4. ECOLOGICAL ASSESSMENT

4.1 Introduction

4.1.1 This Chapter covers ecological issues arising as a consequence of the proposed KT13 stream channelisation works. The objectives of this ecological assessment are as follows:

· to establish an ecological baseline for the KT13 study area, focusing on key habitats and species present;

· to assess the ecological impacts of the proposed channelisation works;

· to detail effective ecological mitigation measures to reduce or eliminate significant impacts;

· to determine whether residual, post-mitigation impacts are acceptable; and

· to assess the post-mitigation acceptability of the proposed project.

4.2 Baseline Ecological Conditions

4.2.1 A habitat survey was conducted in an area of 500 m radius around the existing KT13 streamcourse and the proposed project boundary (Figure 4.1). Appendix C shows the form for ecological field surveys. Twelve habitat types were identified within the Study Area :

• bare ground/works in progress;

• agricultural land;

• fishpond;

• hillside grassland;

• low-lying grassland/fallow land;

• marsh;

• orchard/horticultural land;

• river/stream;

• shrubland;

• urban and industrial area;

• woodland;

• drainage channel.

4.2.2 The same habitat types were found during the dry-season and wet-season surveys, but minor seasonal changes were observed in the habitat areas of fallow fields and agricultural lands. Some areas used as agricultural land in the dry season became fallow fields in wet season. No seasonal variation was found in other types of habitat between the dry and wet seasons. Hence, wet-season habitat surveys were considered to represent the baseline conditions of habitats within the Study Area. The areas of each habitat type within the Study Area and Project Area are listed in Table 4.1 and Table 4.2 respectively. Note that the Project Area refers to the proposed alignment which includes the natural stream courses and the proposed channel within the site boundary, whilst the Study Area includes the proposed alignments of the secondary drainage channel KT13 and the surrounding area up to 500 m from the site boundary.

Table 4.1

Habitat Types and Areas within the Study Area

Habitat type	Area (ha)	%
Bare ground/works in progress	8.70	4.7
Agricultural land	13.32	7.1
Fishpond	2.30	1.2
Hillside grassland	29.44	15.8
Low-lying grassland/fallow land	17.05	9.1
Marsh	1.18	0.6
Orchard/horticultural land	4.66	2.5
River/stream	3.07	1.6
Shrubland	13.50	7.2
Urban/industrial area	64.14	34.4
Woodland	26.69	14.3
Drainage Channel	2.63	1.4
Total	186.68	100

Table 4.2

Habitat Types and Areas within the Project Area

Habitat type	Area (ha)	%
Bare ground/works in progress	0.133	4.32
Fishpond	0.083	2.69
Low-lying grassland/fallow land	0.781	25.35
Marsh	0.009	0.29
Orchard/horticultural land	0.687	22.30
River/stream	1.000	32.46
Urban/industrial area	0.380	12.33
Woodland	0.008	0.26
Total	3.081	100.0

Woodlands

4.2.3 Two areas of woodland are present within the Study Area.

4.2.4 A large woodland is established behind Ma On Kong and Ho Pui Villages. It is dominated by Schefflera octophylla, Pinus massoniana, Aporusa chinensis, Celtis sinensis, Bridelia tomentosa, Cinnamomum camphora, Rhus chinensis and Rhus succedanea. The woodland is mainly natural but some modification by afforestation has occurred. Thus, some commonly used species for forestry plantation such as Tristania conferta, (an exotic tree) can also be found within the woodland. The understorey is quite rich in shrubs including Litsea rotundifolia, Ilex graciliflora, Ilex asprella, Phyllanthus cochinchinensis, Berchemia racemosa, Rhodomyrtus tomentosa, Melastoma sanguineum and Rhaphiolepis indica. An insectivorous herb, Drosera spathulata was observed on some damp rock faces. The Ma On Kong woodland is of much ecological importance because of large size, species diversity, and ecological linkage with the Tai Lam Country Park.

4.2.5 A small woodland present near the Ho Pui Egretry (Plate 4.1) is dominated by Euphoria longan, Rhus chinensis, and Ficus hispida. Euphoria longan is a common fruit tree planted by villagers while Ficus hispida and Rhus chinensis are common wild tree species. A Reeve’s Smooth Skink Scincella reevesii was found in this woodland on the day of the habitat survey. Typical species of butterflies in woodland habitats include Cupha erymanthis, Cyrestis thyodamas, Discophora sondaica, etc. were also observed in the patch of woodland close to the egretry.

Shrublands

4.2.6 Two patches of shrubland are present within the Study Area. One is located behind Ma On Kong Village while another one is located on the hillside east of the toll gate area of the Route 3. The Ma On Kong shrubland (Plate 4.2) is adjacent to the woodland and hence may be of higher ecological importance. The shrub species found are common in (and native to) Hong Kong, including Litsea rotundifolia, Rhodomyrtus tomentosa, Berchemia racemosa, Rhus succedanea, Rhaphiolepis indica, Phyllanthus cochinchinensis, and Wikstroemia indica. The small tree, Aporusa chinensis is also very common in this shrubland. The shrubland east of the toll gate area is isolated from the country park by the Route 3. Dominant plants include Aporusa chinensis, Litsea rotundifolia, Rhodomyrtus tomentosa, Rhus chinensis, Rhus succedanea, Schefflera octophylla, Phyllanthus cochinchinensis, Ficus hirta, Glochidion eriocarpum, Ilex asprella, Macaranga tanarius and Lantana camara. There is dust deposition on the plants at the edge of the shrubland since it is very close to some construction sites. Other disturbance includes cutting, planting, and burning, mainly caused by grave worshippers during Ching Ming and Chung Yeung Festivals.

Marsh

4.2.7 One small freshwater marsh (1.2 ha) was found at Ma On Kong (Plate 4.3). It is close to small areas of abandoned fishponds and the woodland behind the village. The dominant wetland species including Colocasia esculenta, Alocasia macrorrhiza, Commelina nudiflora, and Rumex maritimus are common to Hong Kong. Other freshwater plants such as Polygonum hydropiper, Polygonum perfoliatum, Solanum nigrum, Echinochloa crus-galli and Eleusine indica can also be observed occasionally. Several Little Egrets, Egretta garzetta were observed flying over the marsh during the survey.

Fishponds

4.2.8 Fishponds within the Study Area are very small and scattered. The remaining fishponds located at Ma On Kong are of moderate ecological value as they are linked with a stream, marsh, woodland and some wetland on both sides of the stream and are thus a component of the wetland environs of Ho Pui Egretry. However, the area of these ponds has recently been reduced by filling, considerably decreasing their area and reducing their ecological value, especially as a feeding area for waterbirds. Wetland plants found on the bunds are common species in Hong Kong including Rumex maritimus, Solanum nigrum, Colocasia esculenta, Polygonum hydropiper, Commelina nudiflora, Paspalum sp. and Panicum sp.. Aquatic species, Eichhornia crassipes and Lemna minor were also observed on the water surface of some ponds (Plate 4.4).

Rivers/Streams

4.2.9 Two streams are present within the Study Area; KT 13 which flows through Ho Pui and KT 12 which passes to the south of Cheung Po. The two streams join the primary drainage channel within the West Rail construction area. All vegetation along these sections has been removed and the path was also being changed. The upper watercourse of KT13 is semi-natural with a length over 800m (Plate 4.5). The water quality is poor with strong odor owing to the discharge of domestic sewage and livestock waste. However, the stream passes through the Ho Pui Egretry and is associated with ponds nearby to provide a feeding ground for avifauna. During fauna surveys, several Chinese Pond Herons, Ardeola bacchus and one Little Egret, Egretta garzetta were found foraging in the stream. A Checkered Keelback, Xenochrophis piscator was also found swimming in the downstream from the egretry.

4.2.10 A second natural stream section is present within the Study Area, the mid-course of KT12 to the south of Cheung Po village. The water quality of this section of KT12 is quite good and a locally rare fish Hongkong Bitterling Rhodeus ocellatus is present in this section of the stream. Wetland plants observed include Commelina nudiflora, Solanum nigrum, Alocasia macrorrhiza, Microstegium ciliatum, and Cyperus alternifolius. An ecological evaluation of KT12 and the impacts of proposed channelisation of this stream has been detailed elsewhere (BBV 2002).

Low-Lying Grasslands/Fallow Land

4.2.11 Twelve areas of low-lying grassland/fallow field (Plate 4.6) are present within the Study Area, of which two are moderate in size (4-5 ha) and others are small. The two moderate sized low-lying grasslands/fallow fields located south of the village are very close to the egretry, and thus are of higher ecological importance. The dominant plants species are common in abandoned paddy fields, including Solanum nigrum, Amaranthus spinosus, Panicum maximum, Panicum paludosum, Commelina nudiflora, Eleusine indica, Gnaphalium purpureum, Polygonum perfoliatum, Alocasia macrorrhiza, Ageratum conyzoides, Microstegium ciliatum, Colocasia esculenta, Ipomoea cairica, Mikania micrantha, Rumex maritimus, Polygonum chinense, Hedychium coronarium, etc. A notable species of butterfly Danaid Eggfly (Hypolimnas misippus) was seen in disused agricultural land close to the pig farms at the southern end of the site.

Hillside Grasslands

4.2.12 Several areas of hillside grassland are present within the Study Area (Plate 4.7). Various kind of common grass were found, such as Miscanthus floridulus, Imperata cylindrica, Arundinella setosa, Microstegium ciliatum, Bothriochloa intermedia, Rhynchelytrum repens, Setaria pallide-fusca, etc. A common hillside fern, Dicranopters linearis was also observed. This grassland is a plageoclimax maintained by hillfires.

Agricultural Lands

4.2.13 Several areas of agricultural land are present within the Study Area, of which one is large (over 7 ha) and others are small (about 1 ha). The largest one is located around the Cheung Po village (Plate 4.8). The common crops in these cultivated fields are vegetables such as Ipomoea aquatica, Lactuca sativa and flowers such as Gladiolus gandavensis and Lilium longiflorum. Some fruit trees were also observed here including, Musa paradisiaca, Euphoria longan, and Prunus persica. This type of habitat is widespread in Hong Kong, as well as South China.

Orchard/Horticultural Lands

4.2.14 Three small orchards are present within the Study Area. Common fruit trees Euphoria longan, Litchi chinensis, and Clausena lansium can be found (Plate 4.9). The Ma On Kong egretry is situated in the smallest patch of orchard, to the northwest of the proposed alignment.

4.2.15 The horticultural land is located at Ma On Kong village. Both native and exotic ornamental plants can be found, including Araucaria heterophylla, Chrysalidocarpus lutescens, Livistonia chinensis, Caryota ochlandra, Podocarpus macrophyllus, Bauhinia blakeana, Ficus benjamina, Ficus rumphii, Erythrina speciosa, etc.

Urban and Industrial Areas (UIA)/Bare Ground/Works In Progress

4.2.16 UIA includes villages, roads, animal farmhouses and the Route 3 highway. A large site for the West Rail occupies the area from Cheung Po to Pat Heung Road.

Drainage Channel

4.2.17 The Study Area (but not the Project Area) contains recently constructed concrete-lined drainage channels supporting little or no flora and fauna.

4.3 Floral Survey Results

Overall Species Survey

4.3.1 A total of 205 floral species were recorded within the Study Area, of which 155 are native to Hong Kong, and the other 50 are exotic species (Floral list is given in Appendix E). Aporusa chinensis, Bridelia tomentosa, Celtistetrandra, Cinnamomum camphora, Pinus massoniana, Rhus chinensis, Rhus succedanea, and Schefflera octophylla are the most common tree species present in the woodlands. In the low-lying area, Euphoria longan and Musa paradisiaca are the most abundant trees planted by the villagers for their fruits, while Ficus hispida is the most common wild tree species. An exotic species of Central America origin, Lantana camara is the most common shrub found within the Study Area. Although it is not a native species, it is a valuable food plant for butterflies. Mikania micrantha, an aggressive climber, is common. Like Lantana camara, this is also an exotic species and it is now spreading throughout the SAR. Other common climbers include Ipomoea cairica, Polygonum chinense, and Polygonum perfoliatum.

4.3.2 Three species protected under the Forestry Regulations, Michelia alba, Michelia figo, and Rhododendron championae were found during the floral surveys. All of these species are outside the works area. Michelia alba (Plate 4.10) and Michelia figo (Plate 4.11) were present in the villages of Cheung Po, Ho Pui, and Ma On Kong. These are not naturally occurring but planted by the villagers for amenity purposes. These two species are not included in the floral list as the habitat in which they are located is of low ecological importance. A group of azaleas, Rhododendron championae (Plate 4.12) were found within the shrubland on the hillside behind Ho Pui. These azaleas are far from the Project Area and therefore not expected to be affected by the proposed works. A large banyan tree Ficus tinctoria located in Ho Pui is at least 60 years old. Its location is about 40 to 50 m away from the boundary of the Project Area. It is neither a rare nor protected species. Because of the resistant nature of this species, it is not expected to be affected by the proposed channelization works upstream.

4.3.3 No rare/protected species or species of conservation importance were found within the Project Area and the dominant floral species observed included Bidens pilosa, Cynodon dactylon, Ipomoea cairica, Mimosa pudica, Commelina nudiflora, Alocasia macrorrhiza, Amaranthus spinosus, Panicum maximum, Panicum paludosum, Microstegium ciliatum, Mikania micrantha, Solanum nigrum, and Ipomoea cairica. Ficus hispida and Lantana camara are the most common tree and shrub species found within the Project Area respectively.

Transect Survey

4.3.4 Along the transect placed in the upstream-section alignment, a total of nine and ten species were encountered in five quadrats placed along the transects during dry- and wet-season surveys respectively (Table 4.3a). The transect ran along the stream section that is within the Project Area (Figure 4.2). The common riparian grass, Panicum maximum had the highest percentage cover, 32% and 26% respectively in both seasons. Ficus hispida is the most abundant tree species found along the stream.

Table 4.3a

Belt Transect Floral Survey Results (Upstream Section)

Species	Coverage (%)
	Dry Season	Wet Season
Amaranthus viridis	2.0	1.0
Bidens pilosa	1.0	3.0
Boehmeria nivea	NP	2.0
Cuscuta chinensis	0.4	NP
Eleusine indica	NP	1.0
Ficus hispida	6.0	2.0
Lantana camara	NP	2.0
Microstegium ciliatum	8.6	23.0
Mikania micrantha	5.0	13.0
Panicum maximum	32.0	26.0
Panicum paludosum	7.0	10.0
Solanum nigrum	16.0	NP
Total	78.0	83.0

Note:

1. The percentage values = the mean of the five quadrats

2. NP = not present

4.3.5 In the mid-section alignment, the habitat type was low-lying grassland. A total of 14 and 15 species were recorded during the dry- and wet-season surveys respectively (Table 4.3b). Bidens pilosa, Cynodon dactylon, Ipomoea cairica, Mimosa pudica and Urena lobata were the most common species occupying the grassland.

Table 4.3b

Belt Transect Floral Survey Results (Mid-Section)

Species	Coverage (%)
	Dry Season	Wet Season
Amaranthus viridis	0.4	0.6
Bidens pilosa	2.0	1.8
Bothriochloa ischaemum	NP	0.2
Pueraria phaseoloides	NP	0.2
Cuscuta chinensis	0.1	NP
Cynodon dactylon	1.5	1.4
Digitaria sanguinalis	0.4	0.4
Eleusine indica	0.8	0.6
Erechthites hieraciifolia	NP	0.4
Erigeron floribundus	0.4	0.6
Ipomoea cairica	70	75
Lantana indica	0.1	NP
Mikania micrantha	0.8	0.6
Mimosa pudica	5.5	8.0
Sesbania cochinchinensis	0.2	0.2
Sida rhombifolia	0.2	0.2
Urena lobata	1.5	1.6
Total	83.9	91.8

Note:

1. The percentage values = the mean of the five quadrats

2. NP = not present

4.3.6 The two transect surveys showed that similar vegetation composition between different seasons and all flora species found were very common and typical species throughout the HKSAR.

4.4 Fauna Survey Results

Bird, Amphibian & Reptile, Mammal, Insects, Aquatic Invertebrates and Fish Surveys

4.4.1 Faunal surveys accordingly to the methodology shown in Table 4.4 were conducted on during the period from April to July 2000, as is detailed in Table 4.5. The bird, amphibian, reptile, mammal and insects surveys essentially covered the area 300m from the proposed channel in accordance with the EIA Study Brief. Reference is made to study of active egret nests in August 2000 by Kwok et al (2001). The findings indicate that there were no active nests in Ma On Kong in August 2000.

Overall Fauna Survey Results

(i) Birds

A total of 27 bird species was recorded in an extended survey to a radius of 300 m from the discharge points at KT13 (see Table 4.5a). These included three raptors seen flying over the site: Crested Serpent Eagle and Bonelli’s Eagle on 22 May, and Eurasian Hobby on 10 April. This compares with a total of 45 species recorded during the Hong Kong Bird Watching Society (HKBWS) breeding bird survey (Carey et al. 2001). Results from the two surveys are presented and compared in Table 4.4. Note, however, that the 1 km square which includes most of the Study Area also includes part of the Shek Kong catchwater forest on the hills above. Thus, a number of forest species were recorded in the 1 km square during the HKBWS survey which would not be expected in the present Study Area.

For species where breeding was not proven, likely status is based on Carey et al. (2001), Kwok and Corlett (1999) for forest species, Leven (1998) for farmland species and Leven (2001) for shrubland species. Percentage of 1 km squares in Hong Kong in which a species was present is derived from Carey et al. (2001).

Table 4.4

Faunal Surveys of KT13 – Summary of Methodologies

Faunal Group	Survey Dates	Time / Duration	Methodology
Birds – general surveys	25.05.00 26.05.00 30.05.00 22.06.00	1000-1400h 1000-1400h 1000-1400h 1000-1400h	The entire Study Area of KT13, including areas up to a distance of 300m from the discharge points, was walked on four separate dates during the period May-June 2000, and all wetland-associated birds which were seen utilizing the Study Area (i.e. foraging or perching, but not flying over) were identified with the aid of 8 x 32 binoculars.
Birds – breeding ardeids	10.04.00 30.04.00 22.05.00 26.05.00 22.06.00 26.06.00 01.07.00 16.07.00	0730-0930h 0730-0930h 0730-0930h 0730-0930h 0730-0930h 0730-0930h 0730-0930h 0730-0930h	The two egretries at Ho Pui and Ma On Kong were visited at least once per month in the early morning during the period April to July 2000. During each visit, the following data were collected for each ardeid species present:: number of adults present, number of active nests, number of juveniles, and size of obvious broods. Observations were made with 10 x 50 or 8 x 40 binoculars.
Birds – ardeid flight-lines	29.05.00 03.06.00 14.06.00 23.06.00 01.07.00	0600-1100h 0600-1100h 0600-1100h 0600-1100h 0600-1100h	Observations of flight lines were made from the Shek Kong Water Catchment road on the south side of the valley. This was 80 m above sea level, and as such allowed excellent views of the valley. Observations commenced at or close to dawn, and continued for five hours. This was to coincide with the period of peak activity, which has been shown to significantly higher in the early morning (Young 1993). Observations were made with 10 x 50 or 8 x 40 binoculars, and a 30 x 70 telescope, and involved following individual adult egrets until they landed at the foraging site. At which point the location at which they landed was recorded using a 1:20,000 map of the study area, and the habitat type selected. If birds flew out of sight before they landed, this was noted, and the location at which they were lost from sight. For data analysis, the distance flown was taken as the point at which the bird was lost from sight. While this results in an under estimate of the average distance flown, it is preferred to ignoring these data which would result in even greater bias as bird lost from sight were not surprisingly, relatively distant when this happened. By excluding these data, the average distance flown would be significantly reduced and this would place greater value on those habitats closer to the egretry.
Mammals	22.05.00 25.05.00 26.05.00 30.05.00 22.06.00 16.07.00	1000-1400h + 1930-2130h 1000-1400h 1000-1400h 1000-1400h 1000-1400h + 1930-2130h 1930-2130h	Non-flying terrestrial mammals were surveyed on four days and three nights during the period May-July 2000. During day-time surveys, mammal signs such as scats, prints and burrows were searched for over the entire Study Area, including up to 300m from the discharge points. Any such signs encountered were recorded and species identification was made based on the surveyor’s knowledge of the signs left by different mammal species in Hong Kong. During night surveys, the surveyor walked the Study Area with a strong flashlight and any mammals encountered were identified on sight.
Reptiles	22.05.00 25.05.00 26.05.00 30.05.00 22.06.00 16.07.00	1000-1400h + 1930-2130h 1000-1400h 1000-1400h 1000-1400h 1000-1400h + 1930-2130h 1930-2130h	Reptiles were surveyed on four days and three nights during the period May-July 2000. During day-time surveys, the entire Study Area, including up to 300m from the discharge points, was walked and the surveyor investigated microhabitats such as piled material, large stones, tree trunks, buildings, drainage channels and the stream channel itself. All reptiles observed were identified in the field, occasionally with the aid of binoculars. During night surveys, the Study Area was walked and foraging reptiles were spotlighted using a strong flashlight, and identified on site.
Amphibians	22.05.00 25.05.00 26.05.00 30.05.00 22.06.00 16.07.00	1000-1400h + 1930-2130h 1000-1400h 1000-1400h 1000-1400h 1000-1400h + 1930-2130h 1930-2130h	Amphibians were surveyed on four days and three nights during the period May-July 2000. During day-time surveys, the entire Study Area, including up to 300m from the discharge points, was walked and the surveyor identified all species observed. During night surveys, the Study Area was walked with a strong flashlight and all amphibians seen were identified. In addition, amphibians were identified from male breeding vocalizations.
Fish	09.05.00 15.05.00 12.07.00	Daytime	As the streams are heavily polluted, visual observation was employed.
Butterflies	25.05.00 26.05.00 30.05.00 22.06.00	1000-1400h 1000-1400h 1000-1400h 1000-1400h	Butterflies were surveyed on four days, during the warmest part of the day, during the period May-June 2000. In each survey the entire Study Area, including up to 300m from the discharge points, was walked and all adult butterflies observed were identified. Identifications were facilitated by use of close-focusing 8 x 32 binoculars. A long-handled net was occasionally employed to catch species which required examination in the hand for proper identification.
Dragonflies	25.05.00 26.05.00 30.05.00 22.06.00	1000-1400h 1000-1400h 1000-1400h 1000-1400h	Dragonflies were surveyed on four days, during the warmest part of the day, during the period May-June 2000. In each survey the entire Study Area, including up to 300m from the discharge points, was walked and all adult dragonflies observed were identified. Identifications were facilitated by use of close-focusing 8 x 32 binoculars. A long-handled net was occasionally employed to catch species which required examination in the hand for proper identification.

Table 4.5a

Bird Species Recorded in the KT 13 Study Area during the Present Study and in the HKBWS Breeding Bird Survey and Their Probable Status

Species		Recorded in:		Per Cent of 1 km squares in HK	Breeding habitat	Likely current status in Study Area*
Common name	Scientific name	Present study	HKBWS breeding survey
Little Grebe	Tachybaptus ruficollis	-	Present	2.8	Fishponds	Former breeder**
Little Egret	Egretta garzetta	Breeding	Breeding	14.5	Tree and bamboo clumps near wetlands	Breeding at egretry
Cattle Egret	Bubulcus ibis	Breeding	Breeding	9.7	Tree and bamboo clumps near wetland agriculture	Breeding at egretry
Chinese Pond Heron	Ardeola bacchus	Breeding	Breeding	21.2	Tree and bamboo clumps near wetlands	Breeding at egretry
Black Baza	Aviceda leuphotes	-	Present	5.5	Forest, forest-shrubland mosaic	Summer visitor in catchwater forest
Black Kite	Milvus migrans	-	Present	37.7	Forest, forest-shrubland mosaic	Casual visitor
Crested Serpent Eagle	Spilornis cheela	One sighting	Present	7.0	Forest	Occasional visitor from forest
Crested Goshawk	Accipiter trivirgatus	-	Breeding	9.6	Forest	Resident in catchwater forest
Bonelli’s Eagle	Hieraaetus fasciatus	One sighting	-	N.A.***	Mountains	Casual visitor
Eurasian Hobby	Falco subbuteo	One sighting	Present	2.1	Scattered trees in grassland	Casual summer visitor
White-breasted Waterhen	Amaunornis phoenicurus	Present	Present	18.2	Wetland, streams	Resident along streams
Rock Dove	Columba livia	-	Present	10.8	Anthropogenic habitats	Resident around settlement
Spotted Dove	Streptopelia chinensis	Present	Present	62.4	Anthropogenic habitats	Resident
Emerald Dove	Chalcophaps indica	-	Present	4.3	Forest, shrubland	Resident in forest, shrubland
Chestnut-winged Cuckoo	Clamator coromandus	-	Present	9.7	Forest, shrubland	Summer visitor in forest, shrubland
Large Hawk Cuckoo	Hierococcyx sparverioides	-	Present	19.7	Forest, fung shui woods	Summer visitor in forest, fung shui woods
Indian Cuckoo	Cuculus micropterus	-	Present	21.1	Forest, shrubland	Summer visitor in forest, shrubland
Common Koel	Eudynamys scolopacea	Present	Present	40.0	Farmland, anthropogenic habitats	Resident
Greater Coucal	Centropus sinensis	Present	Present	31.8	Farmland, grassland	Resident
Lesser Coucal	Centropus benghalensis	-	Present	27.7	Grassland	Resident in grassland
Collared Scops Owl	Otus bakkamoena	-	Present	4.5	Forest, fung shui woods	Resident in forest, fung shui woods
Little Swift	Apus affinis	Present	Present	39.2	Aerial feeder, breeds in towns	Casual visitor
Great Barbet	Megalaima virens	-	Present	6.4	Forest	Resident in catchwater forest
White-Throated Kingfisher	Halcyon smyrnensis	Present	-	21.7	Shrubland, farmland, near wetland and streams	Breeding resident near stream
Barn Swallow	Hirundo rustica	Present	Breeding	50.3	Farmland, anthropogenic habitats	Breeding summer visitor
White Wagtail	Motacilla alba	-	Present	5.8	Farmland	Breeding resident
Red-Whiskered Bulbul	Pycnonotus jocosus	Present	Breeding	71.8	Shrubland, anthropogenic habitats	Breeding resident
Chinese Bulbul	Pycnonotus sinensis	Present	Breeding	87.2	Ubiquitous	Breeding resident
Long-Tailed Shrike	Lanius schach	Present	Breeding	33.8	Farmland, grassland	Breeding resident
Oriental Magpie Robin	Copsychus saularis	Present	Breeding	45.4	Anthropogenic habitats	Breeding resident
Blue Whistling Thrush	Myophonus caeruleus	-	Present	18.9	Forest, shrubland	Breeding resident
Masked Laughingthrush	Garrulax perspicillatus	Present	Breeding	51.2	Farmland, anthropogenic habitats	Breeding resident
Hwamei	Garrulax canorus	-	Present	43.3	Shrubland	Breeding resident in shrubland
Yellow-Bellied Prinia	Prinia flaviventris	Present	Breeding	62.4	Grassland, grassland-shrubland mosaic	Breeding resident
Plain Prinia	Prinia inornata	-	Present	8.2	Grassland near wetland, wetland	Former resident**
Common Tailorbird	Orthotomus sutorius	Present	Breeding	67.1	Forest, fung shui woods, shrubland	Breeding resident
Great Tit	Parus major	Present	Breeding	37.4	Forest, fung shui woods	Breeding resident
Fork-Tailed Sunbird	Aethopyga christinae	-	Breeding	16.4	Forest, fung shui woods, shrubland	Breeding resident
Japanese White-Eye	Zosterops japonicus	Present	Present	52.9	Forest, fung shui woods, shrubland	Breeding resident
White-Rumped Munia	Lonchura striata	Present	Breeding	11.2	Forest, fung shui woods, shrubland	Breeding resident
Eurasian Tree Sparrow	Passer montanus	Present	****	49.4	Anthropogenic habitats	Breeding resident
Black-Collared Starling	Sturnus nigricollis	Present	Breeding	29.2	Farmland, anthropogenic habitats	Breeding resident
Common Myna	Acridotheres tristis	Breeding	-	2.4	Farmland	Breeding resident
Crested Myna	Acridotheres cristatellus	Present	Breeding	53.1	Farmland, anthropogenic habitats	Breeding resident
Black Drongo	Dicrurus macrocercus	-	Present	32.3	Farmland, anthropogenic habitats	Breeding resident
Blue Magpie	Urocissa eythrorhyncha	-	Breeding	17.4	Forest	Breeding resident
Common Magpie	Pica pica	-	Present	40.0	Anthropogenic habitats	Breeding resident

* Where current status is not qualified by reference to a particular habitat, status refers to the immediate environs of KT 13 (within c. 50 m) and hence a species is likely to be impacted by the development project.

** Suitable habitat in the Study Area has been destroyed by filling since the HKBWS survey.

*** Exact distribution not published by Carey et al. (2001) for security reasons.

**** Breeding map omitted by Carey et al. (2001) in error.

4.4.2 The habitat utilization of this total of 47 species recorded from the Study Area in respect of the habitats present in the Study Area is indicated in Table 4.5b. Note that this Table includes species which are not tied to one habitat hence the totals are more than the total number of species recorded. For example, White Wagtail regularly makes use of streams and favours this habitat but is not dependent upon it.

Table 4.5b

Characteristic Habitat Utilization and Status of Bird Species Recorded from the Study Area

Habitat type	Resident (probably breeding)	Non-Breeding Visitor
Bare ground / works in progress	0	0
Agricultural land	15	7
Fishpond	10	3
Hillside grassland	6	5
Low-lying grassland / fallow field	10	3
Marsh	4	4
Orchard / horticultural land	27	4
Plantation	11	5
River / stream	7	4
Shrubland	12	8
Woodland	13	4
Mountains*	2	3
*Habitat not present in the Study Area

4.4.3 Thus, the majority of bird species recorded in the Study Area are those of anthropogenic habitats (especially cultivated land) or woodland and shrubland. Species in the former category are common and widespread in Hong Kong (see Table 4.5a above) whilst the woodland and shrubland in the Study Area is remote from the development area. Accordingly, those species most likely to suffer adverse impacts which are of conservation significance are those wetland-dependant species which habitually utilize streams and streamside as detailed in Table 4.6.

Table 4.6

Wetland Dependent Species Recorded in the Study Area

Species	Status in Study Area
Little Egret	Breeding in egretry; limited use of the stream for foraging (see Flight-line study below).
Cattle Egret	Breeding in egretry, feeds in wet agricultural land.
Chinese Pond Heron	Breeding in egretry; limited use of the stream for foraging (see Flight-line study below).
White-breasted Waterhen	Several individuals recorded foraging in the stream; probably dependent on this habitat.
White-throated Kingfisher	Observed in the downstream section of KT 13. May depend on the stream for nesting and, in part, for feeding

(ii) Amphibians and Reptiles

Five anurans, five lizards and three snakes were recorded at KT13 and to 300 m from the discharge points, as follows:

Asian Common Toad Bufo melanostictus- several individuals spotted both upstream and downstream of the egretry.

Günther’s Frog Rana guentheri - present at scattered localities across the Study Area.

Paddy Frog Rana limnocharis - present at scattered localities across the Study Area.

Brown Tree Frog Polypedates megacephalus - one individual in vegetation beneath the egretry and one on the stream bank vegetation downstream of the egretry.

Asiatic Painted Frog Kaloula pulchra - present at scattered localities across the Study Area.

Bowring’s Gecko Hemidactylus bowringii - individuals seen on several buildings across the Study Area.

Changeable Lizard Calotes versicolor – recorded in shrubby vegetation beside the road which runs parallel to the west of the downstream section.

Chinese Skink Eumeces chinensis - one individual on the path next to the stream, c. 100 m upstream of the egretry.

Long-tailed Skink Mabuya longicaudata - one individual on the road beside the egretry; one in Ho Pui village adjacent to the stream.

Reeve’s Smooth Skink Scincella reevesii - one individual in woodland beneath the egretry.

Checkered Keelback Xenochrophis piscator - one individual swimming in the stream c. 50m downstream of the egretry.

Red-necked Keelback Rhabdophis subminiatus – a juvenile seen on a unsurfaced track close to the northern alignment.

Chinese Cobra Naja atra - one dead individual on the stream bank close to the egretry (Plate 4.13).

All species are common and widespread in Hong Kong (Karsen et al., 1998).

(iii) Mammals

A dead Musk Shrew Suncus murinus was found in Ho Pui village on 16 July (Plate 4.14). This species is considered to be common in New Territories villages (Hill & Phillipps, 1981).

(iv) Dragonflies & Damselflies

A total of 24 species were recorded across the extended study site (see Table 4.7), the most notable records being those of the stream damselflies Prodasineura autumnalis and Pseudagrion rubriceps. Several individuals of the large stream libellulid Zygonyx iris were observed soaring over the woodland behind Ho Pui village. This species breeds in clear, fast-running streams (see Wilson, 1995). Hydrobasileus croceus and Neurothemis tullia, in disused agricultural land at the upstream section, were other notable records.

Table 4.7

Odonates Observed in the Study Area

Ischnura senegalensis

Ceriagrion auranticum

Copera marginipes

Prodasineura autumnalis

Pseudagrion rubriceps

Brachydiplax chalybea

Orthetrum glaucum

Orthetrum luzonicum

Orthetrum pruinosum

Orthetrum sabina

Brachythemis contaminata

Crocothemis servilia

Diplacodes trivialis

Neurothemis tullia

Neurothemis fulvia

Pseudothemis zonata

Trithemis aurora

Trithemis festiva

Rhyothemis variegata

Tramea virginia

Hydrobasileus croceus

Pantala flavescens

Tholymis tillarga

Zygonyx iris

No. of species recorded: 24

(v) Butterflies

A substantial total of fifty species were recorded across the extended study area (see Table 4.8). This is a significantly larger number of species than was recorded at KT2, KT4-7, KT12 and KT14-15 where a range of 9 – 26 species was recorded during the same time period with a similar survey effort (BBV, 2002). The most notable species recorded was Danaid Eggfly Hypolimnas misippus seen in disused agricultural land close to the pig farms at the southern end of the site. Some species more typical of woodland habitats, e.g. Cupha erymanthis, Cyrestis thyodamas, Discophora sondaica, Euthalia phemius, Faunis eumeus and Pantoporia hordonia and Melanitis leda were recorded in the patch of woodland close to the egretry.

Table 4.8

Butterfly Species Observed in the Study Area

Euploea midamus

Ideopsis similis

Elymnias hypermnestra

Lethe confusa

Lethe rohria

Melanitis leda

Mycalesis mineus

Ypthima baldus

Faunis eumeus

Discophora sondaica

Euthalia phemius

Cyrestis thyodamas

Charaxes bernardus

Ariadne ariadne

Athyma selenophora

Cupha erymanthis

Hestina assimilis

Hypolimnas bolina

Hypolimnas misippus

Junonia atlites

Kaniska canace

Neptis hylas

Pantoporia hordonia

Symbrenthia lilaea

Abisara echerius

Zemeros flegyas

Iraota timoleon

Acytolepis puspa

Famegana alsulus

Zizeeria maha

Zizina otis

Pieris canidia

Hebomoia glaucippe

Eurema blanda

Eurema hecabe

Graphium agamemnon

Graphium doson

Graphium sarpedon

Papilio bianor

Papilio demoleus

Papilio helenus

Papilio memnon

Papilio paris

Papilio polytes

Papilio protenor

Astictopterus jama

Erionota torus

Parnara guttatus

Polytremis lubricans

Potanthus confucius

(vi) Aquatic Invertebrates

Aquatic invertebrates were not sampled due to heavily polluted state of the water.

(vii) Fish

The surveys were conducted on 9^th and 15^th May and 12^th July 2000. No fish were observed.

Discussion

4.4.4 All of the survey work reported above was undertaken on a limited time-scale in one season (summer), and therefore cannot purport to be a fully comprehensive record of the fauna likely to be present. In particular, seasonal birds are under-represented due to the survey timing, which has excluded most winter visitors and passage migrants. However, it is considered that the key concern with respect to avifauna is the potential impact on wetland dependent bird species, in particular the ardeids using the Ho Pui egretry. No additional pertinent data would have been gathered in this respect from a dry (winter) season survey. Accordingly, notwithstanding these caveats, the following observations can be drawn.

Birds

4.4.5 The site is considered to be potentially of ecological importance for a breeding (or potentially breeding) bird species if it meets the following criteria:

• It supports a breeding population of a species which occurred in less than 5% of 1 km grid squares in Hong Kong in the HKBWS breeding bird survey (Carey et al., 2001);

• Or, it supports 5% or more of the Hong Kong breeding population of a species;

• Or, it supports 1% or more of the Hong Kong population of a species for which the Hong Kong population is considered to be of global significance by Carey et al. (2001);

• Or, it supports a significant population of a species which is declining in Hong Kong; the definition of “significant” in this context to be assessed on a species-by-species basis with reference to the relevant literature, in particular Fellowes et al. (2002) and Carey et al. (2001).

4.4.6 Species of conservation importance or concern could be recorded in the Study Area, but in circumstances under which the Study Area population does not satisfy the above criteria. This applies, in particular, to wide-ranging species such as raptors where a casual observation does not indicate that the habitat over which the bird is passing is of any importance to the species. In the present study, this exclusion applies to raptors such as Bonelli’s Eagle and Crested Serpent Eagle which are species of mountains and forest respectively and would not make use of Study Area habitats to any significant extent.

4.4.7 Another difficulty applies to species which are likely to be under-recorded during general surveys because they are cryptic or nocturnal. Such species include owls, which are usually only recorded at night (night-time surveys were not a requirement of the HKBWS breeding survey). Caution, has, therefore, to be applied in assessment of the conservation importance of cryptic species in order that undue importance is not attached to sightings of such species.

4.4.8 As a further point, exotic species are not considered to be of conservation significance unless the population meets global threat criteria as defined in BirdLife (2000).

4.4.9 Accordingly, Table 4.9 indicates those species of potential conservation concern recorded in the Study Area and highlights those which meet the foregoing criteria and are, therefore, of actual conservation concern with respect to this EIA.

Table 4.9

Bird Species of Potential and Actual Conservation Concern with respect to Their Use of the Study Area

Note: Those species of actual conservation concern (the presence of which is significant in determining the acceptability of the proposed development and for which appropriate mitigation measures should be entertained) for this EIA are shown in bold.

Species	Conservation concern criterion to be satisfied	Is criterion satisfied?
Little Grebe	Breeding population of a species occurring in less than 5% of 1 km squares in Hong Kong	No, already eliminated from the Study Area by habitat destruction.
Little Egret	Supports at least 5% of the Hong Kong breeding population of a species.	No, 4 prs. at Ho Pui is only 1.6% of 2000 Hong Kong breeding population.
Cattle Egret	Supports at least 5% of the Hong Kong breeding population of a species.	Yes, 18 prs. at Ho Pui comprises 27% of 2000 Hong Kong breeding population.
Chinese Pond Heron	Supports at least 5% of the Hong Kong breeding population of a species.	Yes, 6 prs. at Ho Pui is 5% of 2000 Hong Kong breeding population.
Bonelli’s Eagle	Breeding population of a species occurring in less than 5% of 1 km square in Hong Kong.	No, casual observation of a bird flying over does not indicate that the Study Area is of importance for this species.
Eurasian Hobby	Breeding population of a species occurring in less than 5% of 1 km squares in Hong Kong.	No, casual observation of a bird flying over does not indicate that the Study Area is of importance for this species.
Emerald Dove	Breeding population of a species occurring in less than 5% of 1 km squares in Hong Kong.	Yes, recorded in 4.3% of 1 km squares in Hong Kong, indicating requirement for semi-natural forest. However, potential habitat in the Study Area is remote from the development area.
Collared Scops Owl	Breeding population of a species occurring in less than 5% of 1 km squares in Hong Kong.	Yes, recorded in 4.5% of 1 km squares in Hong Kong. However, a nocturnal species which is probably much more widespread than HKBWS survey suggests.
White-throated Kingfisher	Declining as a breeding species in Hong Kong.	Yes, presence in suitable habitat suggests that it may breed in the Study Area.
Common Myna	Breeding population of a species occurring in less than 5% of 1 km squares in Hong Kong.	No, this is an exotic species with a huge global range.

4.4.10 Thus, as noted in Table 4.9, the bird species of conservation concern present in the Study Area are Cattle Egret and Chinese Pond Heron (breeding colony of importance), Emerald Dove and Collared Scops Owl (restricted breeding distribution in Hong Kong) and White-throated Kingfisher (declining in Hong Kong). However, Emerald Dove is unlikely to use any habitat close to the development area, whilst Collared Scops Owl probably only meets the conservation concern category because it is an elusive nocturnal species. Accordingly, the key species, the needs of which should be taken account when the acceptability of the proposed development is evaluated, are Cattle Egret, Chinese Pond Heron and White-throated Kingfisher. Since the Ho Pui egretry supported 27% of the Hong Kong breeding population of Cattle Egrets in 2000, any potential impacts on this species are of great concern.

Amphibians and Reptiles

4.4.11 Thirteen species of herpetofauna were recorded at KT13. Five of the six snake species encountered during the surveys were shared between the Ma On Kong and Ho Pui sites, including one species – Checkered Keelback Xenochrophis piscator – usually associated with wetland habitats.

4.4.12 Significantly, no obligate stream species (amphibian or reptile) were recorded, undoubtedly as a consequence of severe pollution and other anthropogenic disturbance at all sites. However, seasonal marshes are present at KT13, and due consideration should be given to protection of the area during stream channelization works at the site.

4.4.13 Sightings of herpetofauna within the KT 13 channel involved only one snake (Checkered Keelback Xenochrophis piscator). Channelization would remove refuges from the streams and therefore make them even less attractive for breeding amphibians. Rehabilitation of the rivers, by reduction or removal of major pollution sources (e.g. pig farms), would, conversely, lead to greater use by amphibians and perhaps eventual recolonization by obligate stream species such as Green Cascade Frog Rana livida.

4.4.14 With the exception of Checkered Keelback none of the reptile species encountered is particularly associated with wetland habitats.

Dragonflies & Damselflies

4.4.15 KT13 was rich in odonates (24 species). Obligate stream species were encountered in this channel (P. autumnalis, P. rubriceps and Zygonyx iris). The presence of these species at the heavily polluted KT13 may be attributed to non-breeding colonization from the nearby, slightly polluted, KT12.

4.4.16 Although some lowland dragonflies (e.g. Pantala flavescens, Brachythemis contaminata, Orthetrum sabina) can breed successfully in almost any body of water, many species require a specific substrate such as submerged vegetation, woody or weedy margins, or coarse gravel as oviposition site, and/or larval habitat, and partially submerged objects such as boulders or emergent vegetation as larval emergence sites (see e.g. Corbet, 1999). Insensitive channelization which resulted in removal of such substrates would therefore also remove many breeding odonates, including the seemingly ubiquitous Ischnura senegalensis, as well as more sensitive species.

Butterflies

4.4.17 Butterflies are not especially associated with wetland habitats. However they tend to be numerous and species-rich where floral diversity is high, and where woodland edge is a common element of the landscape. There were 50 species recorded in KT13 during the extended survey.

4.4.18 Only one rare species, Danaid Eggfly Hypolimnas misippus, was recorded during the surveys. This appeared to be an irruptive species during 2000 (G.T. Reels, pers. obs.).

Aquatic Invertebrates

4.4.19 KT13 was very badly polluted and hence aquatic invertebrates were not surveyed. However, it is reasonable to assume that aquatic invertebrate communities would show characteristics intermediate between those of the KT2 and KT7 streams (see BBV (2002)).

4.4.20 It is highly likely that a reduction in the pollution load in the channels would lead to eventual re-colonization by pollution-sensitive species. River channelization, on the other hand, would inevitably lead to a loss of appropriate microhabitats (e.g. coarse gravel, boulders, trailing vegetation, and submerged woody material) for many invertebrate species, and a consequent decline in species diversity. Any deposits on the channel bottoms would be dominated by fine sediments and organic particles, which can clog the gills and feeding apparatus of some animals, as well as restricting water movement within the substrate, thereby reducing oxygen levels (Dudgeon & Corlett, 1994).

Fish

4.4.21 No fish were observed during the site visits. The streams are heavily polluted, making it impossible for fish fauna to survive.

Overall stream characteristics

4.4.22 Despite the stream itself being heavily polluted at the present time, the KT13 Study Area is a rare and ecologically significant lotic habitat, characterized by a gravel streambed and meandering platform; streams with this type of geomorphology are restricted to northern New Territories, which have been heavily impacted by pollution and, more recently, large-scale channelization. Channelization not only removes the unique habitat, it also swamps out all the aquatic fauna from this species rich habitat.

4.5 Egretry and Flight-line Survey Results of Ho Pui and Ma On Kong Egretries

4.5.1 A safe nest site is a fundamental requirement for successful breeding in birds. In colonial species such as ardeids (herons and egrets), pressure for breeding sites is thought to be a major factor influencing coloniality (Kushlan and Hafner 2000). Coloniality brings with it many ecological advantages such as predator avoidance and assistance in locating the most profitable food resources. With it also come potentially major disadvantages, not least of all vulnerability to disturbance or loss of the nesting site, which for local populations at least can have severe negative impacts.

4.5.2 The purpose of this aspect of the study was to survey the two egretries that may be affected as part of the project, and to investigate foraging areas and habitats of the ardeids during the breeding season.

Egretry Surveys

4.5.3 The Ho Pui and Ma On Kong egretries (Figure 4.3) were visited on 10^th and 30^th April, 22^nd and 26^th May, 22^nd and 26 June and 1^st and 16^th July. Plate 4.15 and Plate 4.16 show the two egretries. Plate 4.17 shows a Chinese Pond Heron in the Ma On Kong Egretry. The data collected were compared to published data for the area, notably Young (1993), Young and Cha (1995), Carey (1998), and Wong et al. (undated).

4.5.4 Results

(i) Population size

A maximum of 42 pairs of egrets was present in the two egretries. Three species were found breeding Chinese Pond Heron Ardeola bacchus (20 pairs) Cattle Egret Bulbulcus ibis (18 pairs) and Little Egret Egretta garzetta (4 pairs). The details of these are given in Tables 4.10 and 4.11.

Table 4.10

Number of Pairs of Breeding Egrets at Ho Pui during Summer 2000

	April	May	June	Maximum
Chinese Pond Heron	6	6	3	6
Little Egret	4	2	2	4
Cattle Egret	10+	18	8	18
No. of pairs	20+	26	13	28

Table 4.11

Number of Pairs of Breeding Egrets at Ma On Kong during Summer 2000

	April	May	June	Maximum
Chinese Pond Heron	0	4+	14	14
Little Egret	0	0	0	0
Cattle Egret	0	0	0	0
No. of pairs	0	4+	14	14

(ii) Brood Size

Brood sizes recorded at the two egretries are given in Tables 4.12 and 4.13.

Table 4.12

Brood Size and Frequency at Ho Pui Egretry during June 2000

Brood size	Chinese Pond Heron	Little Egret	Cattle Egret
1	0	0	0
2	0	0	2
3	3	3	5

4.5.5 At least 47 Cattle Egret chicks or juveniles were noted on 22 June 2000, representing an average of 2.6 young per pair for 18 pairs.

Table 4.13

Brood Size and Frequency at Ma On Kong Egretry during June 2000

Brood size	Chinese Pond Heron
1	0
2	0
3	5

4.5.6 At least 40 young were present on 22 June 2000, representing an average of 2.85 young per pair for 14 pairs.

Discussion

Population Size

4.5.7 Combined, the two egretries at Ma On Kong and Ho Pui accounted for 27% and 16% of the Hong Kong breeding population for Cattle Egret and Chinese Pond Heron respectively in 2000 (Kwok et al., 2001). The percentage of the Little Egret breeding population was, however, only 2%. The Hong Kong population of Chinese Pond Heron is of global importance, and that of Little Egret may be of regional importance (Carey and Young, 1998).

4.5.8 Numbers of breeding pairs of egrets increased at both Ho Pui and Ma On Kong during 2000. At Ho Pui, this ended the decline noted there since 1997 (Figure 4.4). However, the long-term decline of Chinese Pond Heron continued, although there was a slight increase during 2000, the six pairs breeding are considerably lower than the peak of 32 in 1991. The overall increase in 2000 was due to higher Cattle Egret numbers, which were close to the peak of 22 pairs in 1997.

Figure 4.4

Number of Pairs of Egrets Breeding at Ho Pui during 1991-2000

4.5.9 At Ma On Kong, numbers of breeding egrets increased by 50% during 1999 (Figure 4.5). Only Chinese Pond Heron breeds at Ma On Kong, and although both Cattle Egret and Little Egret bred in 1995, they have not since. Numbers of Chinese Pond Herons have increased steadily since 1997, although there was a marked drop from 12 to 3 pairs between 1996 and 1997.

Figure 4.5

Number of Pairs of Egrets Breeding at Ma On Kong during 1995-2000

4.5.10 The increase in numbers of Chinese Pond Herons at Ma On Kong may in part explain the decrease at Ho Pui, although even if the numbers at the two egretries are combined there are marked fluctuations overall, indicating additional factors. However, as can be seen from Figure 4.6, in recent years combined numbers of pairs of Chinese Pond Herons breeding at Ma On Kong and Ho Pui have been more stable.

4.5.11 A new egretry found at Kam Tin Shui Mei (grid reference 974856 on Lands Department Map HM20C) during summer 2000 may also be of influence. At least 14 pairs of Chinese Pond Heron were present at this egretry during 2000 (Table 4.14). The age of the Kam Tin Shui Mei egretry is not known, and it is impossible to determine whether Chinese Pond Herons breeding there originated from Ho Pui. It is notable that at 34 pairs, the number of Chinese Pond Herons breeding at all three egretries in the Kam Tin Valley in 2000 is only two pairs greater than the peak of 32 pairs at Ho Pui in 1991 (Figure 4.6).

Table 4.14

Number of Pairs of Breeding Egrets at Kam Tin during Summer 2000

	April	May	June	Maximum
Chinese Pond Heron	nc	14	12	14
Little Egret	nc	2	2	2
Cattle Egret	nc	0	0	0
No. of pairs	nc	16	14	16

nc = not counted

Figure 4.6

Numbers of Chinese Pond Herons Breeding at Ho Pui, Ma On Kong and Kam Tin Shui Mei during 1991-2000

4.5.12 Unfortunately, no data exist on breeding success for previous years for the three egretries covered here. However, limited details of breeding success given in Wong et al. (undated) at other egretries in Hong Kong are of interest in this context. A summary of these is given in Table 4.15.

Table 4.15

Summary of Brood Size in Hong Kong Egretries during 1999

(after Wong et al. undated).

Brood size	Chinese Pond Heron	Little Egret	Cattle Egret
1	3	23	1
2	0	26	0
3	1	12	0
4	0	1	0

4.5.13 While these results are not directly comparable, they do indicate that the breeding success at the three egretries under investigation was high during 2000.

4.5.14 Cattle Egrets are less piscivorous than other egrets; they primarily feed on insects, amphibians and reptiles (Voisin, 1991). As such they prefer freshwater wetlands to saline or brackish habitats (Wong et al., undated). Food availability is a major factor influencing chick mortality, with starvation being noted in some studies as the major cause of mortality in Cattle Egret chicks (Siegfried, 1972). The direct affects of weather on chick mortality in Hong Kong are unknown, but are likely, in some years at least, to be significant.

Environmental Factors Influencing Utilization

4.5.15 Some nests sites have apparently been lost in the past year through clearance of bamboo for the construction of village houses (R. Fan pers comm. to P. J. Leader). In addition, the invasive exotic creeper Mikania micrantha appeared to be colonizing parts of the main egretry. This species has been known to cause sections of other egretries in Hong Kong to become abandoned (Young and Cha, 1995), and is potentially a serious problem.

Summary

4.5.16 There was an increase in numbers of breeding ardeids at both Ho Pui and Ma On Kong during 2000. A further egretry was located at Kam Tin Shui Mei during the summer. Results of surveys from all three egretries indicate high breeding success. It is suggested that the 2000 breeding season was generally good for egrets, and that the increase in number of egretries in the northwest New Territories may not indicate a long term increase in the local breeding population.

4.5.17 Studies elsewhere (Hafner, 2000) have identified the potential for disturbance at egretries to have a severe negative impact on breeding success, and in order to reduce this likelihood, a buffer area of 100 m from the egretries where construction works should not take place is recommended for the period April to September. The location of the buffer area is shown on Figure 4.13.

Changes in numbers and distribution of egretries during 2000 - 2005

4.5.18 Since a period of five years has elapsed since the egretry survey data was collected, the following updated information has been added detailing changes in the numbers and distribution of breeding egrets during the period 2000 – 2005 (Sources: Hong Kong Bird Watching Society Egretry Counts and Asia Ecological Consultants Ltd. unpubl. data).

Table 4.16

Number of Pairs of Breeding Egrets at Ho Pui 2000 - 2005

	2000	2001	2002	2003	2004	2005
Chinese Pond Heron	2	2	1	0	3	0
Little Egret	2	2	2	0	1	0
Cattle Egret	9	5	17	12	9	0
Total	13	9	20	12	13	0

Table 4.17

Number of Pairs of Breeding Egrets at Ma On Kong 2000 - 2005

	2000	2001	2002	2003	2004	2005
Chinese Pond Heron	6	5	12	17	15	16
Little Egret	0	0	0	0	0	0
Cattle Egret	0	0	0	0	0	0
Total	6	5	12	17	15	16

Table 4.18

Number of Pairs of Breeding Egrets at Kam Tin Shui Mei 2000 - 2005

	2000	2001	2002	2003	2004	2005
Chinese Pond Heron	15	11	0	0	0	0
Little Egret	1	3	0	0	0	0
Cattle Egret	0	3	0	0	0	0
Total	16	17	0	0	0	0

Table 4.19

Number of Pairs of Breeding Egrets at Tung Shing Lei (Au Tau) 2000 - 2005

	2000	2001	2002	2003	2004	2005
Chinese Pond Heron	7	3	5	7	19	37
Little Egret	21	6	27	16	38	52
Cattle Egret	0	0	0	0	0	13
Total	28	9	32	23	57	102

Table 4.20

Total Number of Pairs of Breeding Egrets in the Kam Tin Area 2000 - 2005

	2000	2001	2002	2003	2004	2005
Chinese Pond Heron	30	21	18	24	37	53
Little Egret	22	5	29	16	39	52
Cattle Egret	9	8	17	12	9	13
Total	61	34	64	52	85	118

4.5.19 The main trend in egretry distribution over the past six years is the concentration of breeding pairs at Tung Shing Lei, which was first utilized in 2000, since 2004. Meanwhile, the egretry at Shui Mei was abandoned in 2001 and that at Ho Pui was deserted in 2005, whilst the number of pairs at Ma On Kong showed an increase in 2002 and has since remained stable.

4.5.20 The increase in the importance of the Tung Shing Lei egretry has been accompanied by an increase in numbers of Chinese Pond Herons and Little Egrets, but not of Cattle Egrets. Numbers of this species have fluctuated between eight and 17 pairs, with no clear trend. Cattle Egrets were the main species utilizing the Ho Pui egretry during 2000 – 2004, and it seems likely that in 2005 the birds which formerly nested at Ho Pui moved to Tung Shing Lei, which was utilized by Cattle Egrets for the first time that year.

4.5.21 The reason for the abandonment of the Ho Pui egretry in 2005 is not certain. Abandonment of egretries in Hong Kong has usually been attributed to a loss of wetland habitat, disturbance or colonization of the nest trees by climbing plants (Young and Cha 1995). There was no observed change to the trees forming the egretry between 2004 – 2005 and no apparent increase in disturbance. However, a change in land use in this part of the Kam Tin valley may be responsible; some of the nearby agricultural fields utilized by foraging Cattle Egrets in 2004 were planted with maize in 2005, making them unsuitable for use by egrets (J. Allcock, pers. obs.).

4.5.22 Historically, there has been a pattern of rather frequent changes in location of egretries in Hong Kong (e.g. Young and Cha 1995, Wong and Woo 2004). Young and Cha (1995) found no cases where an egretry that was abandoned was re-occupied in subsequent years, and no such re-occupation has been noted during the period from 1995 – 2005 (M.R. Leven pers. obs., Wong and Woo 2004). There is, therefore a significant possibility that the Ho Pui egretry will be permanently abandoned.

Flight Line Surveys

4.5.23 A common method of investigating feeding habitat use by ardeids involves following their flight lines. Individual egrets are observed as they leave the colony and land at their feeding sites. This methodology has been used in a number of studies in Hong Kong (Young 1993, Cornish 1996, Wong 1990, and Wong et al. undated). It is not always possible to follow all individuals, as many are lost from sight. It is possible to follow up on birds lost from sight by identifying habitats being utilized by egrets in the direction of lost birds. However, the results of this approach may be biased towards larger feeding areas that contain higher numbers or densities of birds, or habitats where birds spend longer though they are not foraging. Using this method it is easy to over-estimate the importance of some areas and thus possibly some habitats (Young, 1993).

4.5.24 In this project, investigation of flight lines has been used to address specific questions relating to habitat type and location of the foraging areas of breeding adults at the two egretries. This information was considered essential to establish whether the construction works to be undertaken would impact upon the breeding egrets through habitat loss, or disturbance of foraging areas. Disturbance of the breeding sites is dealt with elsewhere.

4.5.25 Flight line observations were undertaken on 29^th May, 3^rd, 14^th and 23^rd June and 1^st July 2000. On this later date, it was clear that many birds had left the egretry as the majority of the young had fledged.

Results

4.5.26 During the fieldwork, exactly 200 individual egrets were followed as they left either Ho Pui or Ma On Kong. Most birds (89%) were from Ho Pui, and related to Cattle Egret, Little Egret, Chinese Pond Heron, and Night Heron. Further details are given for all species except for Night Heron, which related to a single individual on one date, and was not recorded during the breeding surveys.

(i) Distance flown

Details of distance flown by the birds followed are given in Table 4.21.

Table 4.21

Details of Distances Flown by Egrets at Ho Pui and Ma On Kong Egretries,

Summer 2000 (All measurements in meters)

	All birds	Little Egret	Cattle Egret	Chinese Pond Heron	Chinese Pond Heron (Ho Pui only)	Chinese Pond Heron (Ma On Kong only)
Minimum	80	500	150	80	80	200
Maximum	5700	4700	5700	4040	4040	2900
Average	2178.5	1640.9	2471.6	1317.4	1544.2	1187.0
s.d.	1218.9	1091.7	1216.6	787.3	1075.1	643.2
n.	199*	22	137	40	18	22

*excludes a single Black-crowned Night Heron

4.5.27 The number of individual egrets in each 100 m band radiating out from the egretry was also calculated. These data are presented in Figure 4.7.

Figure 4.7

Distance from Egretry (m), and Number of Birds for Ho Pui and Ma On Kong,

Summer 2000.

(ii) Flight Direction

Direction of flight was divided into 8 sectors, and this was further sorted into distance (Figure 4.8) following Young (1993). Data for each species, with those data for Chinese Pond Heron separated for the two egretries, are given in Figures 4.9 to 4.12.

(iii) Habitat Selection

Of the birds followed, habitat selection was noted for 45%. The remainder either were lost from sight, or landed at a location where it was not possible to determine the habitat (as this was obscured, often by buildings). Details of habitat selection are given in Table 4.22.

Table 4.22

Habitat Selection of Egrets at Ho Pui and Ma On Kong Egretries, Summer 2000.

	All birds	Little Egret	Cattle Egret	Chinese Pond Heron
Active dry agriculture	1 (1%)	0 (0%)	0 (0%)	1 (4%)
Active wet agriculture	39 (43%)	3 (60%)	21 (32%)	17 (74%)
Inactive wet agriculture	36 (40%)	1 (20%)	33 (51%)	3 (13%)
Grass	8 (9%)	0 (0%)	8 (12%)	0 (0%)
Trees	6 (7%)	1 (20%)	3 (5%)	2 (9%)
Total	90 (100%)	5 (100%)	65 (100%)	23 (100%)

Discussion

Distance Flown

4.5.28 As can be seen from Figure 4.7, there was a peak of distance flown from the egretry at about 800 m. Below this distance numbers were very low and there was a general increase (although with marked fluctuations) between 1000 m and 3000 m. The distance 1000 to 3000 m accounted for 62.6% of all flights. It is therefore of potential concern that all of the proposed drainage channels at KT13 and KT12 also fall within this range, with implications on habitat loss, and construction disturbance, of foraging areas.

Flight Direction

4.5.29 Flight direction was concentrated within the NNE and ENE sectors (Figure 4.8) for all three species breeding at Ho Pui, and within the ENE and SEE sectors for the Chinese Pond Herons breeding at Ma On Kong. At Ho Pui, the NNE and ENE sectors accounted for 45% of Little Egrets, 75% of Cattle Egrets, and 83% of Chinese Pond Herons. At Ma On Kong, the ENE and SEE sectors accounted for 68% of the Chinese Pond Herons breeding there.

4.5.30 Given that both egretries lie at the southwest end of a northeast-southwest orientated valley, which has shrub-covered hills to the south and west, and wetlands to the north and east, it is not surprising that the northeast holds the main foraging areas. However, it is again of note that the proposed works for KT12, KT13, KT14 and KT15 are either north to east from Ho Pui or northeast to south east from Ma On Kong, i.e. within the preferred sectors at both.

Habitat Selection

4.5.31 The importance of wet agricultural areas, as foraging sites for the egrets breeding at Ho Pui and Ma On Kong are apparent from Table 4.17. Active and inactive wet agriculture accounted for 83% of the habitat selected by all birds. It was in fact similar for all three species: 80% for Little Egret, 83% for Cattle Egret and 87% for Chinese Pond Heron. There were differences in utilization of active and inactive wet agriculture, with both Little Egret and Chinese Pond heron preferring active wet agriculture, and Cattle Egret preferring inactive wet agriculture.

4.5.32 Rivers, creeks and drainage channels have been recorded as foraging areas for ardeids in similar Hong Kong studies. Young (1993) found that up to 37% of Chinese Pond Herons at Tsim Bei Tsui and up to 22% at Mai Po foraged in nullahs. Wong (1991) found creeks to be the second most important habitat for Little Egrets after fishponds at Mai Po. The reason for the absence of sightings of birds using streams in the present study may simply be a consequence of the relatively small number of observations. It is also possible that some of the birds lost from sight during the observations were foraging within the areas to be channelized. However, it may be that other factors (for example stream width, substrate, adjoining vegetation, pollution and disturbance) resulted in the streams in the Study Area genuinely being little used by ardeids. Nevertheless, on a precautionary basis, it is suggested that since Chinese Pond Herons have frequently been observed using streams elsewhere in Hong Kong, direct disturbance to the KT13 stream should be minimized.

Summary

4.5.33 The flight line data showed that the preferred areas of foraging ardeids were also those where KT12 and KT13 are to be constructed. Whilst it appears that birds make little use of the KT13 stream for foraging, other Hong Kong studies have demonstrated the importance of streams for feeding Chinese Pond Herons. Accordingly, on a precautionary basis, loss of potential feeding habitat (i.e. a natural stream bottom) in KT13 should be avoided. The works program should avoid any construction works in key feeding areas and in KT13 (existing stream course) next to the Ho Pui Egretry site during the period from April to September.

Figure 4.8

Flight Direction and Distance Flown from the Egretry

Figure 4.9

Direction of Flight and Distance Traveled for Little Egret

Little Egret (n=21)

Note: Underlined figures are total for that sector.

Figure 4.10

Direction of Flight and Distance Traveled for Cattle Egret

Cattle Egret (n=137)

Note: Underlined figures are total for that sector.

Figure 4.11

Direction of Flight and Distance Traveled for Chinese Pond Heron

Chinese Pond Heron (Ho Pui) (n=18)

Figure 4.12

Direction of Flight and Distance Traveled for Chinese Pond Heron

Chinese Pond Heron (Ma On Kong) (n=22)

4.6 Ecological Evaluation of Different Habitats

4.6.1 The site (KT13) is generally an important area, supporting good communities of birds, herpetofauna, dragonflies and butterflies. Until 2005 the most important ecological feature of the stream was the presence of the egretry to the east of Ho Pui village. The egretry was not occupied in 2005 but it is not possible at this point to state whether or not it has been permanently abandoned. Tables 4.23 - 4.33 provide an evaluation of the ecological value of the eleven habitats within the Study Area.

Table 4.23

Ecological Evaluation of Woodlands

Criteria	Ma On Kong	Ho Pui
Naturalness	Mainly natural but modified by afforestation	Semi-natural
Size	Large (26.08 ha)	Small (0.61 ha)
Diversity	High	Moderate
Rarity	Not rare	Not rare
Re-creatability	Re-creatable in long term (over 30 years)	Re-creatable in long term (over 30 years)
Fragmentation	Continuous	Fragmented by footpaths, roads
Ecological linkage	Connected with the Tai Lam Country Park	Adjacent to the Ho Pui Egretry
Potential value	High	High if the egretry is re-occupied
Nursery/breeding ground	The woodland potentially provides a breeding ground for birds, butterflies, and reptiles.	Until 2005 the woodland contained an egretry which was an important nursery and breeding ground for Cattle Egret. Little Egret and Chinese Pond Heron were also recorded as breeding there. The woodland potentially provides a breeding ground for other birds, butterflies and reptiles.
Age	Over 40 years	Over 40 years
Abundance/ richness of wildlife	The woodlands contain moderate species richness.	The woodlands contain moderate species richness.
Ecological value	Moderate - high	High in area occupied by egretry, otherwise moderate

Table 4.24

Ecological Evaluation of Shrublands

Criteria	Ma On Kong	East of Route 3
Naturalness	Natural	Mainly natural but with little modification by grave worshippers
Size	Moderate within the Study Area (4.5 ha)	Moderate (9 ha)
Diversity	Moderate	Moderate
Rarity	Not rare	Not rare
Re-Creatability	Readily re-Creatable	Readily re-Creatable
Fragmentation	Continuous	Lightly fragmented by some footpaths
Ecological linkage	Connected with the Tai Lam Country Park	Some linkage with woodland
Potential value	Moderate	Moderate
Nursery/breeding ground	Moderate value	A potential breeding ground for birds, reptiles and insects..
Age	At an early successional stage, the habitat is young	At an early successional stage, the habitat is young
Abundance/ richness of wildlife	Moderate	Moderate
Ecological value	Low - moderate	Low - moderate

Table 4.25

Ecological Evaluation of Marsh

Criteria	Remarks
Naturalness	May be derived from abandoned agricultural fields
Size	Small (1.18 ha)
Diversity	Low to moderate floral diversity but moderate to high in terms of wildlife supported
Rarity	Not rare, but becoming rarer
Re-creatability	Can be re-created within a short time for such young marsh
Fragmentation	No fragmentation
Ecological linkage	Close to the Ho Pui Egretry and provides suitable feeding conditions for breeding egrets
Potential value	Moderate
Nursery/breeding ground	The marsh is potential breeding ground for birds, reptiles and insects.
Age	Young
Abundance/ richness of wildlife	Moderate abundance
Ecological value	Moderate

Table 4.26

Ecological Evaluation of Fishponds

Criteria	Remarks
Naturalness	Man-made habitat
Size	Small (2.3 ha)
Diversity	Low habitat diversity but high in terms of wildlife supported
Rarity	Not rare, but becoming rarer
Re-creatability	Can be re-created artificially
Fragmentation	-
Ecological linkage	Some ponds are adjacent to the egretry
Potential value	Ponds at egretry are of high value
Nursery/breeding ground	The ponds are potential breeding grounds for birds and insects
Age	No information is available
Abundance/ richness of wildlife	Moderate
Ecological value	Moderate

Table 4.27

Ecological Evaluation of Stream (KT13)

	Criteria		Remarks
	Naturalness		Natural
	Size		Moderate (about 1km in length within the Study Area, 3 ha)
	Diversity		Moderate
	Rarity		Not rare but lowland unchannelised streams rapidly becoming rarer
	Re-Creatability		Re-creatable
	Fragmentation		The stream is continuous but the path of the section with the West Rail Working Site is being changed
	Ecological linkage		The stream passes through the egretry location
Potential value		The site has potential as a relatively ecologically important area, if the pollution load in the stream can be brought down.
	Nursery/breeding ground		Prior to 2005, an important breeding ground for egrets and Chinese Pond Heron, as well as invertebrates (including dragonflies and butterflies), amphibians, reptiles and common fish species.
	Age		Old (over 50 years)
	Abundance/ richness of wildlife		High abundance.
	Ecological value		Moderate overall, high in vicinity of the egretry (if egretry present)

Table 4.28

Ecological Evaluation of Low-lying Grasslands/Fallow Land

Criteria	Remarks
Naturalness	Derived from agricultural lands
Size	Small to moderate (from 17.05 ha)
Diversity	Low to moderate
Rarity	Common
Re-creatability	Re-creatable within a short time (1 to 2 years)
Fragmentation	Fragmented by footpaths, local roads
Ecological linkage	Adjacent to woodland, streams, ponds, shrubland and orchard
Potential value	Some area of this habitat are of higher value as they are adjacent to the egretry
Nursery/breeding ground	Potential breeding ground for reptiles, amphibians and insects.
Age	Young
Abundance/ richness of wildlife	Moderate
Ecological value	Moderate

Table 4.29

Ecological Evaluation of Hillside Grasslands

Criteria	Remarks
Naturalness	Natural
Size	Moderate to large in size within the Study Area (29.44 ha)
Diversity	Low
Rarity	Not rare
Re-creatability	Easily re-creatable naturally or artificially within 1 to 2 years
Fragmentation	Continuous
Ecological linkage	Connected with the Tai Lam Country Park
Potential value	Low but can be improved if hill-fire ceases and succession is allowed
Nursery/breeding ground	Low value
Age	Relatively recent
Abundance/ richness of wildlife	Low
Ecological value	Low

Table 4.30

Ecological Evaluation of Agricultural Lands

Criteria	Remarks
Naturalness	Man-made habitat for crop production
Size	Small to large (from 13.32 ha)
Diversity	Low
Rarity	Widespread
Re-creatability	Re-creatable artificially
Fragmentation	Fragmented by footpaths, roads, and other areas of urban environment such as villages
Ecological linkage	May act as movement corridors and foraging habitat for animals from woodlands nearby
Potential value	Low, but can be higher by applying some management practice to enhance the habitat heterogeneity
Nursery/breeding ground	Not important
Age	From recent to old
Abundance/ richness of wildlife	Low
Ecological value	Low

Table 4.31

Ecological Evaluation of Orchard/Horticultural Land

Criteria	Remarks
Naturalness	Man-made habitat
Size	Small (4.66 ha)
Diversity	Low
Rarity	Not rare
Re-creatability	Can be re-created artificially
Fragmentation	Fragmented by footpaths, roads
Ecological linkage	May provide foraging habitat for birds
Potential value	Low
Nursery/breeding ground	Feeding ground for birds and invertebrates. One patch of litchee orchard at Ma On Kong contains an egretry which is an important nursery and breeding ground for Chinese Pond Herons.
Age	From recent to old
Abundance/ richness of wildlife	Moderate abundance but generally very common and widespread species
Ecological value	Generally low, but moderate in the case of the litchee orchard at Ma On Kong used as an egretry.

Table 4.32

Ecological Evaluation of Urban

and Industrial Area(UIA)/Bare Ground/Works in Progress

Criteria	Remarks
Naturalness	Man-made habitat with heavy disturbance from human activities
Size	Small to large (72.84 ha)
Diversity	Very low
Rarity	Not rare
Re-creatability	Artificially re-creatable
Fragmentation	Fragmented within the Study Area
Ecological linkage	None
Potential value	Very low
Nursery/breeding ground	Not important
Age	From recent (Route 3) to old (Ma On Kong village)
Abundance/ richness of wildlife	Very low
Ecological value	Very low

Table 4.33

Ecological Evaluation of Drainage Channel

Criteria	Remarks
Naturalness	Man-made habitat with heavy disturbance from human activities
Size	Small (2.63ha)
Diversity	Very low
Rarity	Not rare
Re-creatability	Artificially re-creatable
Fragmentation	Not particularly fragmented
Ecological linkage	None
Potential value	Very low
Nursery/breeding ground	Not important
Age	Recent
Abundance/ richness of wildlife	Very low
Ecological value	Very low

4.7 Potential Ecological Impacts

Identification of Ecological Impacts

4.7.1 This section identifies all potential impacts caused by the unmitigated channelization works within the Study Area during the construction and operation phases. The next section will evaluate these ecological impacts.

Construction Impacts

4.7.2 Activities during the construction phase include site clearance, site formation, dredging, construction of the bypass culvert, paving of access road, etc. These are likely to have the impacts described below on fauna and flora species and the habitats in which they live.

(i) Habitat Loss /Damage (Direct Impacts)

The direct impact of site clearance works on the existing stream channel is removal of riparian vegetation, loss of natural banks with woody or weedy margins and loss of substrate including submerged vegetation and/or coarse gravel. There will therefore be a direct impact on vegetation and non-vagile animal groups, notably invertebrates and amphibians. There will also be direct habitat loss for all other animal groups using the stream corridors, notably birds, amphibians, reptiles and dragonflies (which have vagile adults but non-vagile larvae).

The project will result in the direct permanent loss of habitats within the project site area as detailed in Table 4.34 below.

Table 4.34

Direct habitat loss arising from construction of the secondary drainage channel KT13

Habitat	Area (ha)	Ecological value of habitat
Bare ground	0.133	Very low
Fishpond	0.083	Moderate
Low-lying grassland / Fallow land	0.781	Moderate
Marsh	0.009	Moderate
Orchard / Horticultural land	0.687	Low
River / Stream	1.000	Moderate
Urban / Industrial area	0.380	Very low
Woodland	0.008	Moderate (because egretry not affected)
Total	3.081

The area of moderate value habitat to be impacted is about 1.881 ha. This comprises a very small area of fishpond, woodland and marsh and larger areas of low lying grassland/fallow land as well as the stream itself. The woodland area which would be lost does not form part of the area occupied by the egretry prior to 2005. It is located at the eastern bank of the meander outside the CA zone (Figure 4.1). Whilst the woodland at Ho Pui as a whole is considered to be of high ecological value if the egretry is present, the area to be lost, which does not form part of the egretry area, is considered to be only of moderate value, as is the woodland as a whole if the egretry is not re-occupied.

Approximately 59 trees will be retained under this Project, while approximately 60 trees will be transplanted and approximately 74 trees will inevitably need to be felled. No rare/protected or species of conservation importance were found.

Of the habitat to be lost, a total of 0.042 ha falls within the CA zone. However, the area to be lost comprises grassland habitat rather than woodland. A total of 8 trees will be affected within the CA zone of which 4 trees will be transplanted and 4 trees will inevitably be felled.

(ii) Habitat Loss /Damage (Indirect Impacts)

Construction activities are likely to have immediate effects on the water table in the low-lying wet grassland derived from abandoned paddy fields, abandoned fish ponds and adjacent freshwater. Lowering of the water table may arise as a consequence of increased drainage, whereas compaction or deliberate or accidental stream diversions may result in temporary flooding.

(iii) Disturbance of Wildlife (Indirect Impacts)

Indirect impacts will arise through disturbance activities affecting disturbance-sensitive animal groups, primarily birds (mammals are also disturbance-sensitive but, as is described above, mammal diversity is low). Disturbance impacts for some bird species may be severe. Though ardeids do show some habituation to regular disturbance sources, they are likely to be particularly sensitive to irregular disturbance associated with construction activities.

Potential Disturbance to the Egretry

The proposed works associated with KT13 are, in part, in very close proximity to the area occupied by the egretry prior to 2005. This applies, in particular to the bypass culvert which is proposed to be constructed immediately adjacent to the north part of the egretry site (Figure 4.3). This close proximity is well within the distance within which ardeids are sensitive to disturbance and therefore if the egretry is re-occupied in subsequent years and if construction activity takes place during the breeding season it is likely to have a considerable negative impact on nest site availability and breeding success and, in a worst case scenario, could lead to partial or total desertion of the egretry.

Some disturbance effects may arise even if construction activity does not take place when the egretry is occupied if it renders all or part of the egretry more accessible and vulnerable to casual disturbance by people. The scale of disturbance which nesting egrets will tolerate is variable. Regular low-key human activities, such as people walking along a path, are generally tolerated, especially where these are already occurring before the beginning of the breeding season. Egrets are more sensitive to irregular disturbance, especially if these involve intrusive activities such as cutting vegetation (even if this does not extend to the nest trees themselves) and/or if they commence after the egretry has been occupied.

In the present case, it is considered that the construction of the bypass culvert would render the egretry (if it is re-occupied) more vulnerable to accidental casual disturbance by people as the removal of vegetation to permit the bypass culvert to be built will leave some of the nest trees more exposed. This disturbance could either be active (i.e. it might be easier for people to gain access to the egretry) or passive in that nesting egrets might avoid nest sites which have become more exposed.

Depending upon the scale of such disturbance and the response of the egrets, there is some risk that the northern section of the former egretry site would be unsuitable for breeding in subsequent years. This, however, remains a hypothetical concern since the egretry was not occupied in 2005 and, if re-occupation occurs, birds will respond depending upon current conditions at the time.

Other Potential Disturbance Impacts

Any changes in the water table of the wet grassland areas would have short-term effects on the fauna of these areas, including changes in their suitability for breeding amphibians and dragonflies and other invertebrates. This would have a consequent effect on their value as feeding areas for ardeids and other waterbirds. This, if combined with construction activity and human disturbance during the breeding season, may have a negative impact on egret numbers and their breeding success.

(iv) Fragmentation of Habitats

The proposed channelized stream passes through a large area of low-lying grassland southwest of Ho Pui village. Site clearance of the Project Area along the stream will increase fragmentation of the grassland. This fragmentation will be increased by the creation of the bypass culvert which will isolate a small section of the grassland in the north-east. Fragmentation will result in barriers to dispersal, especially by less vagile species such as amphibians. In addition, the smaller fragments may fall below a minimum area threshold for utilisation by more disturbance sensitive species such as waterbirds.

(v) Dust Deposition on Vegetation

Some construction activities will generate suspended particulates that can cause vegetation damage, which in turn, have secondary effects on associated fauna, such as insects and birds.

(vi) Increased Sediment Load

During the construction phase, dust and exposed soil may enter the stream through run-off, especially during heavy rainfall periods. However, as mentioned in the habitat evaluation section, the water quality in existing watercourse is already poor. Ecological degradation of the affected habitat may not occur if pollutant levels due to run-off does not exceed those of the watercourse.

Operational Impacts

(i) Habitat Loss

The stream sections within the Project Area are proposed to be widened and channelized. The habitats on both sides of the stream (as well as the microhabitats of the main channel itself) will be lost permanently. Operational maintenance work such as removal of sediment and vegetation will hinder any recolonization of the channelized stream and will prevent all but a transitory recovery of its plant and animal community. As with construction, disturbance may have a severe effect on disturbance-sensitive taxa. If maintenance work was undertaken during the ardeid breeding season disturbance could have significant effects including, at worst, abandonment of the egretry (should it be re-occupied in subsequent years).

Construction will result in permanent loss of some habitats of moderate value including low-lying grassland, woodland and fishpond and orchard/horticultural land. The area of habitats to be lost is small and the areas to be lost are already highly fragmented, hence the direct impacts of the loss of habitat are expected to be small. In general, small and highly fragmented patches of habitat are less species rich and have lower potential to support important populations of species of conservation interest than large, unfragmented habitat patches. Small and highly fragmented patches of habitat are also less likely to support species which are dependent upon that particular habitat type. Hence the direct impacts of the loss of habitat are expected to be small.

If water levels in the wet grassland areas are lowered permanently they will progressively change to dry grassland and, if succession is allowed to proceed, to shrubland. Species associated with wet grassland including a number of dragonflies will be permanently lost. In addition, feeding areas for ardeids will be permanently lost.

(ii) Ecological Barrier

On completion of the project, the channel will become an ecological barrier. Some animals may experience difficulty crossing these barriers when they attempt to move between habitats.

(iii) Continuing Disturbance Impacts

During the initial operational period a part of the area occupied by the egretry prior to 2005 will remain more vulnerable to disturbance than it was prior to the commencement of the project. However, as noted above this is a hypothetical concern only since the egretry was not occupied in 2005 and, if re-occupation occurs, birds will respond depending upon current conditions at the time.

Evaluation of Ecological Impacts

Construction Impacts

(i) Habitat Loss /Damage

The removal of riparian vegetation would impact negatively and significantly on ‘skulking’ birds such as White-Breasted Waterhens. It may also affect White-throated Kingfisher, removing potential nest site locations (bare vertical earth banks) and roosting/hunting perches. The loss of appropriate microhabitats for many invertebrate species will lead to decline in species diversity.

Intra-seasonal effects on the water table in the low-lying wet grassland derived from abandoned paddy fields, abandoned fish ponds and adjacent freshwater are unlikely to have significant adverse impacts as the wetlands are seasonal and the plant and animal communities will be made up of species which are tolerant of short-term changes in water levels. However, there may be some short term effects on the availability of feeding areas for ardeids. If such changes result in a reduction of food availability during the breeding season then ardeid breeding success may be compromised.

The impact of habitat damage due to land used for construction, supporting facilities and other storage areas should largely be temporary. After completion of the works the affected habitats can be restored. Habitats that may be affected include fishponds, low-lying grasslands/fallow fields, woodland, orchard, horticultural land and bare ground. The areas of these habitats to be impacted are small, and the construction period will be relatively short, whilst the habitats are re-creatable within two to three years (at least 20 years, in the case of woodland). Accordingly, the significance of this impact on ponds, woodland and low-lying grasslands/fallow fields is moderate and reversible; while for orchard, horticultural land, and bare ground impacts will be low and reversible.

No rare/protected or tree species of conservation importance will be affected.

(ii) Disturbance to Wildlife

The most significant impacts on wildlife are potential disturbance of breeding egrets should the Ho Pui egretry be re-occupied in subsequent years. The proximity of the Ma On Kong egretry to Route 3 attests to egrets’ ability to acclimatize to traffic disturbance, although the increase in numbers, following the construction period suggests some impacts during that phase of the project. However, the construction disturbance at the Mai Po Village egretry for the New Territories Circular Road in 1987 to 1989, a drainage channel in spring/summer 1990, and the installation of new telephone cables in 1993 are considered factors that have contributed to the decline of this egretry (Young and Cha 1995). Numbers of nesting egrets at Mai Po Village egretry have declined from a peak of 520 (Young and Cha 1995) to 105 in 1999 (Wong et al. undated).

Human intrusion and indirect egretry disturbance linked to human activities are major factors affecting breeding success of the birds. Intense disturbance may cause birds to desert a nesting site permanently and move elsewhere. There is considerable evidence that this is already happening in Hong Kong, with short-term occupation of egretry sites at Kam Tin and in the Lam Tsuen Valley suggesting that egrets may be being forced to utilize suboptimal sites. These factors may have contributed to the abandonment on the Ho Pui egretry in 2005, however, as is discussed in paragraph 4.5.21 it is suspected that in this case the proximal cause of desertion is changes in local farming practice.

Disturbance can be especially disastrous during the early stages of the breeding cycle, when eggs and small young, if left unattended, can be easily lost or die. Conversely, if disturbance prevents to commencement of breeding activities it is of less significance, especially if birds move, and breed successfully, elsewhere. This appears to be what occurred in 2005 where the abandonment of the Ho Pui egretry in 2005 was matched by an increase in numbers at Tung Shing Lei and the breeding of Cattle Egrets (formerly the main breeding species at Ho Pui) there for the first time.

Overall, therefore, and if the egretry should be re-occupied, subsequent disturbance effects on egrets there could range from insignificant and temporary (if the birds adapt readily to disturbance or merely move within the egretry) to moderate (if disturbance displaces egrets and these are unable to breed elsewhere). Within this range it is considered that the disturbance effects will most likely be insignificant or small and the probability of moderate impacts is considered to be low. However, it is considered that the most likely scenario is that the egretry will not be re-occupied and hence there would be no disturbance impacts.

White-throated Kingfisher has been identified as vulnerable to habitat loss due to potential loss of nest sites and perches. Should nest sites be lost there may be a permanent, irreversible, effect leading to the loss of a breeding territory of this species. However, in the case of the present project there is no evidence that actual nest sites will be affected and suitable nest site locations will remain in the section of stream which is not to be channelised. As this is a relatively low density species (Carey et al. 2001) it is unlikely that more than one territory exists within the project area. Accordingly, since a significant area of suitable habitat will remain it is considered unlikely that the project prevent utilisation of the stream and loss of a breeding territory and potential ecological impacts on this species are considered to be low.

Even temporary habitat loss may result in permanent loss of populations of non-vagile taxa of invertebrates. However, given the highly polluted state of the stream it is most unlikely that such populations exist. Accordingly, it is considered that the scale of such impacts is likely to be very low.

(iii) Fragmentation of Habitats

During the construction phase, the natural sections of the stream and the adjacent grasslands will be fragmented. These two habitats are important components of the egretry environment and, in addition, support a number of avifauna, herpetofauna, dragonflies and butterflies. The significance of this impact on such habitat is considered to be low if the egretry is not re-occupied (as the other species are relatively common and widespread) to moderate if the egretry is re-occupied prior to commencement of construction.

(iv) Dust Deposition on Vegetation

Potentially a large area could be affected by dust deposition. However, this situation is temporary and reversible, therefore the significance is low.

(vi) Increased Sediment Load

As reported in habitat survey results, the water quality of the existing stream is already very poor. Thus, the significance of this temporary impact is low.

Operational Impacts

(i) Habitat Loss

The stream sections inside the Project Area will be widened and deepened after the proposed channelization works. Areas adjacent to the stream are expected to be lost permanently. If water levels in the wet grassland areas are lowered permanently they will progressively change to dry grassland and, if succession is allowed to proceed, to shrubland. Species associated with wet grassland including a number of dragonflies will be permanently lost. In addition, feeding areas for ardeids will be permanently lost. Thus, the significance of this impact on this habitat is moderate, for the size of land loss is small (<0.5ha).

Construction of the bypass culvert will result in permanent loss of a small area of low-lying grassland. The direct habitat loss is small, however the loss of any feeding areas of low-lying grassland near to the egretry is of some significance as it exacerbates a trend of continuing habitat loss. Overall, however, given the small area of habitat involved, the significance of the habitat loss is considered to be small.

(ii) Ecological Barrier

The widening and re-alignment of the stream will enlarge the ecological barrier for wildlife wishing to cross over the stream. However, there is no evidence that there are significant populations of non-vagile taxa which might be affected and hence the significance of this impact is considered to be low.

(iii) Disturbance to Wildlife

The creation of the access track associated with the bypass culvert will open up the northern section of the egretry site to significantly higher levels of casual disturbance than is the case at present. Deliberate or accidental human disturbance is, therefore, more likely to occur if the egretry is re-occupied and this re-occupation includes the northern section of the former egretry site. Potential impacts, therefore range from none (if the egretry is not re-occupied) to low (if the egretry is re-occupied prior to the construction phase and then is partially displaced) to moderate (if there has been significant abandonment during the construction phase and re-occupation does not take place). However, given that the egretry was not utilized in 2005, the most likely scenario is that the egretry will not be re-occupied and hence there will be no impacts.

In the longer term, the proposed planting of bamboos along the eastern side of the by-pass culvert will serve to screen the egretry site from human disturbance (and will, in addition, provide potential nesting habitat). Accordingly, it is considered that there will be no residual adverse impacts to the egretry site in the long term.

Cumulative Impact from Concurrent Projects

4.7.3 The KT13 Channel is the most southerly of several drainage projects being undertaken in the Northwest New Territories. Other projects within the vicinity of the KT13 construction are listed in Chapter 2 of this EIA. Most of the works for other channels are at least 1 km from the KT13 channel with the exception of works proposed at KT12 (see BBV 2002). These projects are predicted to have a cumulative impact in reducing the areas of natural streamcourses and permanent or seasonal wetlands (marshes and active or inactive agricultural land) and hence will have a cumulative impact in reducing the areas available for wetland-dependant fauna. However, if mitigation measures proposed in BBV (2002) are implemented these impacts are predicted to be reduced to a low (and acceptable) level.

4.7.4 In the wider area of the Kam Tin valley there have been a number of development projects in recent years which have had significant effects in reducing the areas of natural or semi-natural wetland. These include the construction of the Route 3 road link, channelisation of the Kam Tin River (shown on Figure 1.1), the construction of the KCRC West Rail line and the Kam Tin Bypass. In addition to direct loss of permanent or seasonal wetlands due to the land-take of these projects, dumping of excavated fill material from the first three of these projects has resulted in a major reduction in the size of the permanent and seasonal wetlands to the west of Kam Tin, popularly known as the “Buffalo Fields”.

4.7.5 Whilst there are no direct cumulative linkages between the potential impacts of the proposed channelisation of KT13 and these projects, there may be some indirect effects as a consequence of the long-term progressive reduction in foraging areas for ardeids. Indeed, habitat loss in the wider area may have been a factor in the abandonment of the Ho Pui egretry in 2005, though the proximal cause of this was considered to be a change in agricultural practice (i.e. not a consequence of development activity) in an area that had formerly been a favoured feeding site for Cattle Egrets. Such cumulative egrets remain of concern in respect of the Ma On Kong egretry as much of the affected area lies within the potential foraging range of nesting egrets (Wong et al. 1999). However, since the main wetland area (the Buffalo Fields) lies more than 3 km from the egretry and only 18% of foraging flights recorded from the egretry were of more than 3 km (see paragraphs 4.5.23 – 4.5.33), the effects of these projects, and hence cumulative impacts on the Ma On Kong egretry, are considered to be minor.

4.7.6 As a corollary, much more significance must be attached to cumulative impacts of developments within 3 km of the Ma On Kong egretry, as it is within this zone that 72% of egrets breeding in the egretry were observed to forage (see paragraphs 4.5.23 – 4.5.33). In this zone, as well as the effects of the channelisation of other streamcourses (see paragraph 4.7.3), major cumulative impacts arise as a consequence of small-scale progressive loss of wetland habitat. Such progressive loss of wetland habitat is likely to continue as, in addition to those projects designated under the EIA Ordinance and Government projects for which an Environmental Study is required, small-scale developments conforming to the Outline Zoning Plan will continue to take place.

4.7.7 Much of the Study Area, including all of the triangle formed by the villages of Ho Pui, Ma On Kong and Tai Kek is zoned for Village development. Continuing progressive loss of non-urbanised areas to urbanization is therefore predicted. This habitat loss is likely to include many of the remaining fishponds and permanent and seasonal wetlands. Continued reduction in the areas of habitat available to wetland-dependant fauna will, therefore, occur. In the long term this impact is predicted to be major and may result in the permanent loss of the Ma On Kong egretry and reduce the likelihood of the re-establishment of an egretry at Ho Pui or nearby.

4.8 General Recommendations on Ecological Impact Mitigation

4.8.1 The prime objective of the present project is land drainage improvement. The key to a practical yet environmentally sustainable management scheme is the integration of hydraulic and ecological techniques. The following management measures have largely been adopted internationally by engineers and ecologists as part of a set of standard best practices. These measures also adhere to the aforementioned impact mitigation policies set out in the Technical Memorandum on EIA Process. Recommendations on specific ecological mitigation measures to be undertaken at KT13 are presented in Section 4.9.

Alternative 1: Avoidance

4.8.2 The first and foremost alternative is the avoidance of channelization of streams with particular ecological value.

4.8.3 In the present case, avoidance of channelisation of the section of KT13 above the bridge leading to Ho Pui Village would leave intact the most ecologically valuable section of the stream and would avoid entirely the part of the stream that is adjacent to the Ho Pui egretry site. This option would, therefore, result in there being no adverse ecological impacts on the upper reaches of the stream and, if taken in parallel with the design measures to minimize impacts on the lower part of the stream (as the discussed under Alternative 3 below), will not cause any residual ecological impacts. This option would not however relieve the area from flooding.

Alternative 2: Dredging and Widening without Constructing Permanent Banks and Bottoms Structures; Combined with Scheduling of Construction Works – A Means of Impact Minimization

4.8.4 For streams where avoidance of construction work is not warranted and where increased hydraulic capacity is needed, such work should be limited to dredging and widening only, with a minimum of permanent structures on the bank sides and bottoms, and without excessive removal of the existing stream and riparian vegetation. Preferably, one of the banks should be left undisturbed. If protection of the banks from erosion is necessary, rock armouring or gabion structures may be placed along the margin of the watercourse (Plate 4.18). It must be emphasized that the concreting of banks and bottoms will definitely render the stream sterile ecologically. Not only are the existing ecosystems lost, reestablishment of flora and fauna in the future is also precluded. The use of grasscrete at channels, although less damaging than bare concrete does have value in mitigating visual and landscape impacts, but falls far short of meaningful conservation.

4.8.5 The execution of the construction works should be planned to minimize any potential disturbance to fauna of conservation importance. Generally, in Hong Kong the most sensitive period is the breeding season which for most species occurs mainly during the early to mid wet season from April to July, though for some species breeding activity begins earlier in the year or is completed later.

Alternative 3: Dredging and Widening, and the Placement of Gabions for Structural Stability; Combined with Scheduling of Construction Works – A Means of Impact Compensation/Impact Minimization

4.8.6 Under circumstances where protection of stream bank/bottom from erosion is crucial, gabions, instead of concrete or grasscrete, should preferably be used to provide the required bank stability. Successful overseas cases of such applications are common (please refer to Plates 4.19 to 4.21 for examples). Plate 4.22 shows typical examples of gabion channel in Hong Kong. The main ecological advantages of gabions are the following:

(i) Gabions allow the natural exchange of water between the stream and the substrate, thus connecting the watercourse with the surrounding phreatic strata (if concrete base platform is not used), and allow the growth of riverine flora through the gabions; the flora in turn provide a rich habitat for fauna, including microorganisms which together with the flora are essential food sources for other organisms in the riverine food chain.

(ii) The interstices between the rocks or stones within a gabion make an excellent habitat for biota, especially invertebrates and fish, as well as their eggs, larvae or fry, all of which at various times need this form of refuge. The open form of the gabion also provides refuge from fast currents, predators, changing ambient temperatures and other threats. Gabions thus closely mimic the functions of natural bank/bottom, thereby allowing the stream and riparian ecosystems to largely maintain their biocoenosis, albeit in a somewhat artificially altered state.

(iii) The open form of the gabion is far more desirable than that offered by concrete or grasscrete, the former supports no life and the latter provides little opportunity for biodiversity or recolonization by ecosystems. With gabions, habitats lost to the dredging and widening of the river/stream can largely be compensated by natural reestablishment.

4.8.7 From the engineering perspective, channels constructed with gabions in the last 100 years attest to their suitable hydraulic and structural performance. The design life span of PVC sheathed gabions in a river/stream setting is conservatively estimated to be at least 15 to 20 years. To ensure the continual satisfactory performance of gabions, however periodic inspection and maintenance are, required. When necessary, sections will need to be strengthened or replaced. Also, vegetation growing in the gabions at critical segments of the channel will need periodic trimming to maintain adequate hydraulic conditions. Such routine maintenance would, in any case, be required for alternative construction techniques.

4.8.8 From the aesthetic perspective, as vegetation will fully establish on the gabions within just a few years, these artificial elements will completely integrate with the surrounding natural riverine environs. This establishment of flora takes place naturally, and hydroseeding or planting at gabions is therefore usually not necessary, except where specific plants are required.

4.8.9 Figure 2.4 present a typical possible configuration of gabion for channel lining.

4.8.9 As in the previous alternative, the execution of the construction works should be planned to minimize any potential disturbance to fauna of conservation importance.

4.9 Proposed Ecological Impact Mitigation Measures at KT13

4.9.1 Since Alternative 1 described above fails to meet the design objectives of the project, it is proposed that mitigation of adverse impacts should be mitigated by a combination of measures described under Alternative 2 (for the most ecologically sensitive section of the stream) and Alternative 3 (for the remainder). These measures are detailed below.

Mitigation of operational phase impacts

4.9.2 Mitigation of operational phase impacts arising from stream channelisation is proposed by designing the required channelisation measures at KT13 as detailed in Figures 2.2A, 2.2B and 2.4. The proposed design contains the following elements:

(i) Mid-stream section around Ho Pui egretry site

The mid-stream section of KT13 around the location of the egretry site is not proposed to be channelised as it is considered that any form of artificial channel construction is likely to have major adverse permanent impacts on the egretry site which would prevent its successful re-occupation.

The normal stream flow will continue to follow the existing stream, but a bypass culvert will be formed to divert floodwater from the existing channel when water reaches flood level. The design of this culvert addresses the requirement to avoid modification of the middle section of the stream for the preservation of the Ho Pui Egretry and the associated habitats and will ensure that continuous flow of the existing stream is maintained. Should the egretry be re-occupied prior to the commencement of construction, further mitigation measures are required to compensate for the impacts of this culvert on the egretry.

In order to form the bypass channel it will be necessary to realign a short section of the existing natural stream (c. 25 m) by shortening a meander. The section to be realigned will be restored with gabion banks and natural substrates as stream bed materials. Gabions are necessary to provide stability to the stream banks given the scouring effect at the meander.

(ii) Upstream section (Section B) and Downstream Section (Section A)

The upstream section of KT13 is considered to have intrinsic moderate ecological value and provides habitat for wetland-dependant fauna. Use of a gabion-lined channel is proposed in order to maintain the ecological value of the channel whilst not compromising flood-prevention objectives. Trapezoidal channel with gabion sides and gabion bottom (as shown in Figures 2.2A, 2.2B and 2.4) is proposed.

Once the gabion channels have been constructed the water table in the adjoining areas will be restored and there will be no significant changes in the water table in wetlands. Hence no long-term impact on wetland habitats and their biota in expected.

Potentially adverse impacts arising from the maintenance of the channelized sections will be minimized by restricting routine channel maintenance to annual silt removal by hand or light machinery during the dry season (October to March). The management of woody / emergent vegetation will be limited to manual cutting, to be carried out only when unchecked growth of such vegetation is very likely to impede channel flow.

Bypass channel alignment

4.9.3 Although the construction of a bypass channel would eliminate adverse impacts to the most sensitive section of the existing stream, the construction of the bypass channel also has potential ecological impacts. The selected alignment was chosen after evaluation of alternative alignment options. Alternative alignment options are described in detail in Appendix B, Table B1 and illustrated in Figure B1 and are summarised below:

- Option 1: bypass channel largely to the west of the existing stream, largely passing through filled land (former ponds) but requiring some loss of land zoned Conservation Area, some woodland area and minor realignment of a meander in the section of stream to be bypassed.

- Option 2: bypass channel further to the west of the existing stream than Option 1, avoiding Conservation Area zone but bisecting woodland area and requiring resumption of existing houses as well as significant landscape impact due to need to traverse a low hill.

- Option 3: bypass channel direct route, approximately 50% to the west and 50% to the east of the existing stream (i.e. bisecting the egretry location and crossing the most sensitive stream section) with minor intrusion into Conservation Area.

- Option 4: bypass channel eastern route, passing through Ho Pui village and impacting a section of the existing stream close to the egretry location and wetland in Ho Pui.

- Option 5: “no-build” option.

4.9.4 The “no-build” option is preferable ecologically but will result in failure to satisfy the project objective of eliminating flooding. Of the remaining options, both Options 3 and 4 would have unacceptable impacts on the existing stream and the egretry location. The potential ecological impacts of Options 1 and 2 are similar in scale and importance. Both options would avoid permanent impacts on the egretry location; but both would result in some adverse impacts on habitats of ecological value. However, Option 1 is preferable in respect of its lower landscape impact and avoidance of village houses

4.9.5 Accordingly, Option 1 was selected, but it was then refined to reduce ecological impacts to an acceptable level. Avoidance of some potential adverse impacts was achieved by minor adjustments to the alignment to reduce the extent of intrusion into the Conservation Area and to reduce loss of trees. In order to minimize impacts on the section of stream meander to be realigned it is proposed to make use gabion banks and natural bottom instead of concrete walls, and measures to mitigate construction phase impacts and compensate for unavoidable loss of trees are proposed. These proposed avoidance, minimization, mitigation and compensation measures are described in more detail in below.

Mitigation of adverse impacts of bypass channel construction

4.9.6 The northern section of the proposed bypass channel will be immediately adjacent to the northern section of the egretry site. Even with the controls which are proposed to restrict the adverse effects of construction works, it is considered that the proximity of the bypass channel will open up this section of the egretry site to accidental or deliberate disturbance. Since the egretry was abandoned in 2005 and it is likely that it will not be re-occupied, it is most likely that there will be no impacts on breeding egrets. However, in the event that the egretry is re-occupied prior to the commencement of construction and such re-occupation includes the northern part of the area occupied by breeding egrets prior to 2005, there is a significant likelihood that disturbance will result in the abandonment of all or part of the northern section of the egretry. In addition to passing close to the egretry, this northern section of the bypass channel will pass directly through an area which is zoned as Conservation Area (see Figure 4.13), though this area is no longer suitable for egrets occupation (due to recent clearance of trees and bamboos).

4.9.7 Accordingly, temporary chain link fence should be provided along the site boundary near the CA zone and Ho Pui Egretry (Figure 4.13) to restrict access by construction workers or equipment or works to these sensitive areas. Disturbance of the vegetation outside the site boundary is prohibited. Signage should be provided at conspicuous location to warn workers from entering and disturbing the sensitive areas.

4.9.8 Approximately 59 trees will be retained under this Project, while approximately 60 trees will be transplanted and approximately 74 trees will inevitably need to be felled. No rare/protected or species of conservation importance will be affected. Compensatory planting of about 148 heavy standard size trees (in 2:1 ratio) will be provided. Details can be found below and in the Landscape and Visual Impact Assessment (Chapter 10).

4.9.9 It is proposed to mitigate for the loss of trees by planting an area (855 m²) of appropriate tree and bamboo species to provide compensatory habitat in the long term (Figure 4.13). The proposed planting area is at the southern end of the proposed channel. This location has been chosen as it is a relatively large block of land within the project area (compensatory planting can only be provided within the boundaries of the project area) in proximity to a mixture of wetland and fishpond areas. As such, it is considered to be an area with potential for egretry establishment in the long term. The tree and bamboo species to be used in the compensatory planting area are detailed in Table 4.35. The tree and bamboo species are the same or similar to native trees which have to be felled and include species suitable for nesting egrets. Planting should be completed before the commencement of Operation Phase.

4.9.10 In addition, it is proposed to plant Bambusa eutuldoides in a staggered double row as a screen along the east side of the bypass culvert where this is close to the egretry (Figure 4.13). When mature, the bamboos will form a screen, shielding the egretry from disturbance, and will also provide potential breeding sites. To ensure the right species of bamboo is planted, an experienced botanist should be acquired by the contractor to source the correct bamboo species. Moreover, to ensure the planting of the bamboo for screening to be effective, the bamboos should have a minimum stem diameter of 8-10 cm and clump size of 5 shoots per plant.

Table 4.35

Tree and bamboo species to be used in compensatory planting area

Species	Percentage of total
Bambusa eutuldoides	40
Cinnamomum camphora	15
Celtis tetranda	15
Ficus virens	15
Ficus microcarpa	15

Mitigation of construction phase impacts

4.9.11 As discussed above, the principal concern during the construction period is minimizing adverse impacts to nesting egrets should the Ho Pui egretry be re-occupied. Accordingly, monitoring of the egretry site should be conducted as part of the EM&A during the breeding season in order to establish if breeding egrets are present. Monitoring should be carried out during the months of March to August. Notwithstanding the monitoring result, no construction works should take place within 100m of the egretry location (Figure 4.13) during the period from 1st March to end of May. This covers the early nesting period because of early warming up weather in recent years which may lead to some egrets preparing for nesting as early as March. During this period, construction activities within the works area within 100m of the egretry should be limited to transit movements by construction vehicles. If no egret nest is found at the Ho Pui egretry, construction works can be carried out from June to February in the next year subject to approval obtained from AFCD and EPD. Otherwise, the ‘no works’ period should be extended till the end of September.

4.9.12 Prior approval from AFCD and EPD should be obtained before commencement of construction works on the bypass culvert. Adequate time should be allowed for the vetting process by the relevant authority.

4.9.13 Construction phase impacts on the stream as a whole will be minimized by phasing the major construction activities to a maximum of 75 m length at Section A and Section B of the Channel (except for 100 m sections at the extreme upstream and downstream sections of the project works area where no restriction will apply as these areas are remote from the Conservation Area and egretry site). As the two work fronts will be constructed in one direction and since there are more than 300 m separation between Section A and Section B of the Channel, the above 75 m restricted length could be implemented concurrently at both Section A and Section B of the Channel. This will reduce the duration of construction phase impacts at any one location and will minimize the duration of time during which the water table will be drawn-down at any one location, thereby minimizing the adverse effects to adjacent wetland habitats and their plant and animal communities. The water table is expected to be lowered by approximately 600 mm and the lateral distance affected is expected to be of the same magnitude. Draw-down will be temporary and the water table will be restored once construction is completed.

4.9.14 The contractor should ensure continuous water flow to the remaining part of the stream during the construction stage and the ‘no works’ period. A typical temporary drainage diversion plan is shown in Figure 7.3 to demonstrate that keeping continuous water flow is feasible and workable.

4.9.15 The contractor should provide details of the mitigation measures to be implemented during construction stage as part of their working method statement to the Engineer for approval. This should also be reviewed by the Environmental Team Leader with reference to the baseline ecological monitoring results if necessary.

4.10 Conclusions

4.10.1 Predicted significant ecological impacts and proposed mitigation measures to eliminate or reduce these to acceptable measures are summarised below:

Construction Phase Impacts

Disturbance to existing stream habitats and fauna due to channelisation.

4.10.2 In the absence of mitigation measures, disturbance to stream habitats and fauna due to channelisation of sections of the stream is considered to result in a moderate adverse ecological impact. The nature of the project is such that the direct physical impacts on stream habitats and fauna cannot be eliminated or reduced; however, it is proposed to restrict the works area to a maximum of 75 m length at Section A and Section B of the Channel except for the 100 m sections at the extreme upstream and downstream limits of the project area. As the two work fronts will be constructed in one direction and since there are more than 300 m separation between Section A and Section B, the works could be undertaken concurrently at Section A and Section B of the Channel. This will reduce the duration of construction phase impacts at any one location and will reduce the extent of habitat loss for wildlife which may use the streamcourse and surrounding riparian corridor as a feeding or roosting area.

4.10.3 It is considered that with such mitigation, the residual construction phase impact would be reduced to an acceptable level.

Disturbance to egretry due to channelisation

4.10.4 Should the Ho Pui egretry be re-occupied, in the absence of mitigation measures, there is a significant risk that construction activity could cause it to be abandoned again, which would be a major adverse impact. However, by restricting construction works within 100m of the egretry to the non-breeding season direct disturbance impacts to the breeding egrets will be eliminated.

4.10.5 Because the proposed bypass culvert is close to the northern section of the egretry site, the project would make the egretry more vulnerable to casual disturbance in the event of its being re-established in this location. It is considered most likely that the egretry will not be re-occupied. However, in the event of re-occupation prior to the commencement of construction it is considered most probable that any egrets which are displaced from the disturbed part of the egretry will move to another part of the egretry or to another egretry in the area and it is unlikely that they will be prevented from breeding entirely. Accordingly, it is considered that the residual impact of any disturbance would, most likely, be insignificant or small and of only local importance and the risk of the residual impact being of moderate importance (birds being displaced and prevented from breeding) is very small.

Disturbance to habitats and fauna (other than the egretry) due to construction of the bypass culvert.

4.10.6 In the absence of mitigation measures, the construction of the bypass culvert would have only a low adverse impact of habitats and species other than egrets. Nonetheless, it is proposed to provide chain link fence along the site boundary near the CA zone and Ho Pui Egretry to restrict access by the contractor into these sensitive areas. Whilst there may be some impact on breeding habitat of White-throated Kingfisher, a sufficient unchannelised length of the stream will remain to provide breeding habitat for this species. Accordingly, it is considered that specific mitigation measures will not be required to mitigate for minor impacts on this species.

Changes to water table in adjacent wetlands

4.10.7 Construction works are predicted to result in the reduction of the water table in adjacent wetlands by up to 600 mm (both vertically and laterally). The overall area of wetlands affected would, therefore, be very small and the predicted adverse ecological impact is considered to be low. Making provision to limit the working area to a maximum of 75 m length at both Sections of the Channel (except for the 100 m extreme upstream and downstream limits) would further reduce the duration of this adverse impact at any one location. However, this does not alter the overall evaluation of the residual impact as it is already considered to be low in the absence of mitigation.

Operational Phase Impacts

Loss of stream habitat and fauna

4.10.8 In the absence of mitigation measures, loss of stream habitats and fauna due to channelisation of sections of the stream is considered to result in a moderate adverse ecological impact. The proposed use of gabions to form the banks and bottom of the channelised sections of the stream will permit recolonisation by riparian and aquatic fauna and flora. Accordingly, it is considered that, with mitigation measures in place, operational phase impacts on stream habitat and fauna will be very low.

Loss of low-lying grassland/fallow land of moderate ecological value

4.10.9 The proposed secondary drainage channel KT13 will result in the permanent loss of an area of low-lying grassland/fallow land. Though this habitat is considered to be of moderate ecological importance, the area affected is small and hence the scale of the impact is considered to be low. As such, no mitigation measures are considered necessary.

Loss of fishponds of moderate ecological value

4.10.10 The proposed secondary drainage channel KT13 will result in the permanent loss of an area of fishponds. Though this habitat is considered to be of moderate ecological importance, the area affected is small and the loss of habitat and feeding areas for birds and other wildlife is also small. The scale of the impact is, therefore considered to be low and, in the absence of an alternative alignment for KT13, is considered to be acceptable without additional mitigation measures.

Loss of woodland of moderate ecological value

4.10.11 Loss of woodland of moderate ecological value has been reduced to a minimum (only 0.008 ha) by fine-tuning the bypass channel alignment to minimize the requirement for felling of existing trees. Though this habitat is considered to be of moderate ecological importance, the area affected is very small and the ecological value of the habitat and individual trees to be lost is correspondingly small. As such the scale of the impact is considered to be low. No ecological measures to compensate for the loss of individual trees are considered necessary. However, it is proposed to compensate for the habitat loss by carrying out tree planting at the location shown in Figure 4.13.

Disturbance to egretry

4.10.12 In the absence of mitigation measures, channelisation of the section of the stream which passes through the egretry site is predicted to result in a high likelihood of abandonment of the egretry, should it be re-established prior to the commencement of construction. This would be a major adverse ecological impact. Avoidance of channelisation of the section of stream (except for a small localized realignment of the meander) with bypass culvert design which passes through the egretry site together with the restriction of the timing of construction works is predicted to eliminate this potential adverse impact. The bypass culvert design is such that continuous water flow similar to existing condition will be maintained at the unmodified portion of the stream.

4.10.13 However, in the long term, any adverse impacts to the egretry site will be eliminated once the recommended screen planting of bamboo matures. This will screen the egretry site from human disturbance and will also provide potential nesting sites in the event that the egretry is re-established. Long term impacts on the egret population of the area are considered to be insignificant as there will be no net reduction in the available nesting, roosting and foraging habitat, either in the immediate vicinity of the egretry or in the wider area.

4.10.14 Accordingly, there will be no significant permanent residual ecological impacts arising from this project. It is considered, therefore, that in the context of the wider benefits of the project in reducing flooding, the low to moderate impacts arising during the construction period are acceptable.

References

Anon. 1998.Urbanization and streams: studies of hydrologic impacts. US Environmental Protection Agency, Office of Water.

http://www.epa.gov/OWOW/NPS/urbanize/report.html.

Barclay, J.S. 1980. Impact of stream alterations on riparian communities in south-central Oklahoma. U.S. Department of the Interior, Fish and Wildlife Service, Biological Services Program: FWS/OBS-80/17.

Bascombe, M. J., Johnston, G., and Bascombe, F. S. 1999. The butterflies of Hong Kong, Academic Press.

BBV. 2002. Yuen Long, Kam Tin, Ngau Tam Mei & Tin Shui Wai Drainage Improvement, Stage 1. Environmental Study for Phase 2 Works. Binnie Black and Veatch Hong Kong Ltd., Hong Kong.

Carey, G. J. 1998. Egretry Counts in Hong Kong, with particular reference to the Mai Po and Inner Deep Bay Ramsar Site. The Conservancy Association, Hong Kong.

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

Cheung, K.W. 1998. Sightings of three freshwater fish. Porcupine! 18, 9.

Corbet, P. S. 1999. Dragonflies: Behavior and Ecology of Odonata, Cornell University Press.

Dudgeon, D. (1995). Environmental impacts of increased sediment loads caused by channelization: a case study of biomonitoring in a small Hong Kong river. Asian Journal of Environmental Management 3, 69-77.

Dudgeon, D. 1999a. Tropical Asian Streams: Zoobenthos, Ecology and Conservation, Hong Kong University Press.

Dudgeon, D. 1999b. More about Hong Kong freshwater fishes. Porcupine! 19, 17-18.

Dudgeon, D., and Corlett, R. 1994. Hills and streams: An Ecology of Hong Kong, Hong Kong University Press.

Federal Interagency Stream Restoration Working Group. 1998. Stream corridor restoration: principles, processes and practices. US Department of Agriculture, Washington.

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

Hill, D. S., and Phillipps, K. 1981. Hong Kong animals, Government Printer, Hong Kong.

Hafner, H.2000. Heron Nest Site Conservation in Kushlan. J.A. and Hafner H. (eds.) Heron Conservation Academic Press, San Diego.

Hilton, C. 1995. “Feedback.” Porcupine 13, 7.

Hirano, T. 1997. The effect of river bank repair works on wintering waterbird populations. Strix 15: 39-44.

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

Kushlan, J. A. and Hafner, H. 2000. Heron Conservation. Academic Press.

Kwok, H.K. and Corlett, R.T. 1999. Seasonality of a forest bird community in Hong Kong, South China. Ibis 141: 71-79.

Kwok, H.K. and Dahmer, T.D. 2000. Impact of drainage channel on wetland avifauna diversity and implication of grasscrete embankment in Hong Kong. Paper presented at the Workshop on Integration of Impact Assessment and Spatial Planning, International Association of Impact Assessment 2000 Conference, 19-23 June 2000, Hong Kong.

Kwok, H.K., Wong, L.C. and Carey, G.J. 2001. Egretry birds in Hong Kong: with particular reference to the Mai Po Inner Deep Bay Ramsar Site. Summer 2000 Report. Hong Kong Bird Watching Society, Hong Kong.

Krapu, G.L. 1996. Effects of legal drain clean-out on wetlands and waterbirds: a recent case history. Wetlands 16: 150-162

Lau, M. W. N., and Dudgeon, D. 1999. Composition and distribution of Hong Kong Amphibian fauna. Memoirs of The Hong Kong Natural History Society, 22, 1-80.

Leven, M.R. 1998. Birds [of farmland]. Porcupine! 18: 19-23.

Leven, M.R. 2001. Shrubland birds in Hong Kong: community structure, seasonality and diet. PhD Thesis, University of Hong Kong.

Loneragan, N.R. and Bunn, S.E. 1999. River flows and estuarine ecosystems: implications for coastal fisheries from a review and a case study of the Logan River, southeast Queensland. Austral Ecology 24: 431-440.

Moore, N.W. (1997). Dragonflies – Status Survey and Conservation Action Plan. IUCN/SSC Odonata Specialist Group. IUCN, Gland, Switzerland.

Roper-Lindsay, J 1994. Tales of the riverbank – examples of bank restoration on urban rivers. Pp. 125-143 in Collier, K.J. (ed.) Restoration of aquatic habitats. Selected papers from the second day of the New Zealand Limnological Society 1993 Annual Conference. Department of Conservation, Wellington.

Seigried, W. R. 1972. Breeding success and reproductive output of Cattle Egret. Ostrich, 45: 43-55.

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

Vivash, R. and Murphy, D. 1999. Manual of river restoration techniques. River Restoration Centre, Silsoe.

Voisin, C.1991. The Herons of Europe. Poyser, United Kingdom.

Walthew, G. 1997. The status and flight periods of Hong Kong Butterflies. Porcupine! 16, 34-37.

Weller, M.W. 1995. Use of two waterbird guilds as evaluation tools for the Kissimmee River restoration. Restoration Ecology 3: 211-224.

Weller, M.W. 1999. Wetland birds: habitat resources and conservation implications. Cambridge University Press, Cambridge

White, K. 1996. Restoration of channelized streams to enhance fish habitat: a case study of the Middle Fork John Day River, Oregon.

http://www.ies.wisc.edu/research/ies900/kimchannelization.htm

Willard, D.E., Finn, V.M., Levine, D.A. and Klarquist, J.E. 1990. Creation and restoration of riparian wetlands in the agricultural Midwest. pp. 327-337 in Kusler, J.A. and Kentula, M.E. (eds.). Wetland creation and restoration: the status of the science. Island Press, Washington.

Wilson, K. D. P. 1995. Hong Kong Dragonflies, Urban Council of Hong Kong.

Wilson, K.D.P. 1997. An annotated checklist of the Hong Kong dragonflies with recommendations for their conservation. Memoirs of the Hong Kong Natural History Society 21, 1-68.

Wong, F.K.O. 1991. Habitat Utilization by Little Egrets Breeding at Mai Po, Hong Kong Bird Report 1990:185-190.

Wong, L. C. Corlett, R. T., Young, L., and Lee, J. S. Y. 1999. Foraging flights of Nesting Egrets and herons at a Hong Kong Egretry, South China. Waterbirds 22(3): 424-434.

Wong, L. C., Kwok, H. K. and Carey, G. J. undated. Egretry Counts in Hong Kong, with particular reference to the Mai Po and Inner Deep Bay Ramsar Site. The Conservancy Association, Hong Kong.

Wong, L. C. and Woo, C. K. 2004. Summer 2003 Report on Egretry Counts in Hong Kong, with particular reference to the Mai Po Inner Deep Bay Ramsar Site. Hong Kong Bird Watching Society, Hong Kong.

Young, L. (1993) The Ecology of Hong Kong Ardeidae (Aves) with Special Reference to the Chinese Pond Heron at the Mai Po Marshes Nature Reserve. Phd. Thesis, University of Hong Kong.

Young, L., and Cha, M. W. 1995. The history and status of egretries in Hong Kong with notes on those in the Pearl River Delta, Guangdong, China. Hong Kong Bird Report 1994: 196-215.