)
Disturbance of wildlife by noise and visual movement	Off-site	Indirect	Many marshland species affected	Moderate	Permanent	Reversible	Moderate impact
Pollution from toxic chemicals	On-site & Off-site	Direct & indirect	Potentially many species could be affected	Potentially a large area could be affected	Potentially long-lasting	Dependent on chemicals involved	Potentially high impact
Hydrological disruption	On-site & off-site	Direct	None directly affected	Not quantified but probably very low	Permanent	Mostly reversible	Very low impact

4.7 Mitigation Measures

Habitat loss and disturbance

4.7.1 As described previously, the impacts of disturbance, though to a lesser extent, are analogous to habitat loss and therefore mitigation measures for these are treated together here. As summarised earlier, the main area of ecological importance that is subject to high levels of permanent habitat loss and disturbance impacts are the fishponds around the station complex at San Tin / Lok Ma Chau. The agricultural and marshland habitats of Long Valley are subject to moderate levels of disturbance and low levels of permanent habitat loss.

Avoidance

4.7.2 The Spur Line alignment has been selected based on engineering, operational safety and environmental grounds. The basis for the avoidance of tunnelled sections and the selection of a combination of at grade, viaduct, cutting and embankment sections has been described in Chapter 2 of this report. Various options of the alignment were examined in detail from the environmental point of view, in the PPFS and the selected route minimises impacts on ecologically sensitive areas, while remaining within the constraints of other engineering and planning requirements.

Minimization

4.7.3 The station building and footbridge linking it to Huanggang Station should be designed in a manner that keeps disturbance of wildlife to an absolute minimum, including both noise and visual disturbance. Large areas of reflective material (including glass) should not be used on the outer surfaces of the building, as this in known to result in aerial collisions from birds. In terms of landscaping, little can be done to disguise the station from wildlife. The station height has been minimised within the constraints of the required function of the building.

4.7.4 The design of the prier footings of the footbridge should be carried out so as to minimize additional turbulence which may increase turbidity. Appropriate design will help prevent adverse impacts on the ecological resources of Deep Bay. Working practices using concrete should minimize the release of concrete washings into waterbodies, to prevent adverse impacts on downstream water quality and ecology.

4.7.5 The noise reduction system which will be used on the Spur Line comprises an advanced multi-plenum structure which will absorb noise generated from the train and the track. A sophisticated viaduct design provides additional mitigation in the reduction of noise from vibration. These measures are described in Chapter 3. The mitigation measures proposed will enable the ANL of 55 dB(A) to be met in rural areas of Long Valley at locations close to the viaduct. By incorporating such a noise reduction system for human protection, the impacts from noise to wildlife should therefore also be minimised.

4.7.6 During construction, a haul road will be established along the alignment to transport materials and equipment for the Spur Line construction. Limiting the haul road to the south side of the alignment and restricting access of the contracting staff to the area immediately around the viaduct supports will minimize the potential damage to the habitat, particularly in the Long Valley area. The marsh area in the centre of Long Valley is of particular concern and must be protected to the maximum possible extent by further limiting contractor access to the area.

Habitat compensation

4.7.7 As avoidance and minimisation measures are unable to fully mitigate for habitat loss and disturbance impacts then it is necessary to offset potential residual impacts through habitat compensation. The aim of habitat compensation will be to replace habitats of intrinsic ecological value and ecological functions for Species of Conservation Importance that will be lost or degraded. Compensation may be carried out through the enhancement of existing habitats (i.e. by raising the ecological value of the habitat and thereby its carrying capacity for target species) or by restoration or creation. As the only significant potential residual impacts will relate to habitat loss and disturbance in the area of fishponds around the station complex at San Tin / Lok Ma Chau and the agricultural and marshland habitats of Long Valley then compensation will be targeted towards these locations and habitats and their associated Species of Conservation Importance.

4.7.8 Due to limitations on the potential for resuming land for ecological mitigation purposes then most habitat compensation will be carried out through off-site measures. This is, however, desirable as on-site compensation areas would be subject to disturbance from the operation of the railway line. Nevertheless, where there are opportunities for on-site compensation (such as alongside and underneath the viaduct sections) these shall be used for creating habitats for species that are not susceptible to disturbance impacts.

4.7.9 Sites being used for off-site compensation will be either alongside areas subject to habitat loss (such as the San Tin / Lok Ma Chau station area) or are within the same contiguous wetland block (i.e. within Long Valley).

4.7.10 As described in the Implementation Schedule (Chapter 12) it is important that habitat mitigation measures are carried out as soon as possible. This is necessary to allow vegetation communities and associated animal communities to become established so that suitable habitat is available for displaced species when works commence. In particular it is recommended that the construction mitigation area, that is mainly for the compensation of the loss of habitat for Greater Painted-snipe, is completed in advance of the start of works in the Long Valley marsh area. The habitat mitigation measures at Lok Ma Chau around the station site should be phased so that overall carrying capacity for Species of Conservation Importance is maintained throughout the construction period. The measures to enhance the ponds at the western section of the proposed mitigation area should be concurrent with the start of site formation for the station complex. Those ponds should be capable of providing enhanced feeding opportunities either by the provision of the new shallow margins or by the lowering and management of water levels. This western portion is located more than 500m from the station works area so will be little disturbed during the construction period. Complete mitigation should be in place by the time the station becomes operational.

4.7.11 General principles for habitat compensation and preliminary designs for each site are outlined below. Final designs, establishment methods and long-term management plans will be produced as part of the detailed design process for the construction of the railway. These designs will take into account key factors affecting the ecological value of wetland habitats and their suitability for associated species, including:

· Topography

· Soil types

· Hydrology (i.e. water budgets indicating anticipated direct rainfall, groundwater and surface water inputs and losses, seasonal fluctuations and periodicity in surface water levels and soil watertables, and the need for additional water supplies, water courses and water level control structures)

· Water quality with respect to target vegetation and associated species

· Establishment and management of wetland plant communities and animals

· Disturbance

4.7.12 Whilst there are some similarities between the Lok Ma Chau / San Tin and the Long Valley areas, their different ecological characteristics and wildlife communities, as well as different patterns of land use and the different design of the proposed Spur Line at the two sites, necessitates a different approach to mitigation. Accordingly, these areas are described separately below.

Compensation for the loss and disturbance of fishpond habitats in the Lok Ma Chau / San Tin area

4.7.13 As indicated in Table 4.22 the direct loss of fishponds in Lok Ma Chau Station area amounts to 9.5 ha. In addition equivalent habitat loss will occur through impacts of disturbance (see Section 4.5). Whilst disturbance impacts affect all species, analysis of the susceptibility of Species of Conservation Importance to disturbance (as described above) demonstrates that the species most susceptible to disturbance are two species of large birds of prey, Imperial Eagle and Greater Spotted Eagle and three species of large waterbird, Great Cormorant, Grey Heron and Great Egret.

4.7.14 Of these species, the two large birds of prey range widely in the Deep Bay area and are known to exhibit marked avoidance of man-made structures. Calculation of disturbance impacts, and in particular areas of reduced density is, therefore, somewhat problematic. However, the extensive type range of these birds indicates that the loss of this fishpond habitat is therefore a small proportion of their overall range and impacts would consequently probably be small.

4.7.15 It is therefore more realistic to quantify the disturbance effects on the three large waterbirds, about which more information is available on their habitat requirements and behaviour. Furthermore, if the requirement to compensate for the disturbance to these species is met, full compensation for the impacts of direct habitat loss and disturbance to less-sensitive or unaffected Species of Conservation Importance is likely to be achieved.

4.7.16 Calculation of the overall land requirement to compensate for direct habitat loss and disturbance effects for the large waterbirds is derived as shown in Table 4.33. Disturbance from existing sources such as Lok Ma Chau Boundary Crossing were eliminated from the calculations. Estimates of required compensation factors for additional disturbance impacts are based on the individual species assessment of sensitivity to disturbance discussed earlier and takes into account areas that are already subject to existing disturbance within the station complex area, e.g. the Boundary Crossing (Figure 4.3)

Table 4.33

Land requirement to compensate for habitat loss for large waterbirds

Species	Land requirement to compensate for direct habitat loss (ha)	Additional compensation for habitat loss for additional disturbance impacts	Total land requirement (ha)
Great Cormorant	8.5 (fish pond areas)	c. 2 X direct habitat loss	c. 25.5 ha
Great Egret	9.5 (fish ponds and bunds)	c. 1.5 X direct habitat loss	c. 23.8 ha
Grey Heron	9.5 (fish ponds and bunds)	c. 2 X direct habitat loss	c. 28.5 ha

4.7.17 Accordingly, it is proposed that a compensatory wetland c. 28.5 ha in extent would provide full habitat compensation for these large waterbird species and all other Species of Conservation Importance except for Greater Spotted Eagle and Imperial Eagle. As discussed above, these last two species are wide ranging raptors, wintering in the Deep Bay area and occupying very extensive home ranges, but at low density. The impact resulting from loss of fishpond under this project will result in a relatively small direct loss of their habitat. It is therefore considered unlikely that this residual impact will affect the overall population of these species in Hong Kong.

4.7.18 The area of approximately 28.5 ha has been derived from a consideration of compensation requirements for the species shown in Table 4.33. Compensation requirements for other species of conservation importance will be more than compensated within this defined area in accordance with the precautionary principle quoted in TPB PG No. 12B (1999).

4.7.19 The compensation area of approximately 28.5 ha is based on the fact that the existing habitat is already of ecological value. This area was calculated based on a projected enhancement potential derived from a knowledge of the particular species concerned and previous experience of wetland enhancement in Hong Kong and elsewhere. It should also be noted that these wetlands are already impacted by disturbance from the Boundary Crossing. The potential for mitigation also takes into account further potential impacts under other projects (San Tin Main Drainage Channel (ERM 1999a) and expansion of the Lok Ma Chau Boundary Crossing (Binnie 1999)).

4.7.20 A range of options for wetland enhancement have been investigated and considered in respect of the following criteria:

• Technical feasibility for provision in the context of physical features of the area.

• Compatibility with existing land uses, especially maintenance of existing fish farming.

• Potential carrying capacity for target Species of Conservation Importance of different wetland habitat types.

• Potential scope for provision for other wetland Species of Conservation Importance.

• Compatibility with habitat creation requirements for other purposes (e.g. screening).

• Requirement to avoid areas adversely impacted by existing or proposed developments.

4.7.21 Based on these criteria, it is considered that the most appropriate design for compensatory provision is the enhancement of fish ponds. Currently the large biogeographically important populations of herons and egrets that use fishponds in the Deep Bay area primarily feed on the abundant small non-commercial fish and invertebrates (termed ‘trash fish’) that thrive as a by-product of the highly productive commercial fish-farming systems. These include Gambusia affinis (Mosquito Fish), Macrobrachium nipponense (a prawn) and Oreochromis mossambicus (a species of Tilapia).

4.7.22 However, these food resources are generally only readily available to birds when the ponds are drained down for fish harvesting during the winter. Furthermore, only a small proportion of fish ponds are drained at one time and only for short periods. The spatial distribution of birds feeding on ponds is therefore highly dynamic and variable as birds move between fish ponds as they are drained. When recently drained, such fish ponds may contain many hundreds of feeding egrets, herons and other waterbirds. Consequently, a large area of fish ponds is required to support the bird populations and avoid ‘feeding bottlenecks’.

4.7.23 When ponds are full, their use by birds is severely limited due to their relatively steep sides, deep water and their frequent lack of marginal vegetation. These characteristics also limit their biodiversity interest for other taxa groups. The main objective of enhancing fish ponds would therefore be to:

• Increase the value of fish ponds to herons and egrets outside harvesting periods (i.e. draw-down), by increasing food resources and food availability and by reducing disturbance effects. Enhancement of the value of fish ponds to such birds outside harvest periods could reduce the potential for ‘feeding bottlenecks’ thereby possibly reducing the area of fish ponds needed to support the population.

• Increase their overall biodiversity value and suitability for other non-bird Species of Conservation Importance, such as some mammals, amphibians and reptiles, whilst maintaining their current important functions for herons, egrets and other water birds.

• Maintain the commercial viability of fish farming.

4.7.24 Figure 4.5 indicates the location of the proposed area of enhanced fishponds. These would be placed away from the buildings to minimise disturbance impacts from the railway station and others sources such as the Lok Ma Chau border crossing. They would also be contiguous with the main area of fishponds in the WCA and Ramsar site as a whole. Maintaining a contiguous area for compensation which is linked with an existing area of recognised conservation importance is of significant ecological value. As indicated in Appendix G these ponds are also regularly used by waterbirds at present.

Design proposals for enhanced fish pond habitats at San Tin / Lok Ma Chau

Habitat requirements of target species

4.7.25 The key habitat requirements of target species of Conservation Importance that occur within the San Tin / Lok Ma Chau area are summarised in Table 4.34 below.

Table 4.34

Summary of habitat requirements for target Species of Conservation Importance

Key: habitat important for F = Foraging; R = Roosting; B = Breeding.

Species	Deep water	Shallow water	Muddy vegetation free margins	Reedbeds	Other emergent / marsh vegetation	Bare or sparsely vegetated bunds
Amphibians		F,B	F,B	F,B	F,B
Reptiles		F	F	F	F	F,B
Mammals	F	F	F	F	F	F
Greater Spotted Eagle	F	F			F	F
Imperial Eagle	F	F			F	F
Japanese Yellow Bunting				F,R	F,R	F
Schrenck's Bittern		F			F,R
Red-billed Starling			F		F	F
Great Cormorant	F
Chinese Pond Heron		F		F,R	F
Great Egret		F			F,R
Little Egret		F			F
Common Teal	F	F		F,R
Eurasian Coot	F	F
Black-winged Stilt		F	F
Grey Heron		F			F
Northern Hobby
Pheasant-tailed Jacana		F			F
Watercock		F			F
Black-naped Oriole
Common Snipe			F		F,R
Pintail Snipe			F		F,R
Swinhoe's Snipe			F		F,R
Richard's Pipit			F			F
Bluethroat			F	F,R	F,R
Common Stonechat				F,R	F,R	F
Zitting Cisticola				F,R	F,R
Chestnut Bittern		F			F,R
Pallas's Grasshopper Warbler				F,R	F,R

4.7.26 To provide these habitat requirements for target species and to meet the broad objectives for the enhancement of fish ponds it is proposed that main mitigation measures would be:

• Enlargement of small fish ponds to reduce enclosure effects (as small ponds are avoided by many birds)

• Re-profiling of fish pond bunds and bottoms to provide shallow sloping margins and a variable bottom topography that provides increased structural diversity. Shallow sloping margins will also increase feeding opportunities and the availability of fish and invertebrate prey to birds.

• Establishment of marginal emergent vegetation, including reedbeds and other species, to support and provide cover for invertebrates, amphibians, reptiles, passerine birds and mammals. Such vegetation may also provide screening of disturbance sources from feeding herons and egrets.

• Reductions in water depth during the winter, i.e. when deoxygenation problems are unlikely to occur, to increase the availability of fish and invertebrates to wading birds. This is considered to be a potentially very important enhancement. Observations of some fish ponds in the Deep Bay area by members of the study team have revealed that some abandoned shallow ponds are frequently used for feeding by various waders, herons and egrets, often including substantial numbers of the globally threatened Black-faced Spoonbill.

• Manipulation of the fish stocking and feeding / fertiliser regime to optimise the food availability for birds whilst maintaining commercially viable fish farming.

Construction works

4.7.27 As the area already consists of fish ponds it is envisaged that works would be restricted to the modification of existing ponds, rather than construction of new ponds. These modifications will consist of the joining of selected ponds to reduce enclosure effects (see above). Although the exact layout of ponds will be decided upon during the detailed design stage, after further baseline assessments of habitat conditions, it is suggested that the eventual pond layout may be as indicated in Figure 4.6.

4.7.28 Reprofiling of the selected bunds to provide shallow sloping margins within the range of annual water level variation will also be carried out, as indicated in Figure 4.7. If necessary, the bottoms of each pond will also be reprofiled to provide variations in depth during draw down periods for harvesting. This will be based on a simple ridge and furrow approach as indicated in Figure 4.7 appropriately adapted to the individual dimensions of each pond.

Hydrology

4.7.29 As described in Chapter 10 on fisheries, no large river systems with unpolluted water or large quantities of groundwater that may supply fish ponds with clean water exist in Hong Kong. Hence fishponds are dependent on rain water for filling and are effectively self-contained entities, which are not subjected to the dynamic water budgets of other open system wetland habitats. Water may be lost from the system due to evaporation, evapo-transpiration and filtration through the substrate. However, these losses are considered to be sufficiently low to maintain water levels in fish ponds within acceptable fluctuations, as such ponds are present in the area now and are currently being successfully farmed without water supply problems.

4.7.30 Saline intrusions are also not considered to be a problem. As a component of environmental monitoring undertaken for the Fish Pond Study, the potential of saline intrusion was investigated due to ponds close proximity to the coastal area of Inner Deep Bay. However, no saline intrusion was found (Aspinwall & Co, 1997).

4.7.31 The traditional system of using rainfall to fill ponds and transferring water between ponds during harvests is therefore considered to be adequate for the maintenance of this habitat. Although water levels will change with the season due to rainfall and variable evaporation /evapotranspiration rates these changes are considered to be sufficiently small to be advantageous, through the exposure of muddy margins as water levels slowly fall. Large changes that would be detrimental to wetland vegetation are unlikely to occur.

Vegetation establishment

4.7.32 As it is not intended that new bunds will be created between the ponds, the establishment of new vegetation communities will not be necessary. However, it is intended that the existing vegetation on the bunds will be examined during the baseline ecological surveys for the detailed design stage. Requirements for selective plantings of additional desirable species on the fish pond bunds will then be ascertained.

Management

4.7.33 It is proposed that the main management enhancement of the ponds would be to regulate and extend the period of draw-down for harvesting. Under current fish-farming practices harvesting is carried out in winter by reducing the water depth of the pond (by pumping water to another pond for storage) and the gradual netting of the stock over a couple of weeks (see Chapter 10). However, the precise dates of harvesting are dependent on market prices and this can result in food-resource ‘bottle-necks’ if no or few ponds are drained at any one time. The first benefit of regulating drain-down periods in the mitigation area is that it can be timed to allow for staggered food availability, particularly at times when there may be low food availability in other areas of Deep Bay (based on current knowledge).

4.7.34 The second benefit would be ensuring that draw-down periods are for a fixed time of provisionally 20 days. This is longer than normal and would thus further increase feeding opportunities thereby further reducing potential feeding bottle-necks. Although food resources would be depleted with time it will maximise the benefit of trash fish food availability in each pond. The ponds would then be refilled in time for the normal fish stocking period. It is therefore unlikely that any significant detrimental effects on fish production or financial viability of fish farming would occur.

4.7.35 It is also proposed that at any one time a couple of the larger ponds would be taken out of production (on a rotational basis) for two seasons. This would provide a number of more natural 'lake' like habitats thereby increasing habitat diversity within the area. Such habitats would be used by targets species such as herons and egrets, but would be particularly valuable habitats for ducks and non-bird taxa that are susceptible to the highly eutrophic conditions in managed fish ponds. Recent conversion of some fishponds to similar freshwater habitats within the Mai Po Nature Reserve have produced highly valuable habitats which hold large numbers of wintering waterfowl.

4.7.36 The optimal management of fish ponds for biodiversity is the subject of field trials being carried out under the current AFD Wetland Compensation Study. These trials include investigations of the biodiversity benefits and implications on fish farming of changing fish stocking levels, fertilizer and fish food inputs and the extension of winter draw-down periods for harvesting. These trials will be completed in the spring of 2001 and their results will be taken into account in the finalisation of the long-term management regime for the enhanced fish ponds within the mitigation area.

4.7.37 In addition to the management of the fish farming system, some management of the bund vegetation would also be undertaken. This would mainly involve the control of tall vegetation such as reeds. Although some patches of reed would be encouraged to develop for cover and screening purposes (see Figure 4.6) these would need to be controlled. Ponds that are enclosed by tall vegetation tend to be avoided by many of the larger herons and egrets. Management would also be carried out of undesirable invasive weeds if necessary.

Design proposals for marshland habitats at San Tin / Lok Ma Chau

4.7.38 The fishpond enhancement measures described above would be supported by the creation of some areas of reedbed and marshland habitat alongside the station buildings (Figure 4.8). Reedbeds will be established alongside the station complex to function as a barrier to disturbance and as shelter/cover for smaller herons, crakes, passerines, amphibians, reptiles and mammals. Additional marshland habitat would have a high carrying capacity for several such target Species of Conservation Importance as well as providing habitat for some that at present occur in the area only in small numbers (see Table 4.4 above).

4.7.39 An additional area of reedbed will also be created alongside the station for water clean-up purposes (see Chapter 6). This habitat will provide additional habitat for reedbed species, potentially including smaller herons, bitterns and various passerines.

Construction works

4.7.40 The reedbed and marsh area would principally consist of a shallow basin sloping gradually away from the station buildings, to produce a zoned wetland with drier terrestrial habitats on the upper slopes and permanent wet reeds beds and pools at the lower end (Figure 4.8). Construction works outside the station boundary would, therefore, be minimal and would primarily consist of re-profiling of adjoining fish pond bunds and the land in between. Superimposed on the slope would be a ridge and furrow profile perpendicular to the slope away from the station buildings. This would produce further structural diversity within the wetland leading to a scalloped edging to the vegetation zones. Some small depressions of 0.5 – 1.0 m greater depth than the adjoining substrate would also be made to provide additional permanent and temporary pools within the upper marsh / reedbed areas.

4.7.41 It is recommended that implementation of construction works for the compensation area more than 500 m from the station works area i.e. beyond the area of disturbance for most species, be carried out at the start of site formation for the station complex. Compensation could be carried out through formation of shallows by construction or management of water levels.

Hydrology

4.7.42 There is no watercourse with water that would be of a suitable quality for the maintenance of reedbed and marshland habitats for biodiversity purposes. Also ground water within the area is likely to be saline. Therefore it is envisaged that the reedbed / marshland habitats would need to be primarily maintained by rainfall. However, a preliminary water budget for the area suggests that direct rainfall would be insufficient to maintain areas of permanent wetland across most of the mitigation area (see Table 4.35). It is therefore proposed that additional rainfall would be obtained from intercepting some of the runoff from the station roofing area. This would provide sufficient additional water to maintain a reedbed and marshland with standing water with an average depth of 0.5 m when full (as controlled by a drop-gate weir) during the wet season. In addition, the reedbed area to the east of the station will be fed by the flow of treated sewage effluent from the sewage treatment plant in the Lok Ma Chau Station. It is not proposed to pass any of this effluent through the main marsh area, to avoid build up of pollutants in the marsh.

Table 4.35

Predicted water budget and waterlevels for the marsh/reedbed wetland mitigation area at the San Tin / Lok Ma Chau station

RA Area of roof supplying runoff water = 2.4

MA. Area of marsh vegetation (excluding water treatment reedbed) within mitigation area (ha) = 3.5

ECF Correction factor for evapotranspiration of wetland vegetation = 1.4

RC Runoff coefficient for station roof = 0.95

AWD Average water depth (mm) when full (weir level) at end of wet season in August = 500 mm

Marshland requirements

A.	B.	C.		D = C x ECF	E = B - D
Month	Rainfall	Evapo-transpiration		Marsh evapo-trans	Monthly balance
	mm/month	mm/month		mm/month	mm/month
Jan	20.9	82.8		116	-95
Feb	30.5	75.7		106	-76
March	60.9		88.7	124	-63
April	131.2	101.8		143	-11
May	284.6	131.8		185	100
June	232.0	133.4		187	45
July	387.9	153.1		214	174
Aug	313.3	148.1		207	106
Sept	144.6	136.0		190	-46
Oct	32.0	130.9		183	-151
Nov	48.4	105.8		148	-100
Dec	46.4	88.0		123	-77
Total	1,733	1,376		1,927

Additional supply from station roof			Overall balance and change in levels
A.	B.	J = B – RC	M = E + (J*(RA/MA))	N = AWD(t-1)+M
Month	Rainfall	Monthly balance	Overall balance	Average water depth
	mm/month	mm/month	mm / month	mm
Jan	20.9	19.9	-81	173
Feb	30.5	28.9	-56	118
March	60.9	57.9	-24	94
April	131.2	124.6	74	168
May	284.6	270.4	286	454
June	232.0	220.4	196	500
July	387.9	368.5	426	500
Aug	313.3	297.6	310	500
Sept	144.6	137.4	48	500
Oct	32.0	30.4	-130	370
Nov	48.4	46.0	-68	301
Dec	46.4	44.1	-47	255
Total	1,733	1,646

4.7.43 Additional water from the station complex would be fed through culverts in the station boundary bund and then simply allowed to spill over onto the furrowed and sloping reedbed marshland site. Water would then flow with gravity and collect at the slope bottom alongside the perimeter bund. As the soils are considered to have very low hydraulic conductivities, as indicated by the presence of fish ponds within the area, it is anticipated that water losses to the ground will be minimal.

4.7.44 Water levels within the marsh will be controlled by drop-gate weirs with excess water draining into the water course to the west of the station (Figure 4.8).

4.7.45 This hydrological regime will produce a relatively diverse wetland with a hydrological gradient leading to a wetland zonation from dry non-wetland habitats adjoining the station to permanent wetland with standing water at the slope bottom.

Vegetation establishment

4.7.46 Reedbed will be the principal vegetation within the area, occurring in the zones of permanent standing water and high water table (Figure 4.9). This is a highly invasive and vigorous species in Hong Kong and can be easily established. As the development will lead to the destruction of some reed stands, it is considered that the most appropriate means of establishment will probably be by simply spreading top soils containing rhizomatose material. Some selective planting of seedlings may also be carried out to fill in if establishment is poor in some areas.

4.7.47 Upper areas of the marsh will grade from wetland plant communities, that rely on seasonal inundation, to those which only withstand high water tables, to non-wetland species. Appropriate wetland plants will be selected at the detailed design stage after further baseline studies of soil conditions and vegetation communities currently occurring within the vicinity have been undertaken. Establishment methods will probably be similar as for reeds, with topsoil used if materials from areas with appropriate wetland communities are available.

Management

4.7.48 One of the aims of the design of the wetland area will be to reduce management requirements to a minimum, through the creation of a natural sustainable wetland system. No water pumping or day to day water control will be necessary, with water regimes largely following seasonal cycles.

4.7.49 However, some management will be needed to avoid excessive vegetation growth as some open areas of habitat are important for many target Species of Conservation Importance (see Table 4.34 above). Also, with time, management will be needed to arrest succession processes and the loss of wetland habitats to terrestrial ecosystems. Ideally much vegetation control and the maintenance of valuable areas of open mud and water can be carried out by the use of carefully managed grazing animals, such as water buffalo or domestic cattle. This possibility will therefore be examined as part of the detailed design process. If this is not possible, vegetation will be controlled by selective cutting and removal.

4.7.50 Reedbeds will be cut in compartments on a rotational basis, at varying frequencies (to increase habitat diversity), to maintain condition and vigour of the vegetation. Reed encroachment on areas of open water will be controlled by selective spraying of low toxicity, non-persistent herbicides and occasional mechanical removal of rhizomatous material.

Overall residual impacts from habitat loss and disturbance at San Tin/Lok Ma Chau

4.7.51 It is considered that by a combination of fish pond enhancement and the creation of a more intensively managed reedbed and marshland area it is realistic to expect an increase in the carrying capacity of the current fishpond habitat. The extent of this increase will depend on the existing state of the fishpond as an ecological habitat, and the target species being provided for. It is expected that approximately 28.5ha of existing ponds which will be provided can be significantly enhanced, so long as a significant portion of this area is not impacted by other developments.

4.7.52 The proposed areas and land uses of this proposed provision are shown in Table 4.36. This indicates that over most of the habitat compensation area it is intended that habitat management of the enhanced fishpond area will depend on the active continuation of fish farming, albeit constrained to some extent.

Table 4.36

Land requirement for compensatory habitat provision

and other creation of habitats of ecological value

Land use	Land area (ha)	Notes
Enhanced fish ponds	28.5	Subject to management regime discussed above
Marshland	3.5	Wetland area around station
Total compensation areas for habitat loss	32.0
Reedbed	2.0	Required for wastewater polishing
Bamboo / tree / shrub / reedbed / planting	1.0	Required for screening of station complex
Other habitat of ecological value	3.0	Not included in compensatory habitat provision

4.7.53 The area of compensatory habitat provision of approximately 28.5 ha allows the creation of a coherent management unit of ponds to the west of the station site and ensures that the compensation area comprises entire ponds.

4.7.54 The reedbed area around the station will be used for screening and part of the area will be used for polishing of wastewater from the Lok Ma Chau station. In this context, it is recommended that the reedbed be established during the construction of the sewage treatment plant so that it is established before commissioning of the sewage treatment system.

4.7.55 The overall levels of compensation predicted from the above proposed compensation measures for Species of Conservation Importance that regularly occur or have the potential to regularly occur in the San Tin / Lok Ma Chau fishpond area are summarised in Table 4.37.

Table 4.37

The overall levels of compensation predicted from compensation measures for Species of Conservation Importance that regularly occur or have the potential to regularly occur in the San Tin / Lok Ma Chau fishpond area.

Species	Sensitivity to disturbance	Compensation achieved
Amphibians	Nil	Scope for additional habitat provision
Reptiles	Low	Scope for additional habitat provision
Mammals	Moderate	Fully compensated
Greater Spotted Eagle	Very large	Residual impacts expected to be small*
Imperial Eagle	Very large	Residual impacts expected to be small*
Japanese Yellow Bunting	Small	Fully compensated
Schrenck's Bittern	Small	Scope for additional habitat provision
Red-billed Starling	Small-medium	Fully compensated
Great Cormorant	Large	Fully compensated (Baseline species for compensation proposals)
Chinese Pond Heron	Medium	Fully compensated
Great Egret	Large	Fully compensated (Baseline species for compensation proposals)
Little Egret	Medium-large	Fully compensated
Common Teal	Medium-large	Fully compensated
Eurasian Coot	Medium	Fully compensated
Black-winged Stilt	Medium-large	Fully compensated
Grey Heron	Large	Fully compensated (Baseline species for compensation proposals)
Northern Hobby	Large	Fully compensated
Pheasant-tailed Jacana	Medium-large	Scope for additional habitat provision
Watercock	Medium	Scope for additional habitat provision
Black-naped Oriole	Medium	Fully compensated
Common Snipe	Medium	Scope for additional habitat provision
Pintail Snipe	Medium	Scope for additional habitat provision
Swinhoe's Snipe	Medium	Scope for additional habitat provision
Richard's Pipit	Small	Fully compensated
Bluethroat	Small	Scope for additional habitat provision
Common Stonechat	Small	Fully compensated
Zitting Cisticola	Small	Fully compensated
Chestnut Bittern	Small	Scope for additional habitat provision
Pallas's Grasshopper Warbler	Very small	Fully compensated

Note: Those species for which scope for additional habitat provision is indicated above, at present occur only in very small numbers or irregularly due to absence of suitable habitat. Provision of appropriate wetland habitat would significantly increase numbers of these Species of Conservation Importance at this site.

* The level of compensation would require further detailed studies of the habitat needs and range of these species.

Compensation for the loss and disturbance of agricultural and marshland habitats in Long Valley

4.7.56 Compensation for the loss and disturbance of agricultural and marshland habitats in Long Valley should be targeted to a wide range of fauna, including butterflies, amphibians, reptiles, birds and mammals that are Species of Conservation Importance. Such measures should also aim to maintain and enhance the overall biodiversity of the area, in particular through restoration of natural marshland habitats. These measures should include:

• Creation of marshland habitat in advance of commencement of the works as compensation for areas which will be adversely affected.

It is highly recommended that a temporary marshland compensation area be established ahead of construction activities in the Long Valley marsh. This is necessary to mitigate potentially adverse construction impacts within this special habitat.

This compensation area will provide refuge beyond construction disturbance.

• Marshland habitat should be created to mitigate for the loss of wet agricultural land.

Provided that enhancement management is undertaken to compensate for the loss of wet agriculture, then creation of a more ‘natural’ marsh type habitat is a more feasible and sustainable option than the formation of additional wet agricultural areas. Created marshland areas can be designed to allow for an increased carrying capacity for key species adversely affected by loss or disturbance of areas that are currently wet agricultural land. Any such habitat creation should be outside of the area to be affected by disturbance during the construction and operational phases of the works.

• The hydrological regime of the areas of wet agriculture and marshland must be maintained during construction and during operation.

• All access roads, other ground level infrastructure, temporary storage areas, etc. must avoid areas of marsh, and wet agriculture, and must not affect the hydrology of these habitats.

4.7.57 Using the same methodology described for calculation of impacted areas for Lok Ma Chau station and rail viaduct, the direct loss of habitat beneath the viaduct in Long Valley amounts to 0.85 ha for disturbance-sensitive species such as Little Egret, Common Teal, Northern Hobby and Pheasant-tailed Jacana. These species will also suffer disturbance impact losses of 0.85 ha, resulting in a total impact of 1.7 ha (Table 4.38).

Table 4.38

Land requirement to compensate for habitat loss for

Long Valley species sensitive to disturbance

Species	Land requirement to compensate for divert habitat loss (ha)	Additional compensation for habitat loss for additional disturbance impacts (ha)	Total land requirement (ha)
Little Egret	0.85	1 x direct habitat loss	1.7
Common Teal	0.85	1 x direct habitat loss	1.7
Northern Hobby	0.85	1 x direct habitat loss	1.7
Pheasant-tailed Jacana	0.85	1 x direct habitat loss	1.7

4.7.58 During the construction period, disturbance impacts to all species will be mitigated through temporary creation of suitable habitat in an area of fishponds to the west of the River Beas (Figures 4.10 and 4.11).

4.7.59 Advance creation of marshland habitat is highly recommended to maintain the viability of the Greater Painted-snipe population in Long Valley. This population has recently been shown to be one of only two remaining in Hong Kong, the other being at Kam Tin (Leader 1999). The temporary occupation of this area to provide continuity of habitat for Greater Painted-snipe during the construction process is the only requirement for a Temporary Occupation Area (TOA) for purposes of ecological mitigation. After the completion of construction, this area will be handed back to its owner, reinstated to its original state. All other ecological mitigation is to be provided on land that will permanently be wetland.

4.7.60 Long-term impacts of habitat loss for the majority of Species of Conservation Importance will be compensated by the provision of an area of approximately 2 ha of wetland occupying the strip of land beneath the viaduct and 10 m either side (within the railway reserve).

4.7.61 Whilst, as is detailed in Table 4.39, the mitigation area of 2.4ha to be provided within the railway reserve fulfills the requirement to create an area of compensatory wetland of at least 1.7ha (the requirement detailed in paragraph 4.7.55 and Table 4.38) it is acknowledged that, for some species, the noise impact from trains and direct avoidance of the viaduct structure is such that the habitat to be provided within the railway reserve will not compensate in functional terms for the loss. Accordingly, it is necessary to create an area of compensation wetland which is unaffected by these disturbance impacts. Interpretation of the "public purpose" by Justice Department does not, at the present time, permit acquisition of private land under the Railways Ordinance for purposes of mitigation of adverse environmental impacts. Thus, it is necessary to seek Government land for this purpose. In this respect, three areas of land acquired by Government in connection with the Main Drainage Channel Scheme for the River Beas and River Indus have been identified (Figure 4.12). These areas comprise Meanders 1 and 2 of the Beas River described in Main Drainage Channels for Fanling, Sheung Shui and Hinterland EIA Final Report (CES 1997) together with an area originally proposed as a Temporary Occupation Area (TOA) for this project, but actually resumed, also described in this Report.

Table 4.39

Land requirement for compensatory habitat provision

and other creation of habitats of ecological value

Land use	Land area (ha)	Notes
Temporary mitigation area for construction disturbance to Long Valley marsh	1.8	Two fishponds will be temporarily changed to marsh to provide refuge areas for sensitive species during construction stage in Long Valley marsh.
Total temporary ecological mitigation area	1.8
Permanent wetland beneath viaduct	2.4	Compensation for loss of habitat from presence of viaduct for non-sensitive species.
Enhancement of meanders along River Beas	1.4	Compensation for loss of habitat and disturbance impacts due to presence of viaduct, for disturbance sensitive species.
Total permanent ecological mitigation area	3.8

4.7.62 Meander 1 was proposed by CES (1997) to include a pool formed by the abandoned stream channel, with the entire land area to be planted with trees. This meander was listed as a "Type 2" meander. According to CES (1997) a Type 2 meander would have no tributary input (i.e. no permanent water source) and would be redesigned so as not to include retained river habitat. "Marshland" would be "recreated…wherever practicable." On reviewing the proposal for this meander it is apparent that, at best, a seasonal wetland might develop in this meander through rainwater accumulation. However, such a wetland, if formed at all, would be extremely vulnerable to successional change to shrubland at which point it would lose any minor wetland function.

4.7.63 Meander 2 is proposed by CES (1997) to be filled and landscaped with riparian tree cover. Hence it would have no wetland function.

4.7.64 The third area was proposed as TOA during the construction process (see Figure 2.2 of CES (1997)) but was subsequently resumed. It was then proposed to plant trees and shrubs in this area, once no longer required during the construction process. Accordingly, this area would also have no wetland function after the completion of the Main Drainage Channel Scheme.

4.7.65 Whilst tree and shrub planting may be beneficial in some circumstances, the provision of such non-wetland habitat within Long Valley makes little contribution to the area's value as freshwater wetland, and specifically for wetland bird species of conservation importance. Therefore, it is proposed to restore these areas as wetland under the current (Lok Ma Chau Spur Line) project, thus providing a net gain of 1.4ha of wetland in these locations (see Table 4.39) which, together with the area in the railway reserve, will compensate in full for disturbance impact to more sensitive species. [The area within the railway reserve will have a reduced value for these species equivalent to 0.8ha of undisturbed wetland; the total reprovision for these species will, therefore, amount to 0.8ha (within the railway reserve, calculated as 50% of the value of undisturbed wetland for the area within the reserve but not directly under the viaduct) plus 1.4ha through meanders enhancement, comprising a total of 2.2ha - thus fully mitigating for the loss of 1.7ha detailed in Table 4.38. In view of the uncertainties involved in estimating disturbance impacts the additional area of 0.5ha over and above the direct and indirect loss of 1.7ha is considered to be a prudent allowance for the possibility that disturbance impacts are slightly greater than predicted].

4.7.66 These restored wetland areas will be provided with a permanent independent source of fresh water (either piped or by wells) thus ensuring that their required function as permanent freshwater wetlands is not compromised by other operations.

4.7.67 By appropriate wetland design and management, the identified impacts on disturbance sensitive species can be minimised. The management of the compensation areas under the viaduct, the meanders alongside the River Beas Main Drainage Channel, and this created wetland will be co-ordinated to provide habitat for all Species of Conservation Importance which are of ecological importance in Long Valley and are affected by the Main Drainage Channel and Spur Line projects. The project proponent for the Main Drainage Channels will hand over the meanders to be enhanced as soon as they are available, so that KCRC can implement the required mitigation as early as possible.

Design proposals for temporary marsh habitat to the west of the River Beas

4.7.68 One of the main objectives of the creation of temporary marsh habitat will be the maintenance of the viability of the Greater Painted-snipe population within the Long Valley area. Although the primary function will be to provide a refuge for this species, it is anticipated that other target species will also utilise the site, particularly species which select marsh, wet agriculture and abandoned agriculture habitats. These include gallinago snipe, amphibians, and dragonflies.

4.7.69 The habitat will consist of a shallow water marsh with dense aquatic vegetation, some areas of deeper, permanent open water and smaller shallow seasonal pools. The bottom profile will incorporate a ridge and furrow system, which will produce a greater diversity of water depths, and a scalloped vegetation zone (see Figure 4.11). The areas of deeper water will ensure that some water is retained even during exceptionally hot, dry periods, and will provide refugia for snakes and amphibians. The shallow seasonal pools provide particular good habitats for amphibians and some invertebrates, as these tend to have lower numbers of predators, such as fish and dragonfly larvae.

4.7.70 It is recommended that this temporary habitat be established (i.e. suitable for target species) prior to commencement of any works within the Long Valley marsh area. Consideration will need to be given to the wet season problems of heavy earth moving machinery in order to meet this deadline.

Habitat requirements of target species

4.7.71 The following table shows the habitat requirements of species which may be affected by the construction of the viaduct across Long Valley. These are primarily species which use marsh, wet agricultural and inactive agricultural habitats.

Table 4.40

Summary of habitat requirements for target Species of Conservation Importance

	Shallow water	Deep water	Emergent aquatic plants	Permanent open water
Japanese Yellow Bunting
Schrenck's Bittern	F		f
Red-billed Starling
Chinese Pond Heron	F	F	f	F
Great Egret		F
Little Egret	F	F
Black-winged Stilt
Northern Hobby
Greater Painted-snipe	F,r		F,r,b
Pheasant-tailed Jacana	F,r		f.r
Watercock	F,r		F,r
Common Snipe	F,r		F,r
Pintail Snipe	F,r		F,r
Swinhoe's Snipe	F,r		F,r
Bluethroat	F		F
Common Stonechat			F
Zitting Cisticola			F
Chestnut Bittern	F,r		F,r
Pallas's Grasshopper Warbler	F		F

F= foraging

R= Roosting

B= Breeding

4.7.72 Some of the species listed are not catered for within the temporary works area. These are species that although recorded infrequently in the marsh, wet agriculture and inactive agriculture, are not marsh specialists. Such species are not considered to require temporary mitigation, but are catered for in the permanent habitat mitigation.

Construction works

4.7.73 The temporary mitigation area is currently a commercial fishpond. To change it to a shallow water marsh will require the pond to be reprofiled to make it shallower and to provide a water outflow to keep water levels low. Reprofiling will also be required to incorporate the ridge and furrow bottom profile previously described for the marsh area around Lok Ma Chau Station. The ponds will be reinstated after construction of Spur Line is complete.

4.7.74 Given that the soils within the bottom of the fish pond will be severely compacted during the reprofiling works, the introduction of soils suitable to line the marsh will be required. Based on observations in Hong Kong, top soil from the Long Valley area will be suitable for translocation. These soils, which are usually used for agricultural purposes, appear able to support a marsh plant community within a relatively short period if they become abandoned and remain in a wet condition. If possible, such soils will be taken from areas to be affected by the works, or already abandoned areas within the Long Valley area. If a suitable abandoned area can be located, translocation of soil from such an area will help promote growth of marsh plant species through translocation of plants and seeds.

Hydrology

4.7.75 To ensure that this area does fulfil its refuge function, the water levels must be maintained throughout the dry season. To maximise the attractiveness of the marsh to target species, especially Greater Painted-snipe, the water level will need to be maintained at between 5 and 20 cm. Evaporation is expected to be high during the dry season, with October being particularly critical, as shown in the water budget (Table 4.41) and water levels will need to be monitored carefully at this time. Water levels will be highest during the summer months, when some flooding is likely. The drainage of the pond will need to allow for rapid disposal of flood water. There is a drainage channel close to the pond which can receive water from the pond in flood conditions, either by pumping or overflow.

Table 4.41

Water Budget for each Mitigation Area within Long Valley

	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec
Mean Rainfall mm/month	20.9	30.46	60.92	131.16	284.64	232.04	387.9	313.3	144.6	31.95	48.41	46.4
Mean Evapotranspiration mm/month	82.8	75.7	88.7	101.8	131.8	133.41	153.1	148.1	136	130.9	105.8	88
Marsh transpiration mm/month	115.92	105.98	124.18	142.52	184.52	186.76	214.34	207.34	190.4	183.26	148.12	123.2
Water balance mm/month	-95.02	-75.52	-63.26	-11.36	100.12	45.28	173.56	105.96	-45.8	-151.31	-99.71	-76.8
Water deficit mm/month	95.02	75.5	63.3	11.36	0	0	0	0	45.8	151	99.7	77
Water demand per ha l/day/ha	31641.66	25141.5	21078.9	3782.88	0	0	0	0	15251.4	50283	33200.1	25641
Water demand l/sec area (ha)
Meander 1 0.31	0.11	0.09	0.08	0.01	0	0	0	0	0.05	0.18	0.12	0.09
Meander 2A 0.71	0.26	0.21	0.17	0.03	0	0	0	0	0.13	0.41	0.27	0.21
Meander 3 0.13	0.05	0.04	0.03	0.01	0	0	0	0	0.02	0.08	0.05	0.04
Under Viaduct 2.41	0.88	0.70	0.59	0.11	0	0	0	0	0.43	1.40	0.93	0.72
Temporary Mitigation Area 1.76	0.64	0.51	0.43	0.08	0	0	0	0	0.31	1.02	0.68	0.52
Total water demand l/sec 5.32	1.95	1.55	1.30	0.23	0	0	0	0	0.94	3.10	2.04	1.58

Vegetation establishment

4.7.76 The plant species listed in Table 4.42 will be introduced into, or are expected to colonise the temporary mitigation area. Although all of the plant species listed are referred as “wetland species” because they can grow in water or soils and substrates that are periodically lacking in oxygen because of excessive water, the hydrological regime of the site plays an important role in determining the distribution and abundance of those species in the created wetland. Wetland plants are typically arranged in a series of concentric bands or zones determined by water depth. Plants that are able to tolerate similar depths of water grow in similar zone. This differing requirement in terms of water levels is allowed for in the ridge and furrow bottom profile of the marsh areas, which results in variable water depths, and will encourage a comparatively diverse plant community.

Table 4.42

Plant species to be used for marshland habitat creation at the

temporary mitigation area

Kam Tin	R.A.	C.A.
Alocasia macrorrhiza	Uc
Alternanthera sessilis	C	y
Ammania baccifera	C
Apluda mutica	C
Cardamine flexuosa	C
Cyperus malaccensis	C	y
Cyperus pilosa	Uc	y
Coix lachyme-jobi	R	y
Commelina nudiflora	Uc
Cuscuta chinensis	Uc
Echinochloa crus-galli	C
Eichhornia crassipes	Uc	y
Eleusine indica	Uc
Elipta prostrata	Uc
Floscopa scandens	Uc	y
Hedychium coronarium	Uc	y
Hygrophila salicifolia	C	y
Ipomoea reptens	C
Kyllinga monocephala	C
Leptochloa chinensis	C
Ludwigia ascandens	C	y
Ludwigia octovalis	C
Ludwigia perennis	Uc
Mikania micrantha	C
Oenanthe javanica	C
Panicum repens	Vc	y
Paspalum conjugatum	Vc
Paspalum distichum	Vc
Pennisetum alopecuriodes	R
Pennisetum purpureum	Uc
Philydrum lanuginosum	R	y
Polygonum hydropiper	C
Ranuncula soleratus	C
Rotala indica	C	y
Rumex maritimus	Vc
Sporobolus fertilis	Uc
Wedelia trilobata	Uc
Sesbania cochinchinensis	Uc

R.A. – Relative Abundance

C.A. – Commercially available

* vc - very common

c - common

uc - uncommon

r - rare

4.7.77 The proposed plants listed in Table 4.42 are based on the plant species composition of the preferred habitat for Greater Painted-snipe at Kam Tin. This area has a greater species richness than the Long Valley marsh which is attributed to the difference in management practice, as sources of disturbance in Kam Tin marsh (mainly human activities) have ceased. Vegetation has been allowed to colonize the area. Also, the marsh in Kam Tin is more heterogeneous in both water level and appearance than at Long Valley, and supports more species. As there will be control over the water levels and topography of the temporary mitigation area, it is anticipated that a higher plant species diversity can be achieved.

4.7.78 Of the species that are not commercially available, none are rare in Hong Kong, and are expected to colonise naturally. Those that are available will be introduced to provide plant cover as soon as possible. Translocation of top soils will be required to allow for soil compaction following reprofiling works. If possible, such soils should be taken from an area that already supports marsh plant species. Alternatively, pre-prepared areas of marsh could be translocated, in a manner similar to grass turfs. The advantages of the latter system is a greater level of success, and a speedier establishment of the vegetation.

Management

4.7.79 Wetlands are dynamic systems and therefore should ideally be restored and managed as such. However, in this situation, where the scale of mitigation is limited, this is not possible and management will be required to arrest natural ecological succession. Without such management, marshland habitats dominated by herbaceous communities will turn into mature non-wetland habitats dominated by shrubs and then trees.

4.7.80 Observations at Kam Tin and Long Valley have shown that complete clearance of the vegetation from selected areas results in rapid regrowth. It is considered that annual clearance of a selected section of the temporary mitigation area, on an annual basis is the best management option. No more than 25% of the total area should be cleared at one go, and such works should not be undertaken during the summer months. Ideally it should be undertaken in the late winter, prior to the start of the breeding season. In order to reduce disturbance to wildlife, vegetation will be cleared by hand. In view of the ability of Greater Painted-snipe to remain hidden, despite the close proximity of people, clearance by machine or spraying of herbicide are not consider appropriate measures. All invasive species such as reeds, or climbers will be cleared during vegetation control.

4.7.81 The temporary mitigation area should be maintained as a marsh habitat until the construction works in Long Valley marsh area are complete and the original marsh has been returned to its original functional state for less sensitive species

Design proposals for marsh habitats under the viaduct and meanders of the River Beas

4.7.82 The meanders will be designed to provide permanent marsh habitat primarily for Greater Painted-snipe, but also other marsh specialists, including gallinago snipe, amphibians and dragonflies. The two main sites are River Beas meanders number 1 and 2A, as identified in Figure 4.12. Together these will provide almost 1 ha of permanent marsh habitat. Meander 2A will be predominantly shallow water (5-20cm deep) with emergent aquatic vegetation, with two central areas of open water with a maximum water depth of 1m. The shallow areas will provide habitat for Greater Painted-snipe, and other target species, with the deeper water areas aimed at attracting longer legged bird species (Figures 4.15 and 4.16).

4.7.83 Meander No. 1 will be entirely shallow water with emergent aquatic vegetation. The smaller area of this site does not lend itself towards habitat provision for larger birds which are typically more readily disturbed (Figure 4.13).

4.7.84 An additional meander, No. 2 will be designed as temporary marsh habitat that is likely to dry out during the winter months (Figure 4.14).

4.7.85 The area below the viaducts will aim to provide habitat suitable for the smaller target bird species, which are less likely to be affected by the effects of disturbance from the adjacent railway and the physical viaduct structure. As such these areas will comprise a combination of reed beds and very dense emergent aquatic vegetation (Figure 4.17a b c).

Habitat requirements of target species

4.7.86 All target species are freshwater specialists, or wetland specialists. Some of these have more specific habitat requirements, and although the design caters for such species (e.g. Greater Painted-snipe) the range of micro-habitats, water levels, and vegetation types will provide foraging or roosting habitat for all target species (see Table 4.43). Few of the target species require deep water, the exceptions being Great and Little Egret.

Table 4.43

Summary of habitat requirements of target Species of Conservation Importance

	Shallow water	Deep water	Emergent aquatic plants	Permanent open water	Reed beds (under viaduct)	Dense aquatic vegetation (under viaduct
Japanese Yellow Bunting					F,r	F,r
Schrenck's Bittern	F		F
Red-billed Starling						F
Chinese Pond Heron	F	F	F	F
Great Egret		F
Little Egret	F	F
Black-winged Stilt
Northern Hobby
Greater Painted-snipe	F,r		F,r,b
Pheasant-tailed Jacana	F,r		f.r
Watercock	F,r		F,r
Common Snipe	F,r		F,r
Pintail Snipe	F,r		F,r
Swinhoe's Snipe	F,r		F,r
Bluethroat	F		F		F,r	F,r
Common Stonechat			F		F,r	F,r
Zitting Cisticola			F		F,r	F,r
Chestnut Bittern	F,r		F,r
Pallas's Grasshopper Warbler	F		F		F,r	F,r

F= foraging

R= Roosting

B= Breeding

Construction works

4.7.87 The meander areas will need to be reprofiled to create shallow ponds that slope gently away from the water source (Figure 4.16). Incorporated into the bottom profile will be a ridge and furrow system, that will produce greater diversity of water depths and a scalloped vegetation zone. Creation of bunds and ditches will be required as these will be needed to delineate the compensation area and for the distribution of water. However, edge bunds should be low, to allow water from the surroundings to flow into the wetland as this contains the flap valve through which water drains into the channel in flood conditions.

4.7.88 Water levels will be controlled by small weirs in the viaduct area, to direct flow in the required direction and maintain areas of deep and shallow water (Figure 4.17a b c). Translocation of soils will be required due to the anticipated compaction of the base of the meander during reprofiling. Compaction of soils beneath the viaduct may also occur during construction, and regrading of the area is likely to be required to form the wetland area.

Hydrology

4.7.89 Due to complications of water supply (including potential conflicts with farmers) and the need to keep the main mitigation areas permanently wet, the mitigation areas will require a permanent, independent water source. It is considered that the best way to achieve this is pump from the ground water into the mitigation areas. This will also ensure that water quality is reasonably good. Two such pumps will be required, one at one end of the area below the viaduct, and another to service meanders Nos. 1 and 2.

4.7.90 The viaduct pumping location will be at the east end of the wetland, and water will flow west through a series of ponds as shown in Figure 4.17a b c. Weirs will be constructed to maintain water levels and ensure a permanent water presence even in the dry season. At the western end of the viaduct wetland, overflow will discharge into the closest meander, in this case, Meander No. 4 in the River Beas Drainage Channel Report. This meander already has a water feed and a flap valve outlet to the channel, which allows discharge in summer flood events. This additional water flow will enhance the meander to retain a wetland status even during the dry season. The ability to control water levels in such a way is required to prevent the sites becoming dry during the winter months.

4.7.91 Meanders 1 and 2A will receive water from the ground through a pump, as described above, and will discharge via flap valves to the channel. A channel from meander 2A will feed water to meander 1 to retain its wetland status.

4.7.92 Meander No. 2 will rely on rainwater and is expected to act as a seasonal marsh. This meander is only 0.13 ha, and as such is too small allow for the rather fine control of water levels that are required for the creation of permanently wet marsh. The meander also has a flap valve for drainage.

4.7.93 The proposed use of pumped fresh water will have an insignificant effect on existing drainage conditions as pumping will only be undertaken in dry conditions in order to protect the integrity of wetland habitat. Such pumping will, therefore, have no effect on flood risk as it will only take place during the dry winter season when the flood risk is negligible.

4.7.94 Created wetland habitats may, in fact, have a slight beneficial effect in reducing flood risk at the beginning of the wet season as these areas will not generally be operated at full water holding capacity during the dry winter months. There will, therefore, be a slight buffering effect as the created wetland habitats absorb some of the surplus run-off from heavy rainfall at the beginning of the wet season before they reach their wet season water levels. The total amount of water involved will, of course, be unchanged as the volume of rain and the run-off coefficient within the system will be unchanged whether or not the wetland habitats are created. Details of discharges to watercourses should be proposed to DSD at detailed design stage for the various project components.

Vegetation establishment

4.7.95 The vegetation establishment regime shall be the same as for the temporary mitigation areas, with the exception of the creation of reedbeds (phragmites spp,) in the area below the viaduct.

4.7.96 In some areas of the meanders, deeper water will allow the growth of taller reeds and emergent vegetation. The types of vegetation to be used in each meander are shown on Figures 4.13 4.14 4.15 4.16.

Management

4.7.97 Management of the vegetation of meanders 1, 2 and 2A will be as for the temporary mitigation areas. Occasional thinning of vegetation will be required to maintain a clear water flow. The area below the viaducts will require less intensive management, as denser cover may be allowed to become established in this area. The removal of larger, woody plants and some control of the reeds may be all that is required.

Overall residual impacts from habitat loss and disturbance in Long Valley

The overall levels of compensation predicted from the above proposed compensation measures for Species of Conservation Importance that regularly occur or have the potential to regularly occur in Long Valley is summarised in Table 4.44.

Table 4.44

The overall levels of compensation predicted from compensation measures for Species of Conservation Importance that regularly occur or have the potential to regularly occur in the Long Valley area.

Species	Sensitivity to disturbance	Compensation achieved
Amphibians	Nil	Fully compensated
Reptiles	Low	Fully compensated
Mammals	Moderate	Scope for additional habitat provision
Japanese Yellow Bunting	Small	Fully compensated
Schrenck's Bittern	Small	Fully compensated
Red-billed Starling	Small-medium	Scope for additional habitat provision
Chinese Pond Heron	Medium	Scope for additional habitat provision
Great Egret	Large	Scope for additional habitat provision
Little Egret	Medium-large	Scope for additional habitat provision
Black-winged Stilt	Medium-large	Scope for additional habitat provision
Greater Painted-snipe	Moderate	Fully compensated (Baseline species for compensation proposals)
Northern Hobby	Large	Scope for additional habitat provision
Pheasant-tailed Jacana	Medium-large	Fully compensated
Watercock	Medium	Fully compensated
Japanese Quail	Medium-low	Scope for additional habitat provision
Common Snipe	Medium	Fully compensated
Pintail Snipe	Medium	Fully compensated
Swinhoe's Snipe	Medium	Fully compensated
Richard's Pipit	Small	Scope for additional habitat provision
Bluethroat	Small	Fully compensated
Common Stonechat	Small	Fully compensated
Zitting Cisticola	Small	Fully compensated
Chestnut Bittern	Small	Scope for additional habitat provision
Pallas's Grasshopper Warbler	Very small	Fully compensated

Habitat fragmentation

4.7.98 Although fragmentation effects will be relatively low due to the elevation of the tracks through the most ecologically valuable habitats, residual impacts may occur indirectly through disturbance effects and avoidance of the railway structure. The following mitigation measures should be therefore taken to reduce the impacts of habitat fragmentation:

• Additional areas of marshland habitat should be created to compensate for the adverse effects of the high level of fragmentation of marshland and agricultural land which will occur as a result of the works.

This is considered essential due to the relative importance of this type of habitat and its scarcity not just within the Study Area, but in the northwest New Territories as a whole. This requirement is addressed in the compensation areas in Long Valley described in the previous section. Marshland, rather than wet agricultural land, creation is appropriate in order to permit habitat management measures that will enhance the carrying capacity for key species.

• Continuity of wetland habitat should be maintained through the creation of wetland corridors.

The area under the viaduct will form a wetland corridor and where possible, will be linked with the meanders and wetland created alongside the Main Drainage Channels passing beneath the viaduct. See above account of habitat compensation at Long Valley.

• Landscape treatment should maximise views and openness of the area as well as habitat links both along and across the railway.

The use of landscape measures to mitigate for adverse visual impacts should not result in the creation of visual barriers that will inhibit use of the area by key species.

Dust deposition

4.7.99 Dust creation should be minimised using standard procedures, including the damping down through water spraying during periods of dry weather. Details of mitigation measures to be used for minimising dust deposition are included in the Air Quality Impact Assessment.

Increased sediment load

4.7.100The potentially increased sediment load during construction will be controlled and minimised through implementation of good site management practices such as means for the sediment to settle before discharge of the clear supernatant. During operation, sediment load is likely to be minimal and any solids will settle in the oil interceptors and sediment traps that are incorporated in the drainage system. With good management and maintenance programmes, these measures should minimise potential impacts from this source.

Pollution from the railway operations (e.g. oil leaks) or accidents involving toxic chemicals

4.7.101Good storage practices and handling of the chemicals used during the construction period will minimise the opportunity for impact on the terrestrial and aquatic environments. Spillage impacts can be minimised by storing chemicals in bunkers that contain the volume of the containers stored. During operation, there is a potential for oil and grease to be washed off parts of the train during storm events. This will be minimised through train design reducing exposed parts to wet weather, and provision of an oil interceptor within the drainage system. Oil and grease from the track maintenance operations is expected to be minimal because low levels of grease are used on the floating slab trackform. Regular maintenance of the interceptor will maximise its efficiency in trapping oil before release into the environment.

Soil compaction

4.7.102The extent of the area to be compacted for works during construction should be minimised to reduce damage to habitats, particularly in Long Valley and Lok Ma Chau areas. Each area should be re-instated when the works on that area is complete, to re-establish an environment as similar as possible to the original habitat. This may require excavation, removal of compacted (and polluted) material, and replacement with suitable material for re-creation, particularly in the areas of wetland creation and directly alongside the track where wetland is impacted.

Hydrological disruption

4.7.103During construction, a temporary drainage system will incorporate necessary sedimentation areas to minimise sediment lost in surface run-off. In the operational period, the drainage system will be designed to direct rainwater from the box girder section of the rail track, where flows will pass through an oil interceptor and onto a gravel area at ground level. The gravel serves to spread the flows and filter out excess solids, especially in storm flows when most of the flow will bypass the oil interceptor to prevent washout. Encouraging infiltration into the ground will minimise impacts on the hydrology of the area.

4.8 Implementation of Mitigation Measures

4.8.0 The ecological mitigation areas proposed for the whole of Sheung Shui to Lok Ma Chau Spur Line totals approximately 28.5 ha of fishpond habitat, 3.5 ha of reedbed around the station, and 3.0 ha of bamboo planting as landscaping, and reedbed for effluent polishing, within the Lok Ma Chau area. In Long Valley, ecological mitigation areas total 5.6 ha, comprising marsh for temporary construction impacts, 2.4 ha reedbed/marsh beneath the viaduct, and 1.4 ha marsh in the River Beas meanders to be enhanced under this project . The construction of the ecological mitigation areas to compensate for impacts to habitats due to the Spur Line will be undertaken by KCRC.

4.4.1 Management of the ecological mitigation areas will be carried out through the establishment of a Wetland Trust. Short-term and long term arrangements for the wetland management are described below.

Short-term Arrangements of the Wetland Trust

4.4.1 Before operation of the railway, KCRC will be responsible for the design, construction and management of the wetland compensation areas. In the event that the HKSAR Wetland Trust for multi-projects could not be established on time (i.e. approximately May 2002), a contractual agreement between KCRC and a specialist contractor will be entered into where the contractor will take up the management responsibility for the maintenance of the wetland areas. The contract would commence in approximately May 2002 and operate until such time as the HKSAR Wetland Trust is set-up, the target date of which is expected to be early 2004.

4.4.1 As the management agent, the contractor will take over the Environmental Permit (EP) obligations with regard to the Habitat Creation and Management Plan (HCMP) management and maintenance requirements. Once the HKSAR Wetland Trust is in place, the KCRC would seek to transfer the responsibilities for long-term management to the HKSAR Wetland Trust.

4.4.1 Responsibility for the long-term management of the habitat compensation areas will be handed over to the HKSAR Wetland Trust, which is in the process of being established. The appointed trustees of this Trust will oversee the disbursement of funds for the management of wetland compensation areas associated with major development projects, such as those proposed for the Spur Line within this EIA. This will provide a mechanism to ensure that the long term management and maintenance functions associated with the wetland areas are fulfilled. Implementation of the defined responsibilities will be through a management agent appointed by the Trust with experience in wetland management.

4.4.1 The long-term financial support for the management of the ecological mitigation areas will be provided through a donation by KCRC of a lump sum to the Wetland Trust. This lump sum will be invested by the trustees in order to provide income to cover recurrent management and maintenance costs.

4.4.1 The trustees will take over the Environmental Permit (EP) obligations with regard to the Habitat Creation and Management Plans (HCMP) long-term management requirements and will appoint appropriately qualified management agents to manage and maintain the wetlands according to the requirements of the EP and the HCMP.

Programme for Setting up the Wetland Trust

4.4.1 KCRC envisage that the programme for the construction of the Spur Line wetland areas envisages a start on the enhancement of the off-site fishpond area at Lok Ma Chau in November 2000. The initial enhancement work (outside the 500m disturbance zone)is programmed to take a total of 6 months and to be complete in May 2001. Enhancement of the Long Valley ecological mitigation areas will commence 1st quarter 2001, and will be completed before construction of the viaduct directly impacts the ecologically sensitive areas.

4.4.1 KCRC envisage that the Wetland Trust would be set up in early 2004 based on the assumption that it would take about 30 months to set up the Trust after the Wetland Compensation Study report is issued.

4.4.1 Land resumed for the construction of the viaduct sections of the railway in Long Valley will be used for the construction of wetland mitigation areas (as described above) and then passed over to the Wetland Trust together with the old River Beas meanders.

4.4.1 The land around the future station in Lok Ma Chau, which is proposed for mitigation, is all owned by Government. Implementation of the proposed marshland / reedbed mitigation measures adjacent to the station will require formal clearing of the land before mitigation measures can be put in place. The most efficient and effective way of implementing the enhancement measures for fishpond habitats, while minimizing disruption to the existing environment, would be through continued commercial fishpond operations.

4.4.1 Operation of fishponds would be carried out through contracts between the Wetland Trust and a management agent experienced in wetland management. Contracts would stipulate that operation of the ponds must be according to specified environmental protocols documented as Standard Operating Procedures.

4.8.1 The ecological mitigation areas proposed for the whole of Sheung Shui to Lok Ma Chau Spur Line totals approximately 28.5 ha of fishpond habitat, 3.5 ha of reedbed around the station, and 3.0 ha of bamboo planting as landscaping, and reedbed for effluent polishing, within the Lok Ma Chau area. In Long Valley, ecological mitigation areas total 5.6 ha, comprising marsh for temporary construction impacts, 2.4 ha reedbed/marsh beneath the viaduct, and 1.4 ha marsh in the River Beas meanders to be enhanced under this project. The construction of the ecological mitigation areas to compensate for impacts to habitats due to the Spur Line will be undertaken by KCRC.

4.8.2 Management of the ecological mitigation areas will bemay later be carried out through the establishment of a HKSAR wetland management organization, which is envisaged to be a non-profit making, territory wide wetland management body for overseeing HKSAR wetland in the future. The concept of the HKSAR wetland management organsation is currently in the process of being establishedunder discussion. Short-term and long- term arrangements for the wetland management are described below.

Short-term and Long-term Management Arrangements of Wetland

4.8.3 Before operation of the railway, KCRC will be responsible for the design, construction and management of the wetland compensation areas. In the event that the HKSAR wetland management organisation for multi-projects could not be established on timeon completion of the Spur Line Project (i.e. approximately late 2004), a contractual agreement between KCRC and a specialist contractor will be entered into where the contractor will take up the management responsibility for the maintenance of the wetland areas. The contract would commence in approximately late 2004 and and operate until such time as the HKSAR wetland management organisation is set-up and takes over the management and maintenance of the wetland compensation areas.

4.8.4 As the management agent, the contractor will take over the Environmental Permit (EP) obligations with regard to the Habitat Creation and Management Plan (HCMP) management and maintenance requirements. Once the HKSAR wetland management organisation is in place, the KCRC would seek to transfer the responsibilities for long-term management to the HKSAR wetland management organisation.

4.8.5 KCRC envisage that the programme for the construction of the Spur Line wetland areas envisages a start on the enhancement of the off-site fishpond area at Lok Ma Chau in November 2000. The initial enhancement work is programmed to take a total of 6 months and to be complete in May 2001. The complete formation of the enhanced ponds will take until 2004. Enhancement of the Long Valley ecological mitigation areas will commence 1st quarter 2001, and will be completed before construction of the viaduct directly impacts the ecologically sensitive areas.

4.8.6 The land around the future station in Lok Ma Chau, which is proposed for mitigation, is all owned by Government. Implementation of the proposed marshland / reedbed mitigation measures adjacent to the station will require formal clearing of the land before mitigation measures can be put in place. The most efficient and effective way of implementing the enhancement ensures for fishpond habitats, while minimizing disruption to the existing environment, would be through continued commercial fishpond operations.

4.8.7 Operation of fishponds would be carried out through contracts between the wetland management organisation and a management agent experienced in wetland management. Contracts would stipulate that operation of the ponds must be according to specified environmental protocols documented as Standard Operating Procedures.

4.8.8 Standard Operating procedures for fish pond operators would be produced as part of the detailed design process and will cover key factors affecting the ecological value of the habitats, including:

· the physical structure of the fish ponds (including bund and bottom profiles);

· timing of draw-down periods for harvest;

· frequency of complete drainage and re-working of ponds;

· frequency of set-aside from operations (i.e. no stocking) for selected ponds;

· fish stock composition and densities;

· the use of fertilisers and food;

· location of buildings;

· use of pesticides and aquaculture chemicals, and

· management of bund vegetation.

4.8.9 The Wetland Trust wetland management organisation will be responsible for monitoring and reviewing the ecological effectiveness of the compensation measures.

4.8.10 Detailed requirements for the long-term management of the mitigation areas will be identified as part of the detailed design phase and documented through Habitat Creation and Management Plans produced for the fishpond and reedbed / marsh mitigation areas at the Lok Ma Chau / San Tin station and for the marsh mitigation in Long Valley. The aim of these Management Plans should be to:

· Describe the sites by collating key physical and biological information on them.

· Clearly identify the management objectives and measurable targets for the site.

· To anticipate any conflicts between, and problems achieving, the objectives for the site and identify the best methods for resolving them.

· To identify and describe the practical management measures necessary to achieve the objectives for the site.

· To identify detailed monitoring requirements and methods for establishing if objectives are being meet and for measuring the effectiveness of management actions.

· To organise manpower and resources.

· To act as a guide to parties responsible for implementation.

· To facilitate communications with other interested parties.

4.8.11 The required contents and structure of the Management Plans will be identified as part of the detailed design process. It is important that the plans should provide detailed works programmes with clear indications of the timing and responsibilities for management actions.

4.8.12 A summary of the timing of the initial stages of the implementation of habitat compensation measures and responsibilities for these is provided in the implementation Schedule in Chapter 12 of this report.

4.9 Estimate of Residual Impacts

4.9.1 An estimate of residual impacts taking into account the feasible and proposed avoidance, minimisation, and compensation mitigation measures described above is presented in Table 4.45.

Table 4.45

Proposed mitigation measures and predicated residual impacts

Potential impact	Potential mitigation measures				Predicted residual impact
	Avoidance	Minimizing	Compensation
			On-site	Off-site
Habitat loss	Slight realignment of the track to minimise habitat loss and for compatible engineering reasons has already been carried out and taken into account. Further re-routing is not feasible	All habitats of conservation importance that are occupied or ecologically altered should be restored	Temporary lost fishponds, agricultural and marshland habitats to be reinstated. An compensation area will be provided around the Lok Ma Chau Station. On-site habitat compensation feasible within Long Valley for direct habitat loss. Recommendation for early commencement of Works on mitigation areas in Lok Ma Chau and Long Valley to minimize construction impacts.	Permanently lost and disturbed fishponds, agricultural and marshland to be compensated on- or off-site to avoid serious impacts to Species of Conservation Importance .It is recommended that compensation of defined areas of the mitigation areas begins before construction of Spur Line to avoid serious impacts. Fishpond compensation may involve creation of shallows for enhanced feeding.	Provision of compensation areas will result in minimal impacts to Important Habitats (fish ponds and seasonally inundated agricultural habitats) and associated wetland Species of Conservation Importance during construction. Advance construction is highly recommended to minimise impacts. No residual impacts on most species if proposed fishpond compensation is implemented. No significant residual impacts from direct habitat loss on other habitats
Direct mortality	Avoid use of transparent and reflective materials on station and other structures	Mark aerial cables where feasible. Use tree plantings to dissuade large raptors from foraging over railway lines	Not applicable	Not applicable	Very low residual impacts are unavoidable
Habitat fragmentation	As above	Where feasible, tracks will be elevated and suitable habitats restored or maintained underneath to allow passage of Species of Conservation Importance. Where required tunnels should be placed through embanked sections or under cuttings to allow passage of amphibians, reptiles and mammals	Where feasible, new ecological corridors could be constructed to link fragmented habitats. The use of viaducts for most of the alignment will provide corridors, especially in Long Valley.	Not applicable	Compensation of habitat as described will minimise impacts to Important Habitats (fish ponds agricultural habitats and marshland) and associated wetland Species of Conservation Importance during construction. Early construction of compensation areas is recommended. Low impacts are likely during the operation phase.
Disturbance of wildlife by noise and visual movement	Not feasible	Artificial screening and natural buffers (e.g. trees) should be used to reduce noise and the visibility of people in sensitive areas (e.g. close to wetlands)	A compensation area will be provided around the Lok Ma Chau Station to mitigate disturbance impacts.	Necessary to avoid serious impacts to disturbance sensitive Species of Conservation Importance. Compensation area will be provided in Long Valley, management of which will be co-ordinated to mitigate impacts on disturbance sensitive species. Early construction of Long Valley construction mitigation and construction and/or management of selected areas of Lok Ma Chau fishponds is recommended to avoid serious impacts.	Provision of compensation areas will minimise impacts to wetland Species of Conservation Importance during construction. Early construction of the compensation areas is recommended. No residual impacts on most species if proposed compensation is implemented. Minimal impacts in Long Valley.
Dust deposition on surrounding habitats	Not feasible	Standard construction mitigation measures should be used (e.g. watering of dry surfaces to reduce dust)	Not applicable	Not applicable	Overall ecological impacts are likely to be minimal.
Increased sediment load and pollution of watercourses	In the operation of the railway, particularly in relation to the use of chemicals and oil and grease application to tracks, all feasible steps should be taken to avoid accidents.	A suitable drainage system should be established, including sedimentation and infiltration areas in necessary positions to intercept construction site runoff. Contingency measures should be planned for rail operations, in the anticipation of accidents which may lead to potential water pollution impacts	Not applicable	Not applicable	Minimal residual impacts.
Soil compaction	Work in areas which are not sensitive to soil compaction.	Minimize areas of soil compaction.	Habitat creation will include measures to minimize adverse impacts.	Not applicable	Minimal residual impacts.
Hydrological disruption	Direct flow away from sensitive areas	Suitable drainage design to spread run-off as much as possible.	Water management regime will be incorporated into the habitat management plan for compensation areas.	Water management regime will be incorporated into the habitat management plan for compensation areas.	No residual impacts.

4.10 Ecological monitoring and audit requirements

4.10.1 The aim of monitoring of any implemented wetland mitigation scheme should be to establish that clearly defined end objectives are meet according to measurable performance criteria. Such objectives and performance criteria should be based on the broad objectives and targets as outlined here, but should be identified in detail as part of the production of a Management Plan (see above).

4.10.2 However, as overall objectives may take many years to be met, additional operational objectives that indicate a direction of desirable change should also be monitored (e.g. hydrological regimes and plant establishment required to produce certain habitats and wetland functions). Such information may also help to guide management activities as they proceed. In practice, the separation of 'means' and 'ends' objectives can be difficult as there can be overlap in some cases. For example, establishment of a particular vegetation community may be an 'end' objective in itself, but this may also be a 'means' objective for providing suitable habitat for the recolonisation of a target species, another 'end' objective.

4.10.3 A list of wetland attributes, some of which may be form an appropriate basis for measurable performance criteria is provided below. These should be measured according to clearly specified Standard Operating Procedures (including specifications on the timing and frequency of monitoring) developed as part of the Management Planning process.

Pedology

· Change in depth (i.e. topsoil/subsoil regime)

· Change in organic versus mineral content

· Change in nutrient composition

· Change in anaerobic/aerobic condition

· pH

Hydrology

· Water levels and storage capacity (seasonal fluctuations and flood attenuation capacity)

· Inflows (surface, ground, direct)

· Losses

· Surface water flow rates

· Hydraulic gradients

Water chemistry

· pH

· BOD

· dissolved oxygen

· ammonia

· total oxidised nitrogen

· orthophosphate

Ecology

· General habitat extent (e.g. habitat mapping and fixed point or aerial photography)

· Survival, growth and reproduction of key species (especially sown, planted or translocated species)

· Health of planted species e.g. pest or fungal attack.

· Plant species richness and diversity

· Plant community composition (individual species frequency and cover, and NVC community type) formation and zonation

· Vegetation structure

· Natural colonisation and succession

· Maintenance of natural ecological processes

· Presence and abundance of Species of Conservation Importance in relation to baseline levels and target population levels for specific habitat compensation mitigation measures (i.e. within the enhanced fish pond area and reedbed/marsh areas around the San Tin Station, the temporary construction mitigation area, the River Beas meander compensation areas and the Long Valley marshland areas to be created under the viaduct).

· Presence or abundance of selected ecological indicator species

4.10.4 In addition to these general monitoring requirements additional monitoring and research may be required to refine specific management actions required to meet particular objectives. In particular, the ecology and habitat requirements of Greater Painted-snipe are still not fully understood. One study at Kam Tin showed that there were significant differences in the diurnal and nocturnal distribution of this species. Given that this is a target Species of Conservation Importance for compensation at Long Valley it is recommended that baseline data be collected on diurnal and nocturnal habitat requirements prior to the commencement to the works through the use of radio telemetry. This will enable the monitoring of colonisation of the temporary mitigation area without causing excessive disturbance. This radio telemetry work should continue throughout the construction period as both the existing marsh at Long Valley and the temporary mitigation area will be directly affected by the works. The maintenance of a viable population of Greater Painted-snipe throughout the construction period is a primary aim of the mitigation design, and as such should be thoroughly monitored.

4.10.5 Another key area of monitoring and research should be aimed at establishing the impacts of the railway and station construction and operation on the two globally threatened species of eagle: Imperial Eagle and Greater Spotted Eagle. As described above, it is not expected that residual impacts on these species are likely to be significant. However, given there globally threatened status it is proposed that these species should be included as part of the monitoring and research should be carried out to enable enhancement of habitats for these species. Although primarily wetland species within Hong Kong, their specific habitat and particularly food requirements are not well known. Further information on these requirements would enable some refinements of the proposed habitat compensation areas to the benefit of these species. Such findings could also be applied to the management of other areas that are frequented by these wide-ranging species, such as the Mai Po Nature Reserve.

4.10.6 Detailed proposals for research into the ecology and habitat requirements of wintering Greater Spotted Eagle and Imperial Eagle should be developed during the detailed design process and included as part of the Environmental Support Services Contract.

Conclusion

4.10.7 After balancing the engineering, operation, safety, environmental and other considerations, the alignment of proposed Sheung Shui to Lok Ma Chau Spur Line has been determined along its present route. The railway will cross several ecologically valuable habitats. Of these, the low-lying wet agricultural habitats of Long Valley and the fishponds around San Tin / Lok Ma Chau are of most ecological interest because they support a diverse range of flora and fauna, including a number of globally, regionally and locally threatened bird Species of Conservation Importance as listed in Table 4.2. Such wetland habitats have declined rapidly in Hong Kong in recent years and are under continuing development pressure.

4.10.8 Other remnant habitats of ecological value include the small patches of wet agricultural and inactive fishpond, and the fung-shui forest in Chau Tau; as well as some fish ponds in Long Valley, but these habitats are small and fragmented and consequently now of lesser importance.

4.10.9 A review of the effects of the construction and operation of the Spur Line on these valuable ecological resources indicates that:

• Potentially high impacts may occur as a result of disturbance and the loss of fishponds (mainly resulting from the station complex) within the Wetland Conservation Area near Lok Ma Chau.

• Moderate impacts to habitats and high impacts to certain freshwater wetland dependant bird species (in particular, Greater painted-snipe) may occur from the disturbance and some loss of agricultural habitats and marshlands in Long Valley

• Other potential sources of impacts are direct mortality, fragmentation, dust deposition on vegetation, increased sediment load, toxic pollution from accidental spills, soil compaction and hydrological disruption, but predicted impacts from these are considered to be mostly low in habitats of ecological importance.

4.10.10It is predicted that approximately 28.5 ha of fishponds and 6.5 ha of reedbed and marsh (comprising 3.5 ha marsh, 2.0 ha reedbed for effluent polishing and 1.0 ha bamboo/reedbed for screening), provided as habitat compensation measures (through the enhancement of fishponds and creation of reedbeds and marshland) would eliminate residual impacts from the Spur Line operation in the Lok Ma Chau station area for most Species of Conservation Importance. It is recommended that construction impacts in Lok Ma Chau area will be mitigated through early establishment of selected areas of the fishpond compensation area outside the 500 m disturbance zone. Mitigation will be through creation of shallow feeding habitats.

4.10.11It is predicted that 2.4 ha on-site (beneath the viaduct) and co-ordination of ecological compensation with the Main Drainage Channels mitigation areas off-site would provide compensation to eliminate all residual impacts resulting from habitat loss and disturbance for all Species of Conservation Importance in the Long Valley area. The mechanism of co-ordinating the different ecological compensation areas from the Main Drainage Channels and Spur Line to enable impacts to all species to be adequately mitigated, is in the process of being formulated. The project proponents for the Main Drainage Channels will hand over the meanders to be enhanced 1.4 ha as soon as they are available. Wetland establishment and management details have been provided in this report. During the construction period an area of land (approximately 2.0 ha) has been identified for compensation, preferably in advance of construction, to allow creation of a marshland refuge area which will be utilised by wetland dependent species, especially the Greater Painted Snipe, whilst the construction process in the Long Valley marsh area is underway. Water management regimes will be identified for both on-site and off-site compensation areas. In addition, ecological impacts from construction will be minimized through restriction of access in ecological important areas and noise and air impact reduction measures as described in this report.

4.10.12Short-term management of the ecological compensation areas will be arranged, through a contractual agreement between KCRC and a specialist contractor. Responsibility for the long-term management of the ecological habitat compensation areas will be handed over to the HKSAR Wetland Trust wetland management organisation, which is in the process of being established. The trustees wetland management organisation will assign a management agent with experience in wetland management, to fulfil the management and maintenance requirements defined in the Habitat Creation and Management Plan and the Environmental Permit. Financial support of the wetland compensation area will be through a lump sum donation by KCRC, which the trustees will invest to cover the recurrent costs of the wetland management.

4.10.13This approach of compensating both habitat loss and disturbance impacts in the development of mitigation measures, and restriction of contractor access within ecological sensitive areas, is unique in providing protection for Hong Kong’s ecological resources. Such an approach allows long-term management of Hong Kong’s environment.

4.10.14Standard avoidance and minimisation mitigation measures are considered to be adequate to eliminate all other residual impacts from other potential impacts sources.

REFERENCES

Ades G & Reels, G. (1998). Mammals. Porcupine! 18:19-26.

Anon. (1995). Focus on fish ponds: VI. Porcupine! 13:21.

Anon. (1998). Special Features: Focus on Farmland. Porcupine! 18:19-26.

Aspinwall & Company Hong Kong Limited (1997). Study on the Ecological Value of fish Ponds in Deep Bay Area: Executive Summary.

Aspinwall Clouston & Wetlands International - Asia Pacific (1997). Development of a comprehensive conservation strategy and a management plan in relation to the listing of Mai Po and Inner Deep Bay as a Wetland of International Importance under the Ramsar Convention. Agreement No. CE 47/95.

Bergers, P.J.M. (1997). Versnippering door railinfrastructuur. Een verkennende studie. IBN-rapport 262. Instituut voor Bos-en Natuuronderzoek, Wageningen.

Binnie (1999). Lok Ma Chau Boundary Crossing Expansion EIA. For Architectural Services Department.

Binnie Black & Veatch (1998). Sheung Shui to Lok Ma Chau Spur Line PPFS. KCRC.

Binnie Black & Veatch (1999). Expansion of Kiosks and Other Facilities at Lok Ma Chau Boundary Crossing: Final Environmental Study Report.

Binnie-Meinhardt JV (1999). KCRC Sheung Shui to Lok Ma Chau Preliminary Project Feasibility Study - Final Report (Appendix J)

Boring, A.M. (1934). The Amphibia of Hong Kong (1). Hong Kong Naturalist 5: 8-22.

CES (1997). Main Drainage Channels for Fanling, Sheung Shui and Hinterland - Final EIA Report. TDD, Hong Kong Govt.

Chalmers, M.L. (1986). Annotated checklist of the birds of Hong Kong. Hong Kong Bird Watching Society, Hong Kong.

Chong D. W. & Dudgeon, D. Hong Kong Stream Fishes: An Annotated Checklist with Remarks on Conservation Status. Memoirs of the Hong Kong Natural History Society, 19: 77-99.

Chu WH, unpublished. Conservation of Terrestrial Biodiversity in Hong Kong. M.Phil Thesis. The University of Hong Kong.

Chu, W.H. and Xing, F.W. (1997). A checklist of vascular plants found in Fung Shui woods in Hong Kong. Memoirs of the Hong Kong Natural History Society 21:151-158.

Collar, N.J., Crosby, M.J. and Stattersfield, A.J. (1994). Birds to Watch 2. The World List of Threatened Birds. BirdLife International, Cambridge.

Dahmer, T. and Felley, M. (1998). Black-faced Spoonbill (Platalea minor) census, January 1998. Unpub. MS.

del Hoyo, J., Elliott, A. and Sargatal, J. (eds). (1992). Handbook of the birds of the world. Lynx Edicions, Barcelona.

Diskin, D. and J. Pearse (1996). Birding Hong Kong. D. Diskin, PO Box 952 Shatin Central P.O., NT, Hong Kong. 160pp.

Dudgeon D. & Corlett R. (1994). Hills and Steams - An Introduction to Hong Kong. Hong Kong University Press.

Dudgeon D., E. Chan & Reels, G. (1996) - unpublished. Ecological Study of Freshwater Wetland Habitats in Hong Kong.

EPD (1996). River Water Quality in Hong Kong. Monitoring Section, Water Policy & Planning Group, Environmental Protection Department, Hong Kong Government. 167pp.

Erhardt, A. (1985). - Diurnal Lepidoptera: Sensitive indicators of cultivated and abandoned grassland. Journal of Applied Ecology 22:849-861

ERM (1999b). KCRC East Rail Extensions – Sheung Shui to Lok Ma Chau Spur Line Project : Ecological Study.

ERM (1998a) - unpublished. KCRC West Rail (Northern Section) - Initial Environmental Assessment Report.

ERM (1998b). Main Drainage Channels for San Tin - EIA. TDD 1998.

ERM (1998c). West Rail TS900 EIA Study (West Kowloon to Tuen Mun Centre and Northern Section & West Rail Depot): Final Assessment Report. KCRC.

ERM (1999a). Main Drainage Channels and Poldered Village Protection Scheme for San Tin, NWNT: Environmental Impact Assessment Study - Final Assessment Report.

Fasola, M. and Barbieri, F. (1978). Factors affecting the distribution of heronries in northern Italy. Ibis 120:537-540.

Hafner, H., Dugan, P., and Boy, V. (1987). Herons and Wetlands in the Mediterranean: Development of indices for quality assessment and management of Mediterranean Wetland Ecosystems. European Economic Community. Tour du Valet.

Hill, D., Hockin, D, Price, D., Tucker, G., Morris, R., and Treweek, J. (1997). Bird disturbance: improving the quality and utility of disturbance research. Journal of Applied Ecology 34: 275-288.

Hill, D.S. & Phillipps, K. (1981). A Colour Guide to Hong Kong Animals. Government Printer, Hong Kong.

Hockin, D., Ounsted, M., Gorman, M. Hill, D., Keller, V. and Barker, M. (1992). Examination of the effects of disturbance on birds with reference to the role of environmental impact assessments. Journal of Environmental Management 36: 253-286.

Ingelmann, E.(1994). Zum Einflußvon Bahndämmen auf das Wanderverhalten von Erdkröten Bufo bufo L. In: Zusammenfassungen, Jahrestagung der Deutschen Gesellschaft für Herpetologie und Terrarienkunde, Frankfurt/Main.

Karsen, S.J., Lau, M.W.-N. & Bogadek, A. (1998). Hong Kong Amphibians and Reptiles. Second Edition. Provisional Urban Council, Hong Kong.

Lai, Y.-L. & Ng, Y.-S. (1972). Hong Kong Amphibians. New Asia College Academic Annual (Hong Kong) 14: 79-111.

Lau, M.W.N. (1998). Habitat Use by Hong Kong Amphibians, with Special Reference to the Ecology and Conservation of Philautus romeri. Unpublished

Leader, P.J. (in prep.). The Status of Greater Painted-snipe in Hong Kong. Hong Kong Bird Report (1999).

Lopez, A. and Mundkur, T. (eds). The Asian Waterfowl Census 1994-1996. Results of the Coordinated Waterbird Census and an Overview of the Status of Wetlands in Asia. Wetlands International, Kuala Lumpur.

Maunsell Consultants Asia Ltd. (1997). Main Drainage Channels for Fanling, Sheung Shui and Hinterland: Final Assessment Report.

Maunsell Consultants Asia Ltd. (1999). Planning and Development Study on North East New Territories: Technical Paper 6 Environmental Assessment - Draft Report.

Melville, D.S. (1990). Report on Bird Ringing in Hong Kong in 1989. Hong Kong Bird Report 1989: 65-76

Melville, D.S. (1991). Report on Bird Ringing in Hong Kong in 1990. Hong Kong Bird Report 1990: 86-99

Melville, D.S. and Galsworthy A.C. (1992). Report on Bird Ringing in Hong Kong in 1991. Hong Kong Bird Report 1991: 87-103

Melville, D.S., Young L. and Leader, P.J. (1994). The Importance of Fish Ponds around Deep Bay to Wildlife especially Waterbirds, together with a Review of Potential Impacts of Wetland Loss and Mitigation Measures. World Wide Fund for Nature Hong Kong.

Moores, N. (1996). Baikal Teal in South Korea. Hong Kong Bird Report 1995:231-235.

Müller, S. and Berthold, G. (1996). Fauna/traffic safety. Manual for civil engineers. Laboratoire des voies de circulation (LAVOC), Ecole polytechnique fédérale de Lausanne.

Peking University (1995). Environmental Impact Assessment Study on ShenZhen River Regulation Project. Final Report. Shenzhen River Regulation Office of Municipal Government. Shenzhen, PRC.

Perennou. C., Mundkur, T. and Scott, D.A. (1994). The Asian Waterfowl Census 1987-91: distribution and status of Asian waterfowl. AWB Publication No. 86, IWRB Publication No. 24, Kuala Lumpur and Slimbridge.

Reels, G. (1996). Distribution of large mammals in Hong Kong. Porcupine! 15: 36-38.

Romer, J.D. (1970). Revised annotated checklist with keys to the snakes of Hong Kong. Memoirs of the Hong Kong Natural History Society 8:1-22.

Romer, J.D. (1975). Annotated checklist with keys to the lizards of Hong Kong. Memoirs of the Hong Kong Natural History Society 10:1-13.

Romer, J.D. (1979a). Second revised annotated checklist with keys to the snakes of Hong Kong. Memoirs of the Hong Kong Natural History Society 14:1-23.

Romer, J.D. (1979b). Annotated checklist with keys to the adult amphibians of Hong Kong. Memoirs of the Hong Kong Natural History Society 15: 1-14.

Rose, P.M. and Scott, D.A. (1997). Waterfowl population estimates. (2nd ed.). Wetlands International Publication 44, Wageningen.

TPB PG-No.12B (1999) Town Planning Board Guidelines for Application for Developments within the Deep Bay Area Under Section 16 of the Town Planning Ordinance.

Treweek, J. (1999). Ecological Impact Assessment. Blackwell Science, Oxford UK.

Van den Tempel, M.W. (1993). Vogelslachtoffers door het wegverkeer. Techn. Rapport Vogelbesccherming Nederland 11. Vogelbescherming Nederland. Zeist.

Van der Grift, E.A and Kuijsters, R.M.J. (1998). Mitigation measures to reduce habitat fragmentation by railway lines in the Netherlands. ICOWET February 9-12: 166-170.

Viney, C., K. Phillipps, and C.Y. Lam. (1994). Birds of Hong Kong and South China. Government Printer, Hong Kong.

Wilson, K.D.P. (1995). Hong Kong Dragonflies. Urban Council of Hong Kong, 211pp.

world. Vol. 3. Lynx Edicions, Barcelona.

Young, L. (1997). The significance of drained fish ponds for wintering waterbirds at the Mai Po Marshes, Hong Kong. IBIS, 139 (4): 694-698.

Young, L. and Cha M.W. (1995). The History and Status of Egretries in Hong Kong. Hong Kong Bird Report 1994:196-215.

Young, L.(1998). The importance to ardeids of the Deep Bay fish ponds, Hong Kong. Biological Conservation, 84 (3): 293-300.

Zhuang, X. (1993). Forest succession in Hong Kong. Ph.D. Thesis. The University of Hong Kong.

Zhuang, X.Y. and Yau, M.L. (1998). The Role plantations in restoring degraded lands in Hong Kong. In: Wong, M.H.(ed.), Remediation and Management of Degraded Lands. Ann Arbor Press, Chelsea, Michigan.