7. ECOLOGICAL IMPACT ASSESSMENT

7.1 Introduction

7.1.1 This Chapter covers ecological issues arising as a consequence of the proposed drainage improvement works. The objectives of this ecological assessment are as follows:

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

· to assess the ecological impacts of the proposed drainage improvement works;

· to detail effective ecological mitigation measures to avoid, minimize and compensate significant impacts; and

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

Environmental Legislation, Guidelines, Standards and Criteria

7.1.2 The following legislation, guidelines and standards were referred to during the course of the ecological impact assessment:

· Country Park Ordinance (Cap.208);

· Forests and Countryside Ordinance (Cap.96);

· Wild Animals Protection Ordinance (Cap.170);

· HKPSG (Chapter 10 – Conservation);

· EIAO Guidance Note 6/2002;

· EIAO Guidance Note No. 7/2002; and

· EIAO Guidance Note No. 10/2004.

7.1.3 In particular, the ecological impact assessment is conducted based on the criteria in Annex 8 of the EIAO-TM and the guidelines in Annex 16 of the EIAO-TM.

7.2 Study Areas

7.2.1 The Drainage Improvement in Northern New Territories – Package C, is one of drainage improvement works recommended by the Drainage Master Plan Study of the Northern New Territories (DMP Study) which covered the Indus and Ganges drainage basins. The objective of the Project is to alleviate flooding problems in the Man Uk Pin and Lin Ma Hang areas by modifying the existing secondary and local drainage systems.

7.2.2 This EIA Report covers five channels in Man Uk Pin and Lin Ma Hang regions for which an ecological impact assessment is required under the terms of the EIA Ordinance, as follows:

(a) Man Uk Pin Region sub-packages (MUP03, MUP04A, MUP04B and MUP05);

(b) Lin Ma Hang Channel sub-package (LMH01).

Figure 7.1 indicates the locations of the five streams (sub-packages) which are the subject of this EIA together with tributary streams to MUP05 (MUP01 and MUP02) which are Non-DP. Potential secondary and cumulative impacts of the proposed works to MUP01 and MUP02 which are relevant to this EIA are considered where appropriate.

7.2.3 Study Areas for the ecological assessment in this EIA are the areas within 500 m of the proposed project boundary, together with areas downstream from the project boundary for as far as potential ecological impacts (temporary or permanent changes in water quality or flow regime) may have an impact on riparian habitats, communities and flora or fauna. Study Areas are indicated in Figures 7.2 – 7.3. Because of the close proximity of the streams, the Study Area envelopes for MUP03, 04 & 05 show considerable overlap. Accordingly, for certain aspects of the ecological survey and assessment, for example habitat mapping and evaluation, Study Areas for these streams have been combined.

7.3 Survey Methodologies

Habitats

7.3.1 A habitat survey was conducted in an area of 500 m radius around the five existing stream courses and the proposed project boundary (Figures 7.2 – 7.3). Fourteen habitat types were identified within the Study Areas:

• bare ground/works in progress;

• active dry agricultural land;

• active wet agricultural land;

• inactive dry agricultural land;

• inactive wet agricultural land;

• fishpond;

• hillside grassland;

• shrubland;

• plantation;

• orchard/horticultural land;

• woodland;

• river/stream;

• wasteland;

• urban and industrial area.

7.3.2 Habitats were determined from aerial photographs taken during 2004, supported by ground-truthing during January and February 2005 where necessary. Because ground-truthing was conducted during the dry season, seasonally wet areas were identified on the basis of their supporting characteristic wetland plant species.

7.3.3 In order to confirm that there are no significant changes in the intervening period, a follow-up survey was undertaken in early January 2007. The follow-up survey had focused on the habitats described above, with special emphasis on habitats of moderate ecological value at and near the project area such as the stream and its riparian habitats. It is noted that there were no changes to the land use and no improvement to the environment of the site areas. The survey revealed that there were no significant habitat changes in the intervening period. It is therefore confirmed that the habitat survey conducted in 2003 / 2004 is still valid and the results are used for the assessment.

Botanical / Vegetation Survey

7.3.4 A detailed botanical/vegetation survey was also undertaken during January and February 2005 within the Project Area boundaries of the proposed new channels and in adjacent habitats of ecological importance, focusing especially on those habitats (such as wetlands) which might be indirectly affected by impacts of the development. A less detailed botanical survey was undertaken in other habitats of ecological interest in the Study Areas (such as natural woodlands) in order to characterize the habitat types and provide a qualitative estimate of their ecological value but not necessarily to identify all plant species present.

Mammals

7.3.5 Surveys of non-flying terrestrial mammals were conducted in tandem with surveys of avifauna, herpetofauna and insects (see below). Direct observations of mammals were supplemented by searching for mammal signs such as droppings, burrows, diggings and footprints.

7.3.6 As the Lin Ma Hang Lead Mine SSSI, which holds an important colony of bats, is 600 m from LMH01, a literature review of bat data for the area was undertaken. In view of the extensive data available and the fact that no direct impacts on the colony or on bats are predicted, field surveys were not carried out.

Avifauna

7.3.7 For each of the sub-packages monthly surveys were carried out from October 2003 to September 2004, covering a 12 month period. A transect was walked through each of the survey areas to encompass river and stream channels (where it was possible to approach these) and surrounding habitats, to a distance of 500 m from the study channels. The transect route is detailed in Appendix E. As streams such as these are of relatively limited importance in terms of wetland bird species, the aim was to define the general bird community in the area, with particular reference to recording bird usage of the watercourses that might be impacted by new or redesigned drainage channels.

7.3.8 Both daytime and night-time surveys were carried out. Daytime surveys commenced between 30 and 60 minutes after sunrise and continued until approximately 10:30 am in the cooler months of the dry season, and 9:30 am in the hotter months of the wet season. All bird species seen or heard were recorded. For widespread and commonly-occurring species in suitable habitat, only presence or absence was noted. For waterbird species or species of conservation concern (as defined by Fellowes et al. 2002), the number and behaviour of each were noted. Behaviour was categorised as F (Foraging), R (at Rest) or A (in Air). All bird species recorded utilising stream or river channels were counted and their behaviour recorded, using the same categories. Data collected during surveys were recorded in an Excel spreadsheet.

7.3.9 Night-time surveys were carried out primarily in order to survey for Eurasian Woodcock Scolopax rusticola, a nocturnal foraging species. However, they also allowed further data to be collected that would inform an overall assessment of the general avian community in each area. Night-time surveys were carried out using the same methodology as during the day, twice during October to April and once during May to September. In general, however, the night-time surveys produced little data due to the high levels of disturbance from humans and noise from barking dogs.

Herpetofauna

7.3.10 Herpetofauna (amphibians and reptiles) were surveyed every month from October 2003 to September 2004. Two day-time surveys of each sub-package, to a distance of 500 m from the study channels, were originally scheduled for October and November, with one day-time survey scheduled for each remaining month of the survey period. Due to time constraints, however, it was not possible to conduct two surveys in October. Accordingly, the second scheduled October survey was delayed until February.

7.3.11 During day-time surveys, a transect route was followed around each sub-package (see Appendix E), following the stream course where this was possible (but also incorporating all major habitat types present), and all species of herpetofauna encountered were identified and counted. In the course of the transect survey, active searching was conducted in or under appropriate microhabitats such as tree trunks, walls, stream margins, fallen branches, piles of wood, stones or other material, and other objects lying on the ground, in order to find cryptic or secretive species.

7.3.12 In addition to day-time surveys, each sub-package was surveyed at night on three occasions, two of which were scheduled for the period October to April and one for the period May to September. In practice, the first two surveys were conducted in April – a month in which amphibian nocturnal breeding activity is reaching a maximum, thereby ensuring that chances of detecting species were high.

7.3.13 During night surveys, the same transect was walked as for day surveys, with the emphasis on identifying and enumerating amphibians by identification of the male breeding vocalizations. This was supplemented by visual observations of amphibians and reptiles made with the aid of a strong flashlight.

Fish

7.3.14 Freshwater fishes were surveyed in MUP04A and MUP05. One-off surveys of these streams were conducted in January 2005. During the surveys, the stream banks were walked and fishes were identified with the aid of binoculars and kick net sampling using a D-net.

7.3.15 Fish surveys were not conducted in MUP03 or MUP04B as these are not true streams but existing or proposed artificial drainage ditches without permanent flow at present (see Figure 7.8) and not, therefore, capable of supporting fish communities.

7.3.16 The fish community in LMH01 had previously been the subject of extensive survey (Chan 2001, KFBG 2004). Accordingly, in the absence of subsequent material changes to the stream, assessment of the potential ecological impact of the project on fish in this stream was based on a review of this published data.

Butterflies

7.3.17 Butterflies were surveyed every month from October 2003 to September 2004, with the same frequency as for day-time surveys of herpetofauna. During the surveys, a transect route was followed around each sub-package (see Appendix E), following the stream course where this was possible (but also incorporating all major habitat types present, to a distance of 500 m from the study channels), and all species of butterfly encountered were identified and counted, with the aid of binoculars. Some species were caught with a long-handled net for inspection in the hand prior to release at the capture site. Care was taken to ensure that no individuals of these highly mobile insects were counted more than once.

Dragonflies

7.3.18 Dragonflies were surveyed every month from October 2003 to September 2004, with the same frequency as for day-time surveys of herpetofauna. During the surveys, a transect route was followed around each sub-package (see Appendix E), following the stream course where this was possible (but also incorporating all major habitat types present, to a distance of 500 m from the study channels), and all species of dragonfly encountered were identified and counted, with the aid of binoculars. Some species were caught with a long-handled net for inspection in the hand prior to release at the capture site. Care was taken to ensure that no individuals of these highly mobile insects were counted more than once.

Aquatic Invertebrates

7.3.19 Aquatic invertebrates were sampled in December 2004. Four replicate kick samples were collected at one or two sampling locations at sub-packages MUP04A, MUP05 and LMH01, depending upon the length of the water course, and the quantity of water present in the channel. Individual samples were preserved on site in 70% ethanol, and subsequently processed at the AEC laboratory. Species of aquatic invertebrate were identified to the lowest practicable taxonomic level (usually family) and counted.

7.3.20 Aquatic invertebrate surveys were not conducted in MUP03 or MUP04B as these are not true streams but existing or proposed artificial drainage ditches without permanent flow at present (see Figure 7.8) and not, therefore, capable of supporting anything other than ephemeral aquatic invertebrate communities.

7.4 Habitats within the Study Area

7.4.1 The areas of each habitat type within the Study Areas and Project Areas are listed in Table 7.1 and Table 7.2 respectively. The Project Areas are those areas within the proposed site boundaries as detailed in Figures 7.2 – 7.3. Accordingly, the Project Areas are the areas where direct ecological impacts are anticipated, either as a consequence of construction works or due to direct hydrological changes, for example where a section of channel is by-passed. Except where otherwise noted, indirect ecological impacts are anticipated to be contained within the 500 m Study Areas’ envelope (though the extent and scale of indirect impacts within the envelope will depend upon habitat type and species’ sensitivity).

7.4.2 Note that the Study Areas shown are based on the 500 m envelope of the project works as originally proposed, and hence represent the actual field survey area. However, the Project Areas shown have been adjusted to reflect design changes made during the course of the study and therefore represent the actual areas which may be directly impacted if the project is approved in its present form.

Table 7.1

Habitat Types and Areas within the Study Areas

Habitat type	LMH01		MUP03,04A&B,05
	Area (ha)	%	Area (ha)	%
Active Dry Agricultural Land	0.42	0.4	11.70	5.6
Active Wet Agricultural Land	-	-	2.63	1.3
Inactive Dry Agricultural Land	12.74	10.8	25.36	12.2
Inactive Wet Agricultural Land	3.95	3.4	4.55	2.2
Bare Ground	-	-	0.57	0.3
Developed Land	4.59	3.9	25.98	12.5
Fishpond	0.19	0.2	0.58	0.3
Hillside Grassland	37.75	32.1	10.91	5.2
Orchard/Horticulture	0.08	0.1	2.26	1.1
Plantation	0.65	0.6	2.43	1.2
Shrubland	9.20	7.8	30.30	14.5
Stream/River	2.05	1.7	1.73	0.8
Wasteland	0.43	0.4	3.98	1.9
Woodland	17.73	15.1	85.41	41.0
Area beyond HKSAR boundary	27.94	23.7	-	-
Total	117.73	100	208.38	100

Table 7.2

Habitat Types and Areas within the Project Areas

Habitat type	LMH01		MUP03		MUP04A&B		MUP05
	Area (ha)	%	Area (ha)	%	Area (ha)	%	Area (ha)	%
Active Dry Agricultural Land	-	-	-	-	<0.01	2.2	0.58	17.7
Active Wet Agricultural Land	-	-	-	-	-	-	-	-
Inactive Dry Agricultural Land	0.02	9.1	0.06	26.6	0.03	6.8	0.68	20.6
Inactive Wet Agricultural Land	0.02	9.1	-	-	<0.01	0.2	0.13	4.1
Bare Ground	-	-	-	-	0.04	10.6	-	-
Developed Land	0.05	22.7	0.16	73.4	0.09	22.5	0.72	22.0
Orchard/Horticulture	-	-	-	-	0.11	28.2	-	-
Plantation	-	-	-	-	-	-	0.25	7.6
Shrubland	-	-
Stream/River	0.10	45.4	-	-	-	-	0.37	11.2
Wasteland	0.03	13.6	-	-	0.12	29.4	0.55	16.8
Woodland	-	-	-	-	-	-	-	-
Total	0.22	99.9	0.22	100.0	0.39	99.9	3.28	100

7.4.3 Habitats and ecological characteristics of the Study and Project Areas and findings of ecological surveys are described below. Full plant and fauna species lists are given by stream in Appendix E.

7.4.4 In the discussion below, for each site a table lists the wetland-dependent species: WD, largely wetland-dependent species: (WD) or species of conservation concern: CC as defined by Fellowes et al. (2002) that were recorded during surveys. In addition, for faunal groups which are not permanently aquatic, the mean number recorded, the range and an indication of whether the species was recorded utilising the relevant stream or river channel are also provided. In the case of herpetofauna and insects, ‘utilising’ is taken as the animal being within 5 m of the watercourse when recorded.

7.5 Ecological Survey for LMH01

Study Area

7.5.1 The LMH01 Study Area is largely undeveloped and comprises the valley of the Lin Ma Hang Stream together with surrounding hills. The Lin Ma Hang Valley was formerly farmed but almost all the agricultural land has been abandoned. The dominant habitat in the upland area is grassland but there are also extensive areas of shrubland and natural forest. The human population of this remote area is very small and direct human influence on the habitats of the Study Area is now very small.

7.5.2 There are three sites of known conservation importance within the area: the Lin Ma Hang Lead Mines SSSI, the fung shui wood behind Lin Ma Hang village and Lin Ma Hang stream itself which is listed as an Ecologically Important Stream (EIS) in the ETWB Technical Circular No. 5/2005 and is a planned SSSI.

7.5.3 The Lin Ma Hang Lead Mines SSSI is an important roost site for bats (Table 7.3). Out of 42 water tunnel and mine sites surveyed by AFCD in 2004-05, Lin Ma Hang was the most important roost site for both Greater Bent-winged Bat and Lesser Bent-winged Bat and was also considered to be of importance for the uncommon Chinese Myotis. At approximately 600 m from the proposed works, the roost lies outside the Study Area and will not be directly impacted by the proposed works. Bats from the roost are likely to forage throughout much of the Study Area, but none of these species are considered to show particular associations with streams for foraging (Shek 2006) so are unlikely to occur in concentrations in or near the project area. The limited scale of the proposed works, impacting a short section of bank and associated riparian vegetation at 600 m from the roost, would not represent a significant loss of habitat suitable for foraging bats. Thus, the impacts on the roost at Lin Ma Hang Lead Mine will be negligible.

Table 7.3

Bat numbers at Lin Ma Hang Lead Mine Roost

(Source: Shek and Chan (2005)[1])

Species	Status *	Number in summer 2004	Number in winter 2004-05
Rickett’s Big-footed Bat Myotis ricketti	Common	few**
Greater Bent-winged Bat Miniopterus magnater	Common	936	658
Lesser Bent-winged Bat Miniopterus pusillus	Uncommon	200	216
Intermediate Horseshoe Bat Rhinolophus affinus	Uncommon	few
Least Horseshoe Bat Rhinolophus pusillus	Uncommon	few
Chinese Myotis Myotis chinensis	Uncommon	9	6
Total		1146	898

* As defined by Shek and Chan (2005); ** Not stated but apparently < those species for which numbers are listed.

7.5.4 The fung shui wood at Lin Ma Hang is situated approximately 100 – 300 m from the Project Area. The habitat characteristics of the wood, the fact that it is at a higher elevation than the proposed works area, its distance from the stream and scale scale of the proposed works combine to ensure that the project will have no impacts on the woodland.

7.5.5 The Lin Ma Hang stream is listed as an Ecological Important Stream (EIS) in the ETWB Technical Circular No. 5/2005 and was proposed as a SSSI by Dr. Bosco Chan of the Department of Ecology and Biodiversity, University of Hong Kong, in 1998. The SSSI proposal was made on the basis of the extremely diverse lowland fish community present in the stream, including the largest natural population of the rare Chinese Rasbora Rasbora steineri of any stream in Hong Kong, as well as the only confirmed Hong Kong location of Spiny Eel Mastacembelus armatus. The proposed works to the stream will clearly have a direct impact on the stream itself, although appropriate planning and timing of the construction phase will ensure that the impacts are minimal. Full details of measures which should be taken to mitigate against possible impacts are detailed later in this document (Section 7.9).

Project Area

7.5.6 The project area for LMH01 is limited to short sections of the stream adjacent to the village and the Boundary Fence Road.

Habitats in the Study Area

Water-course and Riparian Vegetation

7.5.7 The stream running through the Study Area is about 1m in depth and 1m to 2m in width, and it can be broadly differentiated into upper, middle and lower stream, each with different habitat characteristics. The upper stream has an earthed bank bordered with agricultural habitat on both sides, with no riparian vegetation except those herbaceous plants established along the water edge. The middle stream is shaded by the trees from the woodland on the west of the stream, with most of its southern bank strengthened by stone-wall. The lower stream is partially shaded by the woodland and has already been channelised and lined with concrete (Figures 7.4 & 7.5). Water flow in dry season was found to be slow and shallow along the stream (Figure 7.5).

7.5.8 Tree species found in the riparian zone in the middle stream are Cinnamomum camphora, Cleistocalyx operculata, Ilex rotunda and Syzygium jambos; whereas species commonly found along the edge of the stream and along the bank are the herbs Polygonum barbatum, Ranunculus scleratus, Rorippa nasturtium-aquaticum, Lindernia crustacean, Gamochaeta purpurea and Ageratum conyzoides, the creepers Ludwigia adscendens, Commelina diffusa and Alternanthera sessilis, the grasses Paspalum districhum and Echinochloa crusgalli, well as the fern Cyclosorus interruptus. Several patches of the herbaceous plant Polygonum japonicum which is categorized as “rare” by Xing et al. (2000) were found along the waters’ edge on the upper part of the stream, but this species is not listed in “Rare and Previous Plants of Hong Kong” (AFCD 2003).

Agricultural Land

7.5.9 Agricultural fields are the dominant lowland habitat type within the Study Area and are composed of a mosaic of active dry, inactive dry, as well as inactive wet agricultural land. The status of the agricultural habitats in rural areas, and hence the ecological functions they provide, is subject to the management practice of the tenant. During the field survey it was noted that whilst most agricultural fields were inactive and dry, farming in some of the land lot had either been recently resumed or suspended with most of the crops or vegetation cleared. Plant species commonly found on the barren dry agricultural lands were the grasses Neyraudia arundinacea and Panicum maxima, as well as banana trees. Inactive wet agricultural land is mainly located in the south of the Study Area and is covered by a mix of grasses (Bothriochloa ischaemum, Apluda mutica), weedy climbers (such as Mikania micrantha, Ipomoea aquatica and Ipomoea cairica) and ferns (Lygodium japonica and Cyclosorus interruptus).

Woodlands

7.5.10 There is an extensive area of semi-natural woodland established on a hill-slope to the north of the stream. The woodland canopy is mainly composed of native tree species at the height of about 8 - 10 m. The sub-canopy of the woodland is well-stratified and densely vegetated with climbers, young sapling trees, as well as shrubs. Trees species commonly found in the canopy are Machilus breviflora, Cinnamomum camphora, Celtis tetrandra and Schefflera octophylla, whereas the sub-canopy is dominated by the trees Ardisia quinquegona, Aporusa dioica, Microcos paniculatus, Diplospora dubia, Syzygium jambos, Aporousa dioica, the shrubs Litsea rotundifolia, Psychotria rubra and Sarcandra glabra, as well as the climbers Gnetum luofuense, Uvaria microcarpa and Desmos chinensis. No locally protected plant species were found within the woodland.

Vegetation

7.5.11 Eight individuals of the incense tree Aquilaria sinensis was found in the upper part of the woodland to the north of the stream; wild plants of this species are under State Protection (Category II) and are listed as vulnerable in the China Plant Red Data Book. This species is widely planted around rural villages and is common in Hong Kong. In any case the specimens are remote from the stream and would not be affected by the proposed development.

Mammals

7.5.12 No species of mammal were recorded during any of the surveys. However, a short study of land mammals, utilising infrared-triggered cameras, was conducted by Kadoorie Farm & Botanic Garden in July 2003 (KFBG 2004). This study found evidence of five species of large land mammals, including Wild Boar Sus scrofa, muntjac Muntiacus sp., Malayan Porcupine Hystrix brachyura, Small Indian Civet Viverricula indica and Chinese Ferret Badger Melogale moschata. In addition, burrows of Chinese Pangolin Manis pentadactyla were observed in the course of the KFBG study.

7.5.13 Among these six species, Malayan Porcupine is considered to be of potential global conservation concern, Indian Muntjac is of potential regional concern, and Chinese Pangolin is of regional concern (Fellowes et al., 2002). Chinese Pangolin is thought to be rare in Hong Kong (Shek, 2003).

Avifauna

7.5.14 This is a moderately species-rich area due to a certain degree of habitat variety and low levels of human disturbance. A total of 43 species was recorded during daytime surveys in this relatively small area, with no further night-time additions. Waterbird species recorded in the stream were Little Egret Egretta garzetta in six months, Green Sandpiper Tringa ochropus in two months and Eurasian Woodcock Scolopax rusticola in one month. These species were recorded as single birds on each occasion, and with the exception of Eurasian Woodcock, are typical of lowland streams, ponds and marshes across the New Territories (Carey et al. 2001). Although Little Egret is listed in Fellowes et al. (2002) as a species of conservation concern, this assessment is based largely on the breeding population; no breeding sites for this species are known in the vicinity of LMH01. Other species recorded using the stream were Common Kingfisher Alcedo atthis and Grey Wagtail Motacilla cinerea. Both are widespread in Hong Kong but are, respectively, wholly and largely wetland dependent.

7.5.15 Eurasian Woodcock was highlighted as a focal species of concern in the Study Brief. It was recorded during the surveys only at two sites. At Lin Ma Hang it was seen foraging in the stream in daytime on 29 January 2004. As Carey et al. (2001) state, Eurasian Woodcock is a scarce winter visitor and passage migrant to wooded areas; there is no close correlation with the presence or otherwise of streams, though it is likely that lower-lying damper areas close to streams are favoured.

7.5.16 The only other species of conservation significance recorded was Crested Serpent Eagle Spilornis cheela, a forest raptor that will be very little affected, if at all, by the drainage scheme, either during construction or operation.

Table 7.4

Wetland dependent bird species and bird species of conservation importance recorded at LMH01 during monthly surveys, October 2003 to September 2004. (WD = wetland dependent; (WD) = largely wetland dependent; CC = conservation significance; Level of Concern based on Fellowes et al. 2002)

Species	Status	Mean/visit	Range	In stream	Level of Concern
Little Egret Egretta garzetta	CC/WD	0.50	0-1	Y	Regional
Crested Serpent Eagle Spilornis cheela	CC	0.17	0-1	N	Local
Green Sandpiper Tringa ochropus	WD	0.17	0-1	Y
Eurasian Woodcock Scolopax rusticola	WD	0.08	0-1	Y
Common Kingfisher Alcedo atthis	WD	0.08	0-1	Y
Grey Wagtail Motacilla cinerea	(WD)	0.25	0-1	Y

Herpetofauna

7.5.17 A total of seven wetland-dependent amphibian species and eight reptile species were recorded at Lin Ma Hang during the surveys. Two reptile species of conservation interest (Indo-Chinese Rat Snake Ptyas korros and Common Rat Snake) were recorded. In addition, Chinese Waterside Skink Tropidophorus sinicus was reported from the Lin Ma Hang stream by KFBG (2004). This species is wetland-dependent.

Table 7.5

Wetland dependent herpetofauna species and species of conservation importance recorded at LMH01 during monthly surveys, October 2003 to September 2004

Species	Mean/visit	Range	At stream	Level of Concern
Asian Common Toad Bufo melanostictus	0.8 (4.0*)	0-7	Y
Gnther’s Frog Rana guentheri	2.2 (10.7*)	0-15	Y
Paddy Frog Rana limnocharis	1.6 (7.3*)	0-10	N
Brown Tree Frog Polypedates megacephalus	0.9 (5.3*)	0-7	N
Spotted Narrow-mouthed Frog Kalophrynus interlineatus	0.2 (1.3*)	0-2	N
Asiatic Painted Frog Kaloula pulchra	0.1 (0.7*)	0-2	Y
Ornate Pigmy Frog Microhyla ornata	1.0 (5.3*)	0-6	N
Chinese Waterside Skink** Tropidophorus sinicus	na	na	Y
Indo-Chinese Rat Snake Ptyas korros	0.06	0-1	Y	Potential regional
Common Rat Snake Ptyas mucosus	0.06	0-1	N	Potential regional

* night surveys only; **KFBG 2004

Fish

7.5.18 Because this taxon group has been extensively studied in recent years, no additional fish surveys were conducted at LMH01. A total of 18 species, including four species of conservation interest, have been reported from the stream by Chan (2001) and KFBG (2004) and one further species, the Striped Loach Schistura fasciolata, was listed in the EIA Study Brief as occurring. These are listed below. Chinese Rasbora Rasbora steineri and Spiny Eel Mastacembelus armatus are very rare in Hong Kong (KFBG 2004), while Small Snakehead Channa asiatica and Snakehead Murrel Channa striata are uncommon (Lee et al. 2004). The latter species is, however, introduced.

Table 7.6

Freshwater fish species recorded at LMH01 by Chan (2001) and/or KFBG (2004). Level of concern based on Fellowes et al., 2002

Species	Frequency in LMH01*	Level of concern/ conservation rating
Chinese Rasbora Rasbora steineri	A	Global
Predaceous Chub Parazacco spilurus	A	‘Vulnerable’, China Red Data Book 1998; ‘Endangered’, IUCN 2000
Nicholsicypris normalis	A
Wild Carp Hemicultur leucisculus	C
Rasborinus lineatus	R
Chinese Barb Puntius semifasciolatus	A
Goldfish Carassius auratus	C
Oriental Weatherfish Misgurnus anguillicaudatus	C
Striped Loach Schistura fasciolata	na
Vietnam Catfish Pterocryptis cochinchinensis	R
Whitespotted Walking Catfish Clarias fuscus	R
Swampy Eel Monopterus albus	C
Spiny Eel Mastacembelus armatus	R	Local
Mosquito Fish Gambusia affinis	R
Tilapia Oreochromis sp.	C
Rhinogobius duospilus	C
Paradise Fish Macropodus opercularis	R
Small Snakehead Channa asiatica	C	Local
Snakehead Murrel Channa striata	C

* R = rare; C = common; A = abundant; na = unknown (but this species is common in Hong Kong (Lee et al. 2004)).

Butterflies

7.5.19 A large total of 52 butterfly species was recorded at LMH01 during the surveys. This included one wetland dependent species and two species of conservation interest. In addition, a further eight butterfly species were reported from Lin Ma Hang by KFBG (2004), including one species of conservation interest.

7.5.20 Glassy Bluebottle Graphium cloanthus and Centaur Oak Blue Arhopala pseudocentaurus are rare in Hong Kong, Short-banded Sailer Phaedyma columella, Chocolate Royal Remelana jangala and Grass Demon Udaspes folus are uncommon. A further three species reported by KFBG (2004), Gaudy Baron Euthalia lubentina, Quaker Neopithecops zalmora and Rare Swift Parnara bada, are also uncommon in Hong Kong (Young & Yiu, 2002).

Table 7.7

Wetland dependent butterfly species and species of conservation importance recorded at LMH01 during monthly surveys, October 2003 to September 2004. (CC = conservation concern; WD = wetland dependent. Level of concern based on Fellowes et al., 2002)

Species	Status	Mean/visit	Range	Level of Concern
Glassy Bluebottle* Graphium cloanthus	CC	Na	na	Local
Small Grass Yellow Eurema brigitta	CC	0.4	0-2	Local
Centaur Oak Blue Arhopala pseudocentaurus	CC	0.1	0-1	Local
Bush Hopper Ampittia dioscorides	WD	0.2	0-1

*KFBG, 2004

Dragonflies

7.5.21 A large total of 25 dragonfly species was recorded from LMH01 during the surveys. This included three species (Blue Sprite Pseudagrion microcephalum, Club-tailed Cruiser Macromia urania and Emerald Cascader Zygonyx iris) of conservation significance. In addition, one further species, Dancing Shadow-emerald Idionyx victor, was reported from Lin Ma Hang stream by KFBG (2004). This species is also of conservation significance.

Table 7.8

Dragonfly species recorded at LMH01 during monthly surveys, October 2003 to September 2004. (WD = wetland dependent; CC = conservation concern; Level of Concern based on Fellowes et al. 2002)

Species	Status	Mean/ visit	Range	At stream	Level of Concern
Chinese Greenwing Neurobasis chinensis	WD	0.6	0-4	Y
Black-banded Gossamerwing Euphaea decorata	WD	0.3	0-4	Y
Common Blue Jewel Rhinocypha perforata	WD	0.9	0-3	Y
Orange-tailed Sprite Ceriagrion auranticum	WD	2.3	0-9	Y
Common Bluetail Ischnura senegalensis	WD	2.5	0-6	N
Blue Sprite Pseudagrion microcephalum	CC/WD	0.2	0-2	N	Local
Orange-faced Sprite Pseudagrion rubriceps	WD	0.4	0-3	Y
Black Threadtail Prodasineura autumnalis	WD	2.8	0-16	Y
Black-kneed Featherlegs Copera ciliata	WD	2.1	0-10	Y
Yellow Featherlegs Copera marginipes	WD	3.3	0-15	Y
Giant River Hawker Tetracanthagyna waterhousei	WD	0.1	0-1	Y
Common Flangetail Ictinogomphus pertinax	WD	0.4	0-2	N
Dancing Shadow-emerald* Idionyx victor	CC/WD	na	Na	Y	Local
Club-tailed Cruiser Macromia urania	CC/WD	0.1	0-1	Y	Global; ‘Endangered’, IUCN 2000
Asian Amberwing Brachythemis contaminata	WD	2.7	0-6	N
Forest Chaser Lyriothemis elegantissima	WD	0.9	0-4	Y
Russet Percher Neurothemis fulvia	WD	0.1	0-1	N
Red-faced Skimmer Orthetrum chrysis	WD	0.5	0-3	Y
Marsh Skimmer Orthetrum luzonicum	WD	0.9	0-5	N
Common Red Skimmer Orthetrum pruinosum	WD	0.5	0-2	Y
Asian Widow Palpopleura sexmaculata	WD	0.2	0-2	N
Wandering Glider Pantala flavescens	WD	1.6	0-2	N
Variegated Flutterer Rhyothemis variegata	WD	0.3	0-3	N
Crimson Dropwing Trithemis aurora	WD	1.1	0-4	Y
Indigo Dropwing Trithemis festiva	WD	0.9	0-4	Y
Emerald Cascader Zygonyx iris	CC/WD	0.6	0-5	Y	Potential Global

*KFBG (2004)

Aquatic Invertebrates

7.5.22 Aquatic invertebrates were sampled at LMH01 on 8^th December 2004. Two sampling locations, approximately 100m apart, were selected. The upstream sampling location was at a riffle section in an exposed stretch of the stream, and the downstream sampling location was at a shaded, slow-flowing stretch. At each sampling location, four replicate kick samples were collected.

7.5.23 A total of 18 species of aquatic invertebrates was recorded at LMH01. Although this total is not particularly high, it includes representatives of several groups – notably the fly family Simuliidae and the caddisfly families Calamoceratidae and Hydropsychidae – which are typical of clean, undegraded streams. The downstream sampling location was slightly more species-rich, with 13 taxa as compared to ten taxa at the upstream sampling location.

Table 7.9

Mean number of aquatic invertebrates per kick sample recorded at LMH01,

December 2004

Taxon	Upstream location	Downstream location
Ceratopogonidae sp.		0.5
Chironomidae spp.	17.25	16
Dolichopodidae sp.		0.25
Simuliidae sp.	0.5	0.25
Tipulidae sp.	0.75
Unid. Diptera pupa	0.5
Calamoceratidae sp.		0.75
Hydropsychidae sp.	2
Gerridae sp.	0.5
Coenagrionidae sp.		0.25
Libellulidae sp.		0.5
Platycnemididae sp.	0.75	0.5
Baetidae sp.	0.5	0.75
Caenidae sp.	2
Palaemonidae sp.	1	3.75
Biomphalaria straminea		0.25
Melanoides sp.		0.25
Oligochaeta spp.		0.5

7.6 Ecological Survey for MUP03, MUP04A&B, MUP05

Study Area

7.6.1 The proposed channels at MUP03 and MUP05 are located along the westbound and eastbound carriageways of Sha Tau Kok Road, respectively; the channel of MUP04 is south of Sha Tau Kok Road, to the west of Loi Tung village. Habitats present around the three sites are roadside plantations, agricultural fields, wasteland, watercourse and developed land.

7.6.2 The Outline Zoning Plan for Man Uk Pin includes an area designated as a Conservation Area which supports dense semi-natural secondary woodland in proximity to the Pat Sin Leng Country Park. At its closest point the Conservation Area comprises a hillslope approximately 150 m from MUP04 and 200 m from the upper reaches of MUP05. In view of the small scale of the proposed drainage improvement works, the distance from the Conservation Area, the fact that the Conservation Area is above (i.e. uphill from) the stream and the Conservation Area habitat type (woodland) there will be no adverse impacts upon this area as a result of the project.

Project Area

7.6.3 The proposed Project Area largely comprises the MUP05 channel, which flows parallel with Sha Tau Kok Road, upstream from Wo Keng Shan Road (below which it has previously been channelised) (Figure 7.6). MUP04A is a very small stream which flows into MUP05 and which drains a wooded and former agricultural catchment to the south of Sha Tau Kok Road (Figure 7.7). MUP03 is a concrete-lined ditch which runs along the south side of Sha Tau Kok Road (Figure 7.8). The proposed project works are the channelisation of MUP05 and MUP04A and the construction or reconstruction of drains MUP03 and MUP4B which are intended to collect surface water run-off from village housing areas on the south side of Sha Tau Kok Road. MUP03 are proposed to flow directly into MUP05 whilst MUP04B would first join MUP04A.

7.6.4 Two additional streams, MUP01 and MUP02, are also proposed to be modified but are classified as Non-DP. These streams are also tributaries of MUP05 but with catchments to the north.

7.6.5 The immediate environs of the project areas are largely village housing and agricultural land, much of which is inactive. MUP05 supports a corridor of riparian vegetation and some sections are well shaded by large trees.

Habitats in the Study Area

Watercourses

7.6.6 The watercourse of MUP05 increases in width from 1 m to over 10 m from east to west and in depth from 1 m to over 2 m. The upstream section of MUP05 is heavily modified and is either lined with concrete or has been box-culverted. The water course was found to be fairly polluted and had little water flow during the dry season. Riparian vegetation, where present, was limited to those concreted sections which are shaded by weedy vegetation and some pioneer tree species. Species commonly found in this stretch of the watercourse are the herbs Bidens pilosa and Polygonum barbatum and the climbers Lygodium japonicum, Mikania micrantha and Ipomoea cairica, as well as small-sized trees Ficus hispida and Macaranga tanarius. Clumps of Alocasia macrorrhiza and Polygonum barbatum were also found within the channel.

7.6.7 The downstream section of MUP05 is semi-natural except for a channelised section and artificial weir at the downstream limit of the project area (Figure 7.9). It is bordered by a mosaic of agricultural habitats to the north, and wasteland or developed stockyard to the south. This section of the watercourse ranges from 1 m to 1.5 m in depth and 1.5 m to 4 m in width (except the channelised section, in which the width is over 12 m). Water flow during the dry season was found to be moderate and shallow. The riparian zone is in general well-vegetated by a mix of trees, shrubs and herbaceous plants (Figure 7.10). The dominant tree species include those present in the adjacent roadside plantation (see above) and also species commonly found in rural villages, including Ficus microcarpa, Dimocarpus longan and Litchi chinensis. These species are mostly located on the northern bank. Other species commonly found in the riparian zone along this stretch include a mix of herbaceous weedy plant established on the bank (such as the herbs Bidens pilosa, Ageratum conyzoides and the climbers Mikania micrantha and Ipomoea cairica), as well as a few wetland plant species (such as the herbs Polygonum barbatum, Rumex maritimus and Ranunculus scleratus).

7.6.8 The upper stream of MUP04A ranges from 0.5 m to 1 m in depth and 1 m to 2 m in width. Water flow during the dry season survey was slow and shallow, and the water’s edge was well-vegetated: patches of plants were often found covering more than 90% of the water surface. Species found are mostly common species including the herbs Polygonum chinense, Polygonum barbatum, Ranunculus scleratus, Rumex dentatus, Lindernia crustacea, and Colocasia esculenta, the creepers Commelina diffusa and Alternanthera sessilis and the climber Mikania micrantha. Several patches of Polygonum japonicum and Azolla pinnata were found in the middle of the upper stream; these two species were categorized as rare in Xing et al. (2000). Polygonum japonicum is mainly found along the water edge in rural stream or ditches, whereas Azolla pinnata is a floating herb only found in ponds, paddy fields and ditches in rural area.

7.6.9 The lower section of MUP04A is partially channelised and is bordered by wasteland to the east and a nursery to the west. The width and the depth of the channel range from 2 m to 3 m and 0.5 m to 1 m respectively, and the water flow during the dry season was found to be slow and shallow. The watercourse is quite polluted and disturbed by the intensive human activities nearby; vegetation growth within the watercourse and along the riparian zone was found to be poor and was dominated by wasteland plant species or planted fruit trees. Nevertheless, patches of common wetland plants, especially the herbs Alternanthera sessilis and Commelina diffusa, were found in areas where the concrete lining had been broken, and a layer of Lemna perpusilla was found floating on the water surface.

Agricultural Land

7.6.10 A mosaic of agricultural habitats, consisting of patches of active dry, inactive dry and inactive wet agricultural fields, are the dominant habitat types found around the proposed channels. No agricultural field was found flooded and cultivated with wet produce in the immediate vicinity of the streams.

7.6.11 The active agricultural fields within the Study Area were planted with dryland crops during the field survey. This habitat type is generally poor in terms of species and structure because of farming practices such as monoculture, weeding and application of pesticide. Natural vegetation cover is very limited and mainly confined to the edges of footpaths or ditches within the field.

7.6.12 In contrast, most of the inactive agricultural fields were well vegetated, and showed evidence of being seasonally wet, with water-logged soil and the presence of a number of native wetland plants. Plant species commonly found on the inactive dry agricultural fields include the grasses Pennisetum alopecuroides, Pennisetum purpureum, Imperata cylindrica, Neyraudia arundinacea and Panicum maxima, as well as the herbs Conyza canadiensis, Ageratum conyzoides and Bidens pilosa.

7.6.13 Dominant species in the fields believed to be seasonally wet were the grasses Bothriochloa ischaemum and Apluda mutica, the creeping/climbing plants Commelina diffusa, Mikania micrantha and Ipomoea cairia, the herbs Lindernia crustacea, Alocasia macrorrhiza, Polygonum barbatum and Ranunculus scleratus and the ferns Lygodium japonica and Cyclosorus interruptus. Species diversity was found to be poor because of the dominance of the two grassy species, except in areas near the edge of the fields where open water and a mix of native wetland plant species, such as the herbs Polygonum barbatum, Saururus chinensis, the creeper Commelina diffusa, and the fern Cycolosorus interruptus, were found.

Wasteland

7.6.14 Wasteland in the area is mostly in small patches located alongside Sha Tau Kok Road and close to Loi Tung village. It derives from derelict land where previous land uses, such as stockyards or temporary work areas, have destroyed or removed the natural soil profile and vegetation. Ecological characteristics are dependent on the magnitude of previous disturbance and the vegetation cover established, and range from barren land to densely vegetated areas. Dominant plant species in the wasteland habitat are the trees Leucanea leucocephala and Macaranga tanarius, the herbs Bidens pilosa, Conyza canadensis and Ageratum conyzoides, the climbers Mikania micrantha and Ipomoea cairica, and the grasses Imperata cylindrica, Neyraudia arundinacea and Panicum maxima.

Woodlands

7.6.15 Roadside plantations are located on each side of Sha Tau Kok Road; on the northern side there are also a number of planters containing shrubs. Most of the vegetation (both trees and shrubs) planted is exotic in origin and is actively maintained. This habitat type is fairly fragmented and patchily distributed in the area of the proposed channels, and has suffered from intensive disturbance either during the landscaping maintenance or unauthorized human activities. Trees found along the northern side of the road are mostly fast-growing exotic species, such as Melaleuca leucadendron, Acacia mangium, Bombax ceiba and Eucalyptus robusta, with an average height greater than 10 m; on the southern side is a row of the exotic Melaleuca leucadendron, 8 to 10 m tall.

7.6.16 A range of native tree species and their saplings are found scattered among the roadside plantation, especially in the planters or along the edge of the watercourse, and form part of the the canopy; examples of such species include Hibiscus tiliaceus, Celtis tetrandra, Bridelia tomentosa and Cinnamomum camphora, as well as other pioneer tree species such as Ficus hispida and Macaranga tanarius.

7.6.17 The planters are mostly planted with shrubby and ornamental plants which are exotic in origin; species commonly found include Schefflera arboricola, Caroyta ochlandra, Chrysalidocarpus lutescens, Aglaia odorata, Hibiscus rosa-sinensis and Calliandra haematocephata. Many common weedy herbaceous plants are also often found within the planters, such as Ageratum conyzoides and Bidens pilosa.

7.6.18 Because of the man-made origin of this habitat type and the limited ecological functions provided by the planted exotic species, the ecological value of this habitat type is considered to be very low.

Mammals

7.6.19 Two species of mammal, Malayan Porcupine Hystrix brachyura and Wild Boar Sus scrofa were recorded at MUP03-05.

7.6.20 Malayan Porcupine is listed by Fellowes et al. (2002) as of potential global concern on the basis of a Global Red List (IUCN) categorisation as vulnerable; it is protected in Hong Kong under Cap.170 of the Wild Animals Protection Ordinance (WAPO). Out of a total of 19 field survey visits between October 2003 and September 2004, evidence of porcupines (in the form of paw prints or scats) was found on three occasions. All records were from the southern part of the study area (i.e. south of Loi Tung).

7.6.21 Wild Boar are common in the North East New Territories, and are of no conservation significance. Out of a total of 19 field survey visits between October 2003 and September 2004, evidence of boar (in the form of hoof prints or scats) was found on ten occasions. Individuals were seen on a further two occasions. The majority of records were from the southern part of the study area (i.e. south of Loi Tung); one sighting was however made on the access road to Man Uk Pin village.

Avifauna

7.6.22 Although the area has a fairly diverse bird community at 50 species, the number of waterbirds utilising the stream channels appears to be rather low, with only three species recorded. All were recorded in MUP05. These comprised Little Egret Egretta garzetta, Chinese Pond Heron Ardeola bacchus and White-breasted Waterhen Amaurornis phoenicurus, which are widespread in Hong Kong. All were recorded in low numbers, however. Little Egret and Chinese Pond Heron are both listed as being of regional concern by Fellowes et al. (2002), based on the breeding populations. There is no known breeding site locally and the presence of small numbers of foraging individuals is not considered to be of significance.

7.6.23 The only other species of significance was Eurasian Woodcock Scolopax rusticola, which is a scarce winter visitor and passage migrant to wooded areas (Carey et al. (2001); this individual was recorded in abandoned agricultural land 50 m away from the stream course. There is no close correlation between Eurasian Woodcock and the presence or otherwise of streams, though it is likely that lower-lying damper areas close to streams are favoured.

7.6.24 Other wetland dependent species recorded utilising the river channel were Common Kingfisher Alcedo atthis and Grey Wagtail Motacilla cinerea. In addition the riparian vegetation was also utilized by Long-tailed Shrike Lanius schach, Oriental Magpie Robin Copsychus saularis, Grey-backed Thrush Turdus hortulorum, Masked Laughingthrush Garrulax perspicillatus, Japanese Bush Warbler Cettia diphone, Asian Brown Flycatcher Muscicapa dauurica, Red-throated Flycatcher Ficedula albicilla and Great Tit Parus major. None of these species are restricted to wetland habitats and none are listed as of conservation concern by Fellowes et al. (2002). No extra species were recorded during night-time surveys.

Table 7.10

Wetland dependent bird species and bird species of conservation importance recorded at MUP03 - 05 during monthly surveys, October 2003 to September 2004. (WD = wetland dependent; (WD) = largely wetland dependent; CC = conservation significance; Level of Concern based on Fellowes et al. 2002)

Species	Status	Mean/ visit	Range	In stream	Level of Concern
Little Egret Egretta garzetta	WD	0.17	0-2	Y	Regional
Chinese Pond Heron Ardeola bacchus	WD	0.25	0-3	Y	Regional
White-breasted Waterhen Amaurornis phoenicurus	WD	0.42	0-1	Y
Eurasian Woodcock Scolopax rusticola	WD	0.17	0-1	N
Common Kingfisher Alcedo atthis	WD	0.17	0-1	Y
Grey Wagtail Motacilla cinerea	(WD)	0.67	0-2	Y

Herpetofauna

7.6.25 A total of eight species of wetland-dependent amphibian (Asian Common Toad, Gnther’s Frog, Paddy Frog, Green Cascade Frog Rana livida, Brown Tree Frog, Spotted Narrow-mouthed Frog Kalophrynus interlineatus, Asiatic Painted Frog and Marbled Pigmy Frog Microhyla pulchra) and eight species of reptile was recorded from the combined area of MUP03 - 05 and the adjacent MUP01 - 02 (which falls within the 500m study boundary for MUP03 - 05). These included one species, Common Rat Snake Ptyas mucosus of conservation significance.

Table 7.11

Wetland dependent herpetofauna species and species of conservation importance recorded at MUP03 - 05 during monthly surveys, October 2003 to September 2004

Species	Mean/visit	Range	At stream	Level of Concern
Asian Common Toad Bufo melanostictus	1.6 (6.0*)	0-12	Y
Gnther’s Frog Rana guentheri	2.2 (6.8*)	0-13	Y
Paddy Frog Rana limnocharis	2.7 (10.3*)	0-23	Y
Green Cascade Frog Rana livida	0.2 (1.0*)	0-2	Y
Brown Tree Frog Polypedates megacephalus	2.6 (11.5*)	0-23	N
Spotted Narrow-mouthed Frog Kalophrynus interlineatus	0.2 (0.8*)	0-2	N
Asiatic Painted Frog Kaloula pulchra	0.3 (1.5*)	0-4	Y
Marbled Pigmy Frog Microhyla pulchra	0.7 (3.3*)	0-8	N
Common Rat Snake Ptyas mucosus	0.05	0-1	N	Potential regional

* night surveys only

Fish

7.6.26 A total of six species of freshwater fish was recorded in the stream at MUP04A during the survey conducted in January 2005. It should be noted that water level in the stream was low and fish were, with the exception of the Guppy Poecilia reticulata generally present in very low numbers during the survey. A further four species of fish have been recorded from this stream in recent years (Chan & Dudgeon 2001, Chan unpublished data), giving a combined total of ten species. Small Snakehead Channa asiatica and Snakehead Murrel Channa striata are uncommon in Hong Kong (Lee et al. 2004) and the former is a species of local conservation concern (Fellowes et al. 2002). Snakehead Murrel is an introduced species.

7.6.27 Five species were recorded in the stream at MUP05 during the survey. A further two species of fish have been recorded from this stream in recent years, including the conservation-significant Small Snakehead (Chan & Dudgeon 2001, Chan unpublished data), giving a combined total of seven species.

Table 7.12

Freshwater fish recorded from MUP04A and MUP05 in January 2005, and other recent records. (Level of Concern based on Fellowes et al. 2002)

Species	Abundance		Level of concern
	MUP04A	MUP05
Freshwater Minnow Zacco platypus		High
Chinese Barb Puntius semifasciolatus	Low	Moderate
Oriental Weatherfish Misgurnus anguillicaudatus	Low
Whitespotted Walking Catfish Clarias fuscus	Literature record
Swampy Eel Monopterus albus	Literature record
Mosquito Fish Gambusia affinis	Low	Low
Guppy Poecilia reticulata	High	Moderate
Mozambique Tilapia Oreochromis mossambicus	Low	Low
Goby sp. Rhinogobius duospilus		Literature record
Paradise Fish Macropodus opercularis	Literature record
Small Snakehead Channa asiatica	Literature record	Literature record	Local
Snakehead Murrel Channa striata	Low

Butterflies

7.6.28 A total of 54 butterfly species was recorded at MUP03 - 05. No species of conservation significance were recorded, although six species (Spotless Grass Yellow Eurema laeta, Yellow Orange Tip Ixias pyrene, Danaid Eggfly Hypolimnas misippus, Chocolate Pansy Junonia iphita, Bush Hopper Ampittia dioscorides and Grass Demon Udaspes folus) are considered uncommon in Hong Kong (Young & Yiu 2002).

7.6.29 A single wetland dependent species, Bush Hopper Ampittia dioscorides (Lo & Hui 2004), was recorded at MUP03 - 05.

Dragonflies

7.6.30 A large total of 35 dragonfly species was recorded at MUP03 - 05. This included a number of stream specialist species, recorded within the 500 m study area (on the main stream in MUP02), including Chinese Greenwing Neurobasis chinensis, Black-banded Gossamerwing Euphaea decorata, Common Blue Jewel Rhinocypha perforata, Black Threadtail Prodasineura autumnalis, Indigo Dropwing Trithemis festiva and Emerald Cascader Zygonyx iris.

7.6.31 One species of conservation significance were recorded. Emerald Cascader has a restricted global range, and is known only from Hong Kong and Hainan. Several individuals were observed along the MUP02 stream on 3^rd May and 28^th June. Note that conservation-significant records of Elegant Clubtail Leptogomphus elegans and Giant Hooktail Megalogomphus sommeri were made in the ES study of MUP02, but these records were made at the northern part of the study area for MUP02, and did not therefore fall within the 500m study area for MUP03 - 05. However, it is highly possible that these species would also be present further downstream on MUP02, thereby bringing them within the 500m study area for MUP03 - 05.

7.6.32 In addition, Wandering Midget and Orange-faced Sprite, whilst not of particular conservation significance, are considered uncommon in Hong Kong (Wilson, 2004).

Table 7.13

Dragonfly species recorded at MUP03 - 05 during monthly surveys, October 2003 to September 2004. (WD = wetland dependent; CC = conservation significance; Level of Concern based on Fellowes et al. 2002)

Species	Status	Mean/ visit	Range	At stream	Level of Concern
Chinese Greenwing Neurobasis chinensis	WD	0.7	0-3	Y
Black-banded Goss’wing Euphaea decorata	WD	0.7	0-6	Y
Common Blue Jewel Rhinocypha perforata	WD	0.01	0-1	Y
Orange-tailed Midget Agriocnemis femina	WD	2.4	0-8	N
Wandering Midget Agriocnemis pygmaea	WD	0.2	0-2	N
Orange-tailed Sprite Ceriagrion auranticum	WD	5.0	0-16	Y
Common Bluetail Ischnura senegalensis	WD	5.6	0-13	N
Orange-faced Sprite Pseudagrion rubriceps	WD	1.6	0-7	Y
Black Threadtail Prodasineura autumnalis	WD	2.1	0-16	Y
Black-kneed Featherlegs Copera ciliata	WD	1.1	0-4	Y
Yellow Featherlegs Copera marginipes	WD	6.1	0-29	Y
Pale-spotted Emperor Anax guttatus	WD	0.06	0-1	Y
Golden Flangetail Sinictinogomphus clavatus	WD	0.1	0-1	N
Asian Pintail Acisoma panorpoides	WD	0.4	0-2	N
Blue Dasher Brachydiplax chalybea	WD	0.4	0-2	N
Asian Amberwing Brachythemis contaminata	WD	4.1	0-22	N
Crimson Darter Crocothemis servilia	WD	0.9	0-3	Y
Blue Percher Diplacodes trivialis	WD	1.3	0-5	N
Amber-winged Glider Hydrobasileus croceus	WD	0.06	0-1	Y
Forest Chaser Lyriothemis elegantissima	WD	0.1	0-1	N
Russet Percher Neurothemis fulvia	WD	0.3	0-2	Y
Pied Percher Neurothemis tullia	WD	0.06	0-1	N
Red-faced Skimmer Orthetrum chrysis	WD	0.3	0-2	N
Common Blue Skimmer Orthetrum glaucum	WD	0.7	0-3	Y
Marsh Skimmer Orthetrum luzonicum	WD	1.7	0-6	N
Common Red Skimmer Orthetrum pruinosum	WD	2.1	0-7	Y
Green Skimmer Orthetrum sabina	WD	3.4	0-10	N
Wandering Glider Pantala flavescens	WD	4.1	0-41	Y
Pied Skimmer Pseudothemis zonata	WD	0.3	0-3	N
Variegated Flutterer Rhyothemis variegata	WD	3.9	0-19	N
Evening Skimmer Tholymis tillarga	WD	0.1	0-1	N
Saddlebag Glider Tramea virginia	WD	1.0	0-5	Y
Crimson Dropwing Trithemis aurora	WD	2.9	0-9	Y
Indigo Dropwing Trithemis festiva	WD	1.3	0-6	Y
Emerald Cascader Zygonyx iris	WD/ CC	0.5	0-5	Y	Potential Global

Aquatic Invertebrates

7.6.33 Aquatic invertebrates were sampled at MUP04A and MUP05 on 8^th December 2004. One sampling location was selected at MUP04A, while two sampling locations, approximately 100m apart, were selected at MUP05. At each sampling location, four replicate kick samples were collected.

7.6.34 Eight species of aquatic invertebrates were recorded at MUP04A. Taxonomic composition was typical of degraded lowland streams, although abundance was low even amongst taxa such as chironomid midge larvae and oligochaete worms, which normally thrive in such conditions.

7.6.35 Twelve species of aquatic invertebrates were recorded at MUP05, with only minor differences in composition and abundance between the two sampling locations. Chironomid midge larvae were numerically dominant at both sampling locations.

Table 7.14

Mean number of aquatic invertebrates per kick sample recorded at MUP04A and MUP05, December 2004

Taxon	MUP04A	MUP05 upper	MUP05 lower
Ceratopogonidae sp.	2
Chironomidae spp.	0.25	85.5	46.25
Dolichopodidae sp.	0.25
Simuliidae spp.		2.5	1.5
Gerridae sp.		0.5
Coenagrionidae sp.			0.25
Platycnemidiae sp.		0.25
Baetidae sp.	0.25	5.75	5.25
Corbiculidae sp.		0.25
Sphaeriidae sp.	8.5
Brotia hainanensis		0.25
Melanoides sp.		2.75	0.5
Planorbidae sp.	0.25
Pomacea lineata	0.25		0.5
Hirudinea sp.		1.75	1.75
Oligochaeta spp.	3.75		0.25

7.7 Ecological Value of Streams and other Habitats Within or Adjacent to the Project Areas

7.7.1 Ecological evaluations are based on the general guiding principle that natural streamcourses are, at minimum, of Low to Moderate Ecological Value unless they are both very small and/or grossly polluted.

LMH01

7.7.2 Ecological evaluations of habitats within or adjacent to the the project area which will be (or may be) impacted by the development are detailed in Tables 7.15 to 7.18.

Table 7.15

Ecological Evaluation of Stream and Riparian Zone at LMH01

Criteria	Stream and Riparian Zone
Naturalness	Semi-natural stream with limited disturbance.
Size	Relatively small.
Diversity	The stream flora is structurally diverse providing a range of micro-habitats; there is a good diversity of dragonfly species and invertebrates typical of clean streams.
Rarity	Unpolluted and not channelised semi-natural streams are a relatively rare habitat in Hong Kong and the number of such streams continues to decline.
Re-creatability	Readily re-creatable provided resources are available.
Fragmentation	Not fragmented.
Ecological linkage	Some functional linkage to adjacent inactive agricultural land.
Potential value	Moderate in the absence of disturbance (constrained by small size).
Nursery/ breeding ground	Significant as a breeding ground for dragonfly and fish species of conservation importance.
Age	N/A
Abundance/ Richness of wildlife	Diverse fish and dragonfly communities. Polygonum japonicum is considered to be rare; four fish species and four dragonfly species of conservation concern have been recorded, including one fish and two dragonflies of global concern.
Ecological value	Moderate to High

7.7.3 Though it is small in size, the LMH01 stream remains in a largely natural state and, most importantly, is unpolluted. Present and previous surveys have confirmed the importance of the stream for fish and dragonflies, both pollution-sensitive groups.

Table 7.16

Ecological Evaluation of Agricultural Habitats at LMH01

Criteria	Inactive Wet and Inactive Dry Agricultural Land
Naturalness	Man-made habitat suffering from varying degree of disturbance.
Size	Relatively small.
Diversity	Plant species diversity and structural complexity is generally poor, moderate diversity of herpetofauna.
Rarity	Habitat is widespread in Hong Kong.
Re-creatability	Readily re-creatable provided resources are available.
Fragmentation	Habitat not fragmented.
Ecological linkage	Limited functional linkage of wet agricultural land with the stream.
Potential value	In the absence of agricultural activity, potential succession of wet agricultural land to semi-natural marsh and dry agricultural land the shrubland and forest.
Nursery/ breeding ground	No significant nursery or breeding ground recorded.
Age	N/A
Abundance/ Richness of wildlife	Indochinese Rat Snake and Common Rat Snake are of potential regional concern; Small Grass Yellow is of local concern.
Ecological value	Low to Moderate

7.7.4 Agricultural land in and adjacent to the project area is typical of village-edge areas in the New Territories, comprising a patchwork of recently utilized and long inactive fields. The extent of active farmland is dependant upon the activities of individual farmers and varies from season to season and year to year and it is likely that some of the inactive but recently utilized land will be returned to active agricultural use.

Table 7.17

Ecological Evaluation of Woodland at LMH01

Criteria	Woodland
Naturalness	Natural woodland dominated by native plant species.
Size	Moderate in size.
Diversity	Moderate in terms of floral diversity.
Rarity	A common habitat in Hong Kong; area is increasing due to vegetational succession.
Re-creatability	Possible if adequate resources are available and in the absence of disturbance, but original habitat characteristics including the community composition and structural complexity may require 10 to 20 years to establish.
Fragmentation	The woodland is not fragmented.
Ecological linkage	Not functionally linked to any highly value habitat in close proximity.
Potential value	Good potential to growth and expanding in terms of size and species diversity
Nursery/ breeding ground	No significant nursery or breeding ground recorded during the survey.
Age	Relatively young woodland (with respect to the size of the trees and the structure complexity and community composition).
Abundance/ Richness of wildlife	Crested Serpent Eagle, Glassy Bluebottle and Centaur Oak Blue are of local concern.
Ecological value	Moderate

7.7.5 There is a small area of natural woodland adjacent to the project area, which is relatively young and dominated by native species. As it is surrounded by shrubland and grassland, it has the potential to expand through natural vegetation succession. The design of the improvement works has been taken into account the ecological value of the woodland such that no trees in the woodland will be affected.

Table 7.18

Ecological Evaluation of Wasteland at LMH01

Criteria	Wasteland
Naturalness	Man-made habitat highly disturbed by human activity including litter-dumping.
Size	Relatively small.
Diversity	Low.
Rarity	Habitat is widespread in Hong Kong.
Re-creatability	Readily re-creatable.
Fragmentation	Fragmented due village development.
Ecological linkage	Not functionally linked to any habitat of ecological importance.
Potential value	Potential value limited due to fragmentation and proximity to village development.
Nursery/ breeding ground	No significant nursery or breeding ground recorded.
Age	N/A
Abundance/ Richness of wildlife	Abundance and diversity of wildlife very low. No rare species were recorded.
Ecological value	Very Low

MUP03, 04A, 04B, 05

7.7.6 Ecological evaluations of habitats within or adjacent to the project area which will be (or may be) impacted by the development are detailed in Tables 7.19 to 7.24.

Table 7.19

Ecological Evaluation of Streams and Riparian Zone at MUP04A

Criteria	Lower Watercourse	Upper Watercourse
Naturalness	The lower watercourse has been channelised.	The upper watercourse is semi-natural.
Size	Small.	Very small.
Diversity	Very poor species diversity and poor habitat diversity.	Species diversity and structural complexity is considered to be moderate.
Rarity	Small channelised streams are common in Hong Kong.	Very small natural watercourses remain quite common in Hong Kong and are less threatened than larger streams.
Re-creatability	Readily re-creatable provided resources are available	Readily re-creatable provided resources are available.
Fragmentation	Not fragmented	Not fragmented.
Ecological linkage	Flows into MUP05 but in view of disturbed character of MUP4A it is doubtful if functional linkages are significant.	Functionally linked to adjacent inactive wet agricultural land.
Potential value	Very low because of its concreted bottom and wall	Moderate in the absence of disturbance especially if surrounding wet agricultural land remains inactive.
Nursery/ breeding ground	No significant nursery or breeding ground recorded.	No significant nursery or breeding ground recorded.
Age	N/A	N/A
Abundance/ Richness of wildlife	Wildlife numbers and diversity low; no species of conservation importance.	Two plant species considered rare were found, Small Snakehead is of local conservation concern
Ecological value	Very Low	Low to Moderate

7.7.7 The lower watercourse at MUP04A has been heavily modified and provides very few ecological resources to wildlife except in those areas where patches of wetland plants are established. It is poor in species and structural diversity, and the riparian vegetation is dominated by common weedy herbaceous plants; the ecological value is thus considered to be low. On the other hand, the upper watercourse is a stretch of semi-natural stream which is bordered by agricultural habitats on both sides. The structural complexity and species diversity of the stream and its riparian flora are moderate, and the degree of shading and vegetation coverage are variable along the stream. However, since it is relatively small and fairly disturbed, the ecological value of the upper watercourse is considered to be low to moderate.

Table 7.20

Ecological Evaluation of Streams and Riparian Zone at MUP05

Criteria	Upper Watercourse	Lower Watercourse
Naturalness	The upper watercourse is a man-made habitat heavily modified and lined with concrete.	The lower watercourse is semi-natural and moderately modified.
Size	Generally narrow and relatively short.	Wide but relatively short.
Diversity	Very poor species diversity and poor habitat diversity.	Species diversity and structural complexity is considered to be moderate; riparian vegetation relatively diverse.
Rarity	Small channelised streams are common in Hong Kong.	Partially channelised watercourses with semi-natural sections are common in Hong Kong.
Re-creatability	Readily re-creatable provided resources are available.	Readily re-creatable provided resources are available.
Fragmentation	Not fragmented.	Not fragmented.
Ecological linkage	Not functionally linked to any natural or semi-natural habitats.	Functionally linked to MUP01 & 02 and MUP04A which are tributary streams.
Potential value	Very low because of its concreted bottom and wall.	Low to moderate as constrained by village development and Sha Tau Kok Road.
Nursery/ breeding ground	No significant nursery or breeding ground recorded.	No significant nursery or breeding ground recorded.
Age	N/A	N/A
Abundance/ Richness of wildlife	Wildlife numbers and diversity low.	Wildlife numbers and diversity moderate. Small Snakehead is of local conservation concern.
Ecological value	Very Low	Moderate

7.7.8 The upper watercourse of MUP05 has been heavily modified, with most of the bank and bottom concreted, and therefore provides very few ecological resources for wildlife, except in those areas where patches of wetland plants are established. It is poor in species and structural diversity and the riparian vegetation is dominated by common weedy herbaceous plants.

7.7.9 In contrast, the lower watercourse is a semi-natural stream, only a small section of which is channelised at present. It is structurally and spatially complex as a result of differences in riparian vegetation along each bank and variation in dimensions along the length.

7.7.10 It should be noted that MUP05 is functionally linked to the stream MUP02 which is a Non-DP. This stream is of high importance for dragonflies including one species of conservation significance recorded within the 500m boundary of the MUP05 project area.

Table 7.21

Ecological Evaluation of Roadside Plantation at MUP03, MUP04A, MUP04B and MUP05

Criteria	Plantation Woodland
Naturalness	Man-made habitat planted with exotic trees.
Size	Very small with high edge-to-area ratio.
Diversity	Very poor.
Rarity	No rare species was recorded.
Re-creatability	Readily re-creatable.
Fragmentation	Fairly fragmented.
Ecological linkage	Limited functional linkage to riparian vegetation on MUP05.
Potential value	Very limited additional potential due to species being exotic and proximity to Sha Tau Kok Road
Nursery/ breeding ground	No significant nursery or breeding ground recorded during the survey.
Age	Based on tree size estimated to be c. 20 years.
Abundance/ Richness of wildlife	Very Poor
Ecological value	Low

7.7.11 Plantation woodland is established on either side of the Sha Tau Kok Road and falls into the project areas for works to the existing streams as well as the proposed additional channels.

Table 7.22

Ecological Evaluation of the Agricultural Habitats at MUP03, MUP04A, MUP04B and MUP05

Criteria	Active Dry Agricultural Land	Inactive Dry & Wet Agricultural Land
Naturalness	Man-made habitat and intensively managed.	Man-made habitat in which farming is currently discontinued.
Size	Relatively small in size	Moderate in size as a whole, but wet areas small
Diversity	Poor species diversity and structural complexity.	Low in plant species diversity and poor in structural complexity.
Rarity	Not a rare habitat in Hong Kong.	Habitats widespread in Hong Kong.
Re-creatability	Readily re-creatable but dependant upon agricultural viability.	Readily re-creatable.
Fragmentation	The habitats are slightly fragmented	Largely unfragmented, especially around MUP04A.
Ecological linkage	Not functionally linked to any habitats of ecological importance.	Some linkages to MUP04A and MUP05 riparian zones and hillside woodland above MUP04A.
Potential value	Limited potential value because of the active farming practice and physical constraints.	In the absence of agricultural activity, potential succession of wet agricultural land to semi-natural marsh and dry agricultural land the shrubland and forest.
Nursery/ breeding ground	No significant nursery or breeding ground recorded.	No significant nursery or breeding ground recorded.
Age	N/A	N/A
Abundance/ Richness of wildlife	Both abundance and richness of wildlife low due to habitat characteristics and human activities.	Relatively diverse wildlife though generally of common and widespread species. Common Rat Snake is of potential regional concern.
Ecological value	Low	Generally Low but Moderate for MUP04A upper section

7.7.12 The agricultural habitats within the Study Area comprise a mosaic of active and inactive dry agricultural fields, as well as patches of inactive wet agricultural fields, some of which are seasonally wet. The close proximity of some fields to rural villages and the Sha Tau Kok Road causes some degree of disturbance; some of the inactive wet agricultural fields are, however, relatively undisturbed, notably those around the upper section of MUP04A. Species diversity of all of the agricultural habitats is generally poor and dominated by either the planted crops or grassy vegetation. The ecological value of the dry agricultural field is considered to be low because of the high level of disturbance, simple community structure and poor habitat diversity. The inactive wet agricultural fields are considered to have moderate ecological value because of the lack of disturbance and the potential ecological importance (for example by providing feeding and breeding habitats) during the wet season.

Table 7.23

Ecological Evaluation of Orchard/Nursery at MUP04A & MUP04B

Criteria	Orchard
Naturalness	Man-made habitat actively managed by humans.
Size	Area moderate but only a small fringe in the project area.
Diversity	N/A (nursery plantation).
Rarity	Habitat is widespread in Hong Kong.
Re-creatability	Readily re-creatable.
Fragmentation	Portion in the project area fragment by village development.
Ecological linkage	Not functionally linked to any habitats of ecological importance.
Potential value	Potential value limited as it is a actively farmed.
Nursery/ breeding ground	No significant nursery or breeding ground recorded.
Age	N/A
Abundance/ Richness of wildlife	Abundance and diversity of wildlife low. No rare species were recorded.
Ecological value	Very Low

Table 7.24

Ecological Evaluation of Bare Ground and Wasteland at MUP04A, MUP04B & MUP05

Criteria	Bare Ground and Wasteland
Naturalness	Man-made habitat highly disturbed by human activity including litter-dumping.
Size	Relatively small.
Diversity	Waste ground supports a low diversity (but typical for the habitat) of widespread early stage colonising ruderal plant species.
Rarity	Habitats are widespread in Hong Kong.
Re-creatability	Readily re-creatable.
Fragmentation	Fragmented due to roads and village development.
Ecological linkage	Not functionally linked to any habitat of ecological importance.
Potential value	Potential value limited due to fragmentation and proximity to village development.
Nursery/ breeding ground	No significant nursery or breeding ground recorded.
Age	N/A
Abundance/ Richness of wildlife	Abundance and diversity of wildlife very low. No rare species were recorded.
Ecological value	Very Low

7.8 Potential Ecological Impacts of Stream Channelisation Proposals

7.8.1 Assessment of potential ecological impacts of the channelisation proposals followed the following protocol:

· General Arrangement (GA) plans for the proposed channelisation works to each of the streams were reviewed and potential ecological impacts in the absence of ecologically-driven design changes or mitigation measures were identified and quantified. Ecological Impacts were categorized as follows:

- direct loss of habitat of ecological importance (permanent and temporary) due to construction works;

- indirect loss of habitat of ecological importance (permanent and temporary) due to construction works (including habitat loss due to disturbance or alteration of ecological conditions such as changes in the water table);

- direct and indirect effects on species of ecological importance due to permanent or temporary habitat loss or modification including downstream impacts due to disruption of stream flow or changes to siltation pattern;

- direct mortality due to construction works (primarily affecting vegetation but also some non-vagile animal groups);

- cumulative ecological impacts as a consequence of the combined and sequential effect of this project and other development projects. In this respect, the only project which was identified for which cumulative ecological impacts required to be assessed was the proposed NENT Landfill Extension, which is currently subject of an Environmental Impact Assessment study (Study Brief issued by the Director of Environmental Protection on 19^th May 2004). Though the boundary of the proposed landfill extension comes to within 250 m of the upper reaches of Lin Ma Hang stream, the proposed landfill extension area is actually over 1 km from the boundary of the current project. Thus, whilst potential ecological impacts of both the current project and the NENT Landfill Extension on the ecology of Lin Ma Hang Stream and its environs (e.g. the bat roost at Lin Ma Hang Lead Mine) require to be assessed, the physical separation of the two projects and their very different character are such that no potential cumulative ecological impacts have been identified. Accordingly, since no cumulative impacts have been identified, these are not included in the evaluation tables below.

· In the case of LMH01, no significant adverse ecological impacts requiring design changes were predicted (see below).

· In respect MUP03 – 05 potential to reduce adverse ecological impacts by design changes following the principle of Avoidance elucidated in EIAO Technical Memorandum Annex 8 were then considered through an iterative process whereby potential measures to minimize adverse ecological impacts were assessed as to their technical feasibility taking into account current stream environmental conditions, stream flow characteristics, flood prevention objectives and DSD design standards and management and maintenance protocols and practices. This process took into account Drainage Services Department Technical Circular No. 2/2004 Protection of Natural Rivers and Streams from Adverse Impacts Arising from Construction Works (DSD 2004), following the general principle of minimizing formation of artificial channels or, where this was not feasible, maximizing the extent of natural or semi-natural or at least relatively “wildlife-friendly” design solutions. Accordingly, for these channels the following analysis details the predicted ecological impacts based on the original designs as shown on GA drawings 382486/REV/027A and 028A (MUP03, 04A, 04B & 05) (Figures 7.11 and 7.12). The revised designs are shown on Figures 7.13 and 7.14 indicating how, and the extent to which, ecological impacts have been avoided through design changes

· The remaining ecological impacts of the designs as now proposed for all five streams are then detailed and additional measures for Minimisation and Mitigation (or Compensation) for impacts during the project detailed design and construction stage are then described. Finally, predicted unavoidable residual impacts, assuming implementation of all proposed mitigation measures are detailed and quantified.

LMH01

7.8.2 The DMP Study originally proposed a bypass channel of approximately 250 m in length and 13 m in width (with concrete base) to provide flood relieve to Lin Ma Hang village (see Figure 2.9). Given the ecological importance of the Lin Ma Hang stream, alternative solution to flooding problem were considered (details are presented in Section 2.5). At LMH01 it is proposed to undertake four short sections (c. 360 m in total) of bank stabilization and reinstatement using gabions. These works are required at sections of the stream where erosion is occurring. In addition, following consultations with Green Groups, it is proposed to undertake some replacement of banks which are lined with concrete at present by gabion-lined banks; this replacement will provide an ecological benefit as the gabion-lined banks will provide a better substrate for colonization by stream flora and habitat for stream fauna than the existing concrete-lined banks. It is also proposed to replace three bridges. The proposed locations of these works are to the south and west of Lin Ma Hang Village and at the point where the stream passes under the Boundary Road. At the latter location it is also proposed to replace the security grille across the stream under the road bridge (see Figure 2.4).

7.8.3 Of the 360 m of channel bank which will be reinstated, direct permanent loss of natural channel bank will be limited to 160 m as the remaining 200 m is already concrete-lined (see Figures 7.4 & 7.5). Loss of natural stream banks will be fully mitigated by the use of gabions for construction of the reinstated banks and replacement of 200 m of concrete-lined by gabion-lined banks. Gabion-lined banks represent a significant ecological improvement over concrete-lined banks, as riparian vegetation is able to colonise the interstices in the fill of the gabion, thus allowing the establishment of a riparian community which resembles that present in natural conditions.

7.8.4 The use of concrete for strengthening of banks is discouraged for ecologically important streams (EIS) such as Lin Ma Hang, and it is considered preferable to limit works to strengthening of sections of bank prone to erosion (ETWB Technical Circular No. 5/2005). The replacement of concrete banks with gabions therefore represents an improvement on existing conditions. As a consequence it is considered that permanent impacts on the ecological value of the stream will not be significant.

7.8.5 Temporary ecological impacts will include some disturbance to wetland birds during the construction period and a risk that works will increase sediment levels in the stream. However, both these impacts are expected to be small (and short term) due to the limited extent of the works and (in the case of wetland birds) because of the limited range of species and small numbers present. Disturbance impacts will be further limited by the fact that most works will occur close to the village.

7.8.6 Impacts to fish and other aquatic fauna could result from increased sedimentation, and the risk of pollution incidents, during the construction stage. Although the fauna of small streams is generally spate-adapted (Dudgeon & Corlett 2004) and as such is resistant to short-term changes in water flow and quality, including sediment load, there is scope to further minimise the potential impacts on these species by imposing conditions on working practices, especially by timing the works to take place during the dry season only, when these species are not breeding and low rainfall will limit run-off into the stream. For details see discussion of Minimisation measures below.

Table 7.25

Significance of Potential Ecological Impacts on LMH01

Type of Impact	Description / Notes	Scale/Importance of Impact
Habitat Quality: Direct Habitat Loss
Natural stream bank	A total of approximately 160 m of natural stream bank will be replaced by gabion lined stream bank. In addition 200 m of existing concrete-lined bank will also be replaced by gabion-lined bank.	Despite ecological value of stream (listed by AFCD as an EIS), impacts considered to be of Low Significance because the total length of stream bank to be impacted is relatively small and is restricted to a number of short sections rather than a single longer section. In addition, replacement of 200 m of concrete-lined by gabion-lined bank will be a net ecological improvement.
Habitat Quality: Indirect Habitat Loss
Wet agricultural land	An area of c. 1.5 ha of inactive wet agricultural land will be disturbed during the construction period.	Impacts Not Significant as the minor works required will only cause minimal disturbance and most of the works will occur close to the village in an area which is already disturbed by human activity.
Natural Stream Bed	No permanent impact to existing stream bed, although potential for construction phase impacts without appropriate planning of construction works.	Despite high quality of natural stream, impacts predicted to be Low because a short length of stream will be impacted and works will be carried out during the dry season to minimise adverse impacts.
Impacts on Species
Effects on fauna and flora	No plant species of conservation importance were identified which might be affected by the development. Fauna of conservation importance recorded include one bird species (Little Egret Egretta garzetta – although this species is not known to breed nearby), one reptile species (Indo-Chinese Rat Snake Ptyas korros), four fish species (Chinese Rasbora Rasbora steineri, Predaceous Chub Parazacco spilurus, Spiny Eel Mastacembelus armatus and Small Snakehead Channa asiatica), one butterfly species (Glassy Bluebottle Graphium cloanthus) and four dragonfly species (Blue Sprite Pseudagrion microcephalum, Dancing Shadow-emerald Idionyx victor, Club-tailed Cruiser Macromia urania and Emerald Cascader Zygonyx iris).	Potential impact to several species of stream fauna of conservation importance (fish, dragonflies) as a result of increased sediment load during construction phase. These stream species are adapted to hydrological changes, however, and these impacts would be minimised by adoption of appropriate measures (see below), including good site practice and timing of works during the dry season. By adoption of these measures, impacts to stream fauna are predicted to be Low. Impact to other species of conservation importance (birds, butterflies, reptiles) Not Significant because small numbers of individuals are involved, species are relatively mobile and are not stream-dependent.
Size of Impacts
Habitat area affected	Small area of agriculture (1.5ha) affected during construction, short stretch of stream bank (total 160m of natural stream, 200m of concrete-lined stream, but in a number of shorter sections) to be impacted permanently.	Impact Not Significant as total size of habitats affected is small and impacts to stream occur in a number of short stretches rather than a single long stretch (banks for most of the length of the stream will not be impacted).
Number of organisms affected	Low predicted direct mortality of stream fauna as a result of increased sediment loads during construction.	Low Impact owing to small scale and timing of proposed works to prevent pollution of stream.
Duration of Impacts
Loss of natural stream bank	Short sections of natural stream bank will be lost permanently.	Impact Not Significant as total length of bank affected is small.
Disturbance to wet agricultural land	Small area will be disturbed during construction work.	Impact Not Significant as area affected is small and impacts are temporary.
Impacts to aquatic fauna	Temporary impacts during construction as a result of increased runoff.	Impact Not Significant if works carried out during dry season and appropriate measures to minimise sedimentation and pollution are adopted.
Reversibility of Impacts
Increased stream sedimentation	Impacts would be relatively small and largely reversed by higher flow rates in following wet season.	Impacts Not Significant as conditions expected to return after works are completed. Only short sections of stream to be impacted, so stream fauna could be expected to recolonise rapidly from unaffected stretches.
Magnitude of Impacts
Disturbance to natural stream and loss of stream banks	Impacts limited to short stretches of stream bank.	Impacts Not Significant for Lin Ma Hang stream as a whole because of small scale of works compared to total length of stream.
Disturbance to wet agricultural land	Habitat loss small and levels of disturbance low and restricted to construction phase.	Impacts Not Significant as area affected is small and agricultural land already suffers low levels of disturbance.

MUP03, 04A, 04B & 05

Initial Design (GA drawings 382486/REV/027A and 028A) (Figures 7.11 and 7.12)

7.8.7 The initial design included the following elements:

- construction of 173 m length of new 2 m wide U-shaped vertical gabion- lined channel with concrete base and grade stone covering, constructed on the south side of Sha Tau Kok Road (MUP03);

- channelisation of 370 m of natural streamcourse MUP04A and transforming the stream into a 4 m wide U-shaped vertical gabion-lined channel with concrete base and grade stone covering;

- construction of 147 m length of new 2 m wide U-shaped vertical gabion-lined channel with concrete base and grade stone covering, constructed on the south side of Sha Tau Kok Road (MUP04B);

- replacement of 300 m of channelised stream (upper watercourse) MUP05 on the north side of Sha Tau Kok Road with a new 3 m wide U-shaped vertical gabion-lined channel with concrete base and grade stone covering;

- channelisation of 600 m of natural streamcourse (lower watercourse) MUP05 with 8 – 12 m wide grasscrete lined channel with v-shaped low flow channel; and

- works to streams MUP01 and 02 (which are non-DP).

Table 7.26

Significance of Potential Ecological Impacts of Initial Proposed Design for MUP03, 04A, 04B & 05

Type of Impact	Description / Notes	Scale/Importance of Impact
Habitat Quality: Direct Habitat Loss
Natural stream, natural stream bank and riparian vegetation.	Permanent loss of 600 m of moderate width semi-natural stream, stream bank and riparian vegetation at MUP05. Permanent loss of 370 m of semi-natural stream, stream bank and riparian vegetation at MUP04A.	Stream quality at MUP05 of Moderate ecological value with well-developed riparian vegetation and natural stream bed. Stream is, however, relatively disturbed. Impacts to MUP04A of Low to Moderate Significance because of moderate size and quality of stream to be affected.
Wet agricultural land	Permanent loss of active and inactive wet agricultural land to permit widening of existing streams and construction of new channels.	Significance of impact generally Low due to poor quality of habitats and high levels of disturbance, however impact for upper section of MUP04A Moderate for agricultural land undergoing succession to semi-natural marshland.
Dry agricultural land	Permanent loss of active and inactive dry agricultural land to permit widening of existing streams and construction of new channels.	Significance of impacts Low because habitat is of poor quality and heavily disturbed.
Orchard & Plantation	Permanent loss of orchard/plant nursery and roadside plantations to permit widening of existing streams and construction of new channels.	Impacts Not Significant as habitat is of low ecological value and heavily disturbed.
Habitat Quality: Indirect Habitat Loss
Wet agricultural land	Areas of inactive wet agricultural land will be disturbed during the construction period. Improved drainage of catchments may alter water tables and result in less land being wet and/or reduction in extent of wet periods.	Impacts generally Not Significant as most areas are small, fragmented and disturbed and of low or very low ecological value; but Low to Moderate Significance for upper section of MUP04A where there is a moderate-sized area of long inactive agricultural land which is undergoing succession to semi-natural marshland.
Impacts on Species
Fauna and flora of conservation importance.	No plant species of conservation importance were recorded which might be affected by the development. Fauna species of conservation importance included two bird species (Little Egret Egretta garzetta and Chinese Pond Heron Ardeola bacchus – although neither is known to breed nearby), one fish species (Small Snakehead Channa asiatica) and one dragonfly species (Emerald Cascader Zygonyx iris). Two further dragonfly species of conservation importance (Elegant Clubtail Leptogomphus elegans and Giant Hooktail Megalogomphus sommeri) were present on the tributary stream MUP02 and might be predicted to stray downstream into the MUP05 project area.	Impact Low as few species of conservation importance present and numbers of individuals low. However, scope for impacts to be reduced further by design revisions (see below). No permanent impacts on flora conservation importance. Despite local presence of Elegant Clubtail Leptogomphus elegans and Giant Hooktail Megalogomphus sommeri, impact to these species considered Not Significant as the stream area at MUP03, 04A, 04B and 05 is not thought to provide suitable breeding habitat for these species at present.
Loss of habitat	Permanent loss of habitat for obligate wetland fauna unable to utilize grasscrete or gabion-lined channels. Predicted decline in abundance and diversity of obligate aquatic fauna species due to reduced niche availability and changed stream characteristics. However, surveys suggested few species of conservation importance were present and numbers were small.	Small scale loss of habitat for wetland fauna, more significant for obligate aquatic species, notably fish and dragonflies.
Size of Impacts
Direct mortality	Some direct mortality of non-vagile fauna and vegetation inevitable but direct mortality of species of conservation importance likely to be very small as numbers are very small.	Significance of impact Low as few individuals will be impacted and most species present are common and widespread. Impact to riparian trees Low to Moderate at several large bankside individuals would be lost.
Loss of streams, stream bank and riparian vegetation	Direct permanent loss of semi-natural streams, stream bank and riparian vegetation totalling 600 m for MUP05 and 370m for MUP04A.	Significance of impacts Moderate due to length of semi-natural stream to be impacted.
Loss of other habitats	Direct loss of wet agricultural land, dry agricultural land and orchard/plantation. Indirect impact to wet agricultural land.	Significance of impact to agricultural and orchard/plantation habitats Very Low as small areas of heavily disturbed and widespread habitats impacted.
Duration of Impacts
Loss of habitats	Permanent direct loss of natural stream, wet and dry agricultural land and orchard/plantation. Indirect loss of some areas of agricultural land will be temporary during construction period only.	Significance of impacts Low to Moderate as impacts to important habitats (especially natural streams) would be permanent and irreversible.
Reversibility of Impacts
Loss of agricultural land, orchard and plantation	Agricultural land, orchard and plantation could be restored following construction, although this is unlikely in practise. Permanent loss of some areas of these habitats unavoidable.	Significance of impact Low as habitats are not natural, are of low quality and are widespread in the territory.
Magnitude of Impacts
Loss of natural stream	Moderate length of semi-natural stream and associated riparian vegetation at MUP05 and MUP04A would be lost by creation of channelised stream.	Magnitude of impact Moderate, as reasonable length of stream would be impacted and loss would include stream bed, stream banks and associated riparian vegetation.

7.8.8 Based on the above assessment, it was considered that the most significant adverse impacts were the habitat loss of the moderate-sized natural streamcourse and associated riparian vegetation on the lower section of MUP05 and the habitat loss of the small-sized natural streamcourse and associated area of inactive wet agricultural land on the lower section of MUP04A. The design review therefore focused on how design changes could avoid (or at least significantly reduce) these impacts.

Revised Design for MUP04A

7.8.9 The revised design for MUP04A is shown on Figure 7.14. Key design changes are as follows:

- the upper section of the stream will not be channelised (i.e. the length of the artificial channel will be reduced from 370 m to 184 m). This amendment avoids all impacts on the ecologically more important upper section of the stream (while still providing the required flood alleviation for the lower section which passes through a village area with disturbed anthropogenic habitats).

- the profile of the lower section has been adjusted in that the gabion channel sides are stepped rather than vertical, whilst the base will have a mattress lining rather than being formed with concrete. These changes will facilitate colonisation by herbaceous riparian vegetation and utilisation by riparian and wetland-associated fauna.

7.8.10 Overall, this revised design eliminates adverse ecological impacts to ecologically significant habitats (natural streamcourse and long-inactive wet agricultural land) at MUP4A and reduces impacts on riparian fauna, including species of conservation importance.

Revised Design for MUP05

7.8.11 The revised design for MUP05 is shown on Figures 7.13 and 7.14. Key design changes are as follows:

- the design of the ecologically important downstream section of the channel, which is currently a natural channel, has been changed. It is now proposed to increase the capacity of the channel by modifying and widening of banks but maintaining the existing natural channel bottom. One natural bank will be retained for 280 m of the channel, whilst both banks will be widened over a length of 320 m (Figure 7.15). Where the channel is widened, a two-stage system will be adopted, whereby the existing streambed will remain untouched while the widened channel bed will be constructed at a higher level than the existing channel, to increase capacity during periods of high flow while maintaining the existing natural stream at other times. The widened area of channel floor will receive a covering of natural rock and fines to simulate as closely as possible natural stream conditions. The widened channel will then be contained within a gabion embankment, on top of which will be built a grasscrete access track (Figures 7.13 & 7.14).

7.8.12 Overall, these design solutions accord with options (iii) (enlargement of channel by modifying one bank only) and (iv) (enlargement of channel by modifying/widening both banks) in Appendix C to the guidelines for protection of natural rivers and streams (ETWB Technical Circular No. 5/2005) and represent a significantly reduced adverse ecological impact compared with the previous design which accorded with option (vii) (i.e. the least preferred option).

Revised Design for MUP03

7.8.13 The revised design for MUP03 is shown on Figures 7.13 and 7.14. Key design changes are as follows:

- the profile of this channel has been adjusted in that the gabion channel sides are stepped rather than vertical, whilst the base will have a mattress lining rather than being formed with concrete. These changes will facilitate colonisation by herbaceous riparian vegetation and utilisation by riparian and wetland-associated fauna. A short section of U-channel is also proposed at MUP03.

7.8.14 Since these changes replace an existing concrete-lined ditch, no mitigation of adverse ecological impacts was required, however the revised gabion channel design improves the potential ecological value of the new channel. The new U-channel is essentially replacing or modifying the existing concrete channel, hence no design changes are proposed.

7.8.15 Table 7.27 details the predicted ecological impacts of the revised designs with the above changes incorporated.

Table 7.27

Significance of Potential Ecological Impacts of Revised Proposed Design for MUP03, 04A, 04B & 05

Type of Impact	Description / Notes	Scale/Importance of Impact
Habitat Quality: Direct Habitat Loss
Natural stream, natural stream bank and riparian vegetation.	Construction phase disturbance to 600 m of natural streambed at MUP05 during widening works, however level of disturbance can be reduced by adoption of suitable protocols. Existing stream channel will be retained in current condition with widened two-stage channel at a higher level; therefore no permanent impact through changes in conditions (for example reduced water depth in wider channel). Permanent loss of one natural stream bank for 280 m and both banks for 320 m, loss of 600 m of semi-natural riparian vegetation at MUP05. Impacts at MUP04A restricted to lower, channelised section; no impacts to natural upper stream.	Provided appropriate measures are taken to limit duration of impacts to stream bed, construction phase impacts to MUP05 of Low Significance. Significance of operational phase impacts Low with design and management measures to minimize and compensate for disturbance and loss of natural banks and riparian vegetation. Impacts to MUP04A Not Significant as short length of stream to be impacted is not natural and of poor quality.
Wet agricultural land	Permanent loss of 0.13 ha of inactive wet agricultural land to permit widening of existing streams and construction of new channels.	Impacts Not Significant due to small area and fragmented nature of habitats to be impacted and current high level of disturbance.
Dry agricultural land	Permanent loss of 1.35 ha of active and inactive dry agricultural land to permit widening of existing streams and construction of new channels.	Impacts Not Significant due to small area and fragmented nature of habitats to be impacted, low quality of habitat and current high levels of disturbance
Orchard & Plantation	Permanent loss of 0.11 ha of orchard and roadside plantation to permit widening of existing streams and construction of new channels.	Impacts Not Significant due to small area and fragmented nature of habitats to be impacted, low quality of habitat and current high levels of disturbance
Habitat Quality: Indirect Habitat Loss
Wet agricultural land	Areas of inactive wet agricultural land totaling c. 1.5 ha will be disturbed during the construction period. Improved drainage of catchments may alter water tables and result in less land being wet and/or reduction in extent of wet periods; however reduction of length of channelisation at MUP04A will protect the surrounding long-inactive agricultural land from this potential impact.	Impacts Not Significant as most areas are small, fragmented and disturbed and of low or very low ecological value. Following redesign of channel at MUP04A to avoid channelisation of upper stream, impacts to long-inactive agricultural land in this area also Not Significant.
Impacts on Species
Fauna and flora of conservation importance.	No plant species of conservation importance were recorded which might be affected by the development. Fauna species of conservation importance included two bird species (Little Egret Egretta garzetta and Chinese Pond Heron Ardeola bacchus – although neither is known to breed nearby), one fish species (Small Snakehead Channa asiatica) and one dragonfly species (Emerald Cascader Zygonyx iris). Two further dragonfly species of conservation importance (Elegant Clubtail Leptogomphus elegans and Giant Hooktail Megalogomphus sommeri) were present on the tributary stream MUP02 and might be predicted to stray downstream into the MUP05 project area.	Impact Not Significant as few species of conservation importance present and numbers of individuals low. Redesign of channel for MUP05 to retain natural streambed ensures no permanent loss of habitat for aquatic species of conservation importance (fish, dragonflies). Despite local presence of Elegant Clubtail Leptogomphus elegans and Giant Hooktail Megalogomphus sommeri, impact to these species considered Not Significant as the stream area at MUP03, 04A, 04B and 05 is not thought to provide suitable breeding habitat for these species at present.
Size of Impacts
Direct mortality	Some direct mortality of non-vagile fauna and vegetation inevitable but direct mortality of species of conservation importance likely to be very small as numbers are very small.	Significance of impact Low as few individuals will be impacted and most species present are common and widespread. Following redesign, significance of impacts to bankside trees also Low as many will be retained along sections of streambank not being impacted and others will be transplanted to suitable locations, where possible.
Loss of streams and stream banks	Construction phase disturbance to 600 m of natural stream during widening works, although measures taken to minimise the scale of impacts to the existing stream bed. Permanent loss of one natural stream bank for 280 m and both banks for 320 m, loss of 600 m of semi-natural riparian vegetation at MUP05. Permanent loss of stream banks represents c. 20% reduction on original design.	Construction phase impacts to natural streambed Low (with due diligence) provided appropriate measures are taken to avoid direct impacts to the existing stream channel, and to minimise potential indirect impacts through increased sedimentation (see mitigation measures below). Loss of stream banks and riparian vegetation Low (with due diligence) as impacts will be largely restricted to one side of the stream, although a moderate length will be impacted on both banks.
Loss of other habitats	Direct loss of 0.13 ha of wet agricultural land, 1.35 ha of dry agricultural land and 0.11 ha of orchard/plantation. Indirect (construction phase) impact to 1.5 ha of wet agricultural land.	Significance of impact to agricultural and orchard/plantation habitats Very Low as very small and fragmented areas of heavily disturbed and widespread habitats impacted.
Duration of Impacts
Loss of natural streams, stream banks and riparian vegetation	Impacts to some stretches of stream banks and riparian vegetation permanent. Existing stream bed will be temporarily impacted during construction period only; two-stage design includes retention of existing stream bed so no permanent impacts anticipated.	Significance of impacts to stream bed Low as these will be temporary during construction period only, following which recolonisation will occur from adjacent un-impacted areas. Impacts to banks Low as two-stage channel and gabion-lined banks will allow long-term recolonisation by riparian vegetation (although it may take a number of years to return to current conditions).
Loss of other habitats	Permanent direct loss of wet and dry agricultural land and orchard/plantation. Indirect loss of some areas of agricultural land will be temporary during construction period only.	Significance of impacts Very Low as impacts would be permanent, although habitats are of very low quality.
Reversibility of Impacts
Loss of natural stream, stream banks and riparian vegetation	Construction phase impacts to natural stream bed will be reversed after construction as a result of natural processes and recolonisation of aquatic fauna from adjacent sections of stream. Loss of riparian vegetation can be reversed in the long term by supplemental planting of appropriate tree species.	Significance of impacts to stream bed and riparian vegetation Low as these are short-term or are largely reversible through appropriate management. Loss of stream banks in current condition permanent and not reversible, although use of gabion embankment to simulate natural conditions mean impacts will be of Low Significance.
Loss of agricultural land, orchard and plantation	Agricultural land, orchard and plantation could be restored following construction, although this is unlikely in practice. Permanent loss of some areas of these habitats unavoidable.	Significance of impact Low as habitats are not natural, are of low quality and are widespread in the territory.
Magnitude of Impacts
Loss of natural stream banks	Moderate length of semi-natural stream banks and associated riparian vegetation at MUP05 would be impacted.	Magnitude of impact Low as much of stream will retain natural vegetation on one bank and use of gabions for reinstated banks will permit re-establishment of semi-natural flora.
Loss of stream bed	The two-stage design ensures existing stream bed will not be impacted for most of the length of the stream course, as the widened channel will be at a higher level than existing stream. Bed of widened channel will be covered with natural rock and fines to simulate as closely as possible the natural stream conditions.	Impacts Low as existing conditions will be retained as far as possible, and widened stream bed will be of similar condition.

7.9 Potential Adverse Ecological Impacts which require to be addressed by Minimization or Mitigation Measures

7.9.1 Table 7.28 details those remaining potential ecological impacts after design changes which are of a predicted scale and importance such that they require to be addressed by minimization or mitigation measures.

Table 7.28

Potential Adverse Ecological Impacts which require to be addressed by Minimization or Mitigation Measures

Stream	Description of Impact	Scale and Importance of Impact
LMH01	Construction stage mortality, habitat loss and disturbance of riparian fauna	Small scale and low importance (temporary and most species tolerant) – so long as appropriate controls on construction methods are followed.
MUP03	No significant ecological impacts	-
MUP04A	No significant ecological impacts	-
MUP04B	No significant ecological impacts	-
MUP05	Construction stage mortality, habitat loss and disturbance of riparian fauna	Low scale and importance – disturbance should be temporary and not affecting significant numbers and/or species of conservation importance.
	Permanent habitat loss for riparian fauna.	Low scale and importance so long as appropriate stream restoration and habitat/vegetation management regime is instituted.
	Permanent loss of natural stream banks	Low scale and importance (with due diligence) – loss should be temporary so long as appropriate stream restoration and habitat/vegetation management regime is instituted.
	Permanent loss of riparian vegetation	Low scale and importance (with due diligence) – loss should be temporary so long as appropriate stream restoration and habitat/vegetation management regime is instituted.

Proposed Measures to Minimise Adverse Ecological Impacts

LMH01

Construction Stage

7.9.2 Evaluation of potential ecological impacts suggested that there could be a small adverse impact on stream fauna of conservation importance (specifically fish and dragonflies) as a consequence of construction works increasing the sediment load in the stream – sediment could smother eggs or larvae. While the small scale of the works should ensure that suitable refuges remain for these species during the period of construction works, the potential adverse impact can be minimised by restricting works in the stream to the dry season. This would be beneficial for the following reasons:

- most fish and dragonflies have eggs and small larvae (the most vulnerable stages) during the wet season;

- heavy rain, washing soil exposed by construction activities into the stream is much less likely during the dry season; and

- lower stream levels during the dry season increase the proportion of the works area where activities are not influenced by the flowing stream.

7.9.3 Given the ecological importance of Lin Ma Hang stream, it is proposed that construction works at LMH01 should be restricted to the dry season period from 1^st November – 31^st March. The small scale of works should allow all construction to be completed within dry season to ensure that the risk of erosion and sedimentation due to heavy rain on the works areas, as well as disturbance impacts to surrounding areas, will be minimised.

7.9.4 In addition, the breaking of existing shotcrete banks at LMH01 should be restricted to hand-held equipment. Concrete should not be used for construction of the gabion banks.

7.9.5 Potential disturbance impacts to surrounding habitats and pollution risks (water quality impacts) to the stream should be minimised by adoption of appropriate site management procedures, as detailed in ETWB TCW No. 5/2005; including among others the location of access to the site and storage of materials and treatment of construction site waste to prevent pollution of the stream. These site management measures are listed in the subsequent section.

Operational Stage

7.9.6 Very little or no management / maintenance of the completed sections of LMH01 are expected. Removal of obstruction should be undertaken only when flooding or safety issues have been identified.

7.9.7 Environmental considerations for maintenance of the proposed gabion channels (see Section 5.8) should be adopted.

7.9.8 Vegetation management should be restricted to the removal of the exotic creeper Mikania micrantha which has previously been found to readily colonise gabion embankments. The establishment of this species would have a detrimental impact on the establishment of natural riparian vegetation. Control of Mikania and other invasive exotic species should be incorporated in the maintenance regime.

MUP05 (natural stream section)

Construction Stage

Streambed

7.9.9 One of the main benefits of the proposed stream widening measures is that the existing natural stream bed is left undisturbed. Accordingly, works should be carried out in such a way that as much as possible of the natural stream bed should be left undisturbed and that where disturbance is essential this should be minimised in terms of area, magnitude and duration to minimise potential impacts to stream fauna and to ensure refuges for these species during the period of the works. Avoidance of the stream bed can be achieved by conducting the earthworks to widen the stream from the landward side, by not lowering the widened channel to the same level as, or below, the existing channel, and by leaving the existing stream untouched except during the final stage, when the newly formed widened stream bed is joined to the existing stream.

7.9.10 In addition, the widened stream bottom should be floored with natural materials (natural rock and fines of varying sizes) to approximate as closely as possible to the rocky components of a natural stream bottom. Natural materials of a smaller particle size (sand and silt grains) will soon be deposited naturally.

7.9.11 In order to minimise potential impacts to stream fauna during excavation of the widened “two-stage” channel, this work should be limited to the dry season as far as possible, between 1^st October and 31^st March. As rainfall is low at this time, erosion is less likely and deposition of sediment downstream of the works should be minimised. This also avoids the time when stream fauna are at the most vulnerable stage in their life cycle (eggs and young larvae). Any essential works outside the dry season should be temporarily isolated from the stream to prevent the risk of pollution or sedimentation affecting the ecological integrity of the stream.

7.9.12 As required to minmize potential water quality impacts (Section 5.6), excavation works at the stream section of MUP05 should be restricted to 300m length at any one time. No restriction is considered necessary for the first 300m upstream concrete drains section. Excavation works at MUP04A should be restricted to 100m to cater for potential cumulative impact on MUP05.

7.9.13 Appropriate site management procedures during the construction phase should be adopted, as recommended in ETWB TCW No. 5/2005, to minimise potential disturbance impacts and pollution risks (water quality impacts) to the stream. This should include the location of access to the site and storage of materials, and treatment of construction site waste to prevent pollution of the stream. These site management measures are listed in the subsequent section.

Stream Banks and Riparian Vegetation

7.9.14 Although 20% of the natural stream banks will not be impacted under the revised design, the nature of the works (including widening of the stream channel) limits the extent to which minimization of adverse impacts to the stream banks and riparian zone during the construction stage is feasible. Disturbance impacts to the riparian zone will be minimised by suitable location of works areas and by adoption of appropriate site management protocols during construction. Where possible, native riparian trees which would be impacted during construction works should be transplanted to suitable sites within the project area (see below).

Operational Stage

Streambed, Gabion Banks and Other Areas within the Operational Limits of the Channel

7.9.15 Management and maintenance of the streambed and channel sides should be limited to the minimum required to prevent flooding and ensure safety. Accordingly, the stream should be permitted to find (and adjust) its own low flow channel and natural changes in the deposition of silt, sand and rock should be tolerated except where a specific flooding or safety issue is identified (in accordance with the DSD Technical Circular (2004)).

7.9.16 Environmental considerations for maintenance of the proposed gabion channels (see Section 5.8) should be adopted.

7.9.17 The provision of natural rock and fines in the widened streambed, and the use of stepped gabion banks, will permit recolonisation of the channel by riparian vegetation following completion of the works, thus mitigating for the loss of natural riparian vegetation. Vegetation management within the channel should therefore be restricted to removing obstructions and preventing tree establishment, while the presence of herbaceous vegetation should be tolerated as much as possible. If clearance of herbaceous vegetation is required to prevent obstruction of water flow, where specific flooding or safety issues have been identified, this should not be undertaken during March – August (the main period during which this vegetation would be used as a breeding/nursery area by fauna). Control of invasive plant species, especially the creeper Mikania micrantha, which has previously been found to readily colonise gabion embankments, should be carried out where necessary to permit the establishment of a native floral community.

7.9.18 It is considered that if the above measures to minimise adverse ecological impacts are instituted, all significant ecological impacts which cannot be avoided would be minimised or reduced to acceptable levels with the exception of the loss of the existing diversity of flora and fauna in the riparian zone, resulting primarily from the loss of many riparian trees along the MUP05 stream. This remaining potential adverse impact will require mitigation (see below).

Proposed Measures to Mitigate for Adverse Ecological Impacts

7.9.19 As discussed above, it is considered that the only remaining adverse ecological impact of significance once avoidance and minimisation measures are taken, is the loss of shaded stream sections at MUP05 due to the loss of bankside trees. Shaded stream sections are ecologically important – for example, buffering stream temperatures, providing habitat for a number of fish and dragonfly species of conservation importance (Fellowes et al. 2002) or providing roosting or nesting sites for wetland birds.

7.9.20 The loss of bankside trees, and associated riparian habitats, should be mitigated through transplanting existing trees to suitable locations wherever possible, and through supplemental planting of native trees and bamboos in locations where the project area includes sufficient space adjacent to the stream but outside the channel itself (in addition to retaining in-situ as much trees as possible). Table 7.29 details appropriate species of trees and bamboos. Species selected are characteristic native species present in the area (see Appendix E) and which are known to be of high value to wildlife.

7.9.21 The proposed landscape compensatory planting of about 740 trees (approximately 1,100 m²) along the MUP channels (see Section 8.11 – Landscape and Visual Impact) will serve dual purpose of landscape impact mitigation as well as mitigating the loss of riparian trees.

7.9.22 The Landscape Plan to be submitted prior to commencement of planting or landscaping works should take into account the recommended plant species listed in Table 7.29.

Table 7.29

Recommended Tree and Bamboo Species for Planting at MUP05

Species	Reasons/Benefits
Celtis tetranda (sinensis)	Characteristic Hong Kong riparian species common in the area. Provides good shade and fruit for birds.
Ficus hispida	Characteristic Hong Kong riparian species common in the area. Provides good shade and fruit for birds, bats and civets.
Ficus microcarpa	Characteristic Hong Kong riparian species present in the area. Provides good shade and fruit for birds and bats.
Litsea glutinosa	Widespread small tree species common in the area. Flowers attractive to insects and fruits to birds.
Sapium discolor	Characteristic Hong Kong riparian species common in the area. Limited shade but abundant fruit for birds.
Schefflera octophylla	Widespread lowland species occurring along streams (and elsewhere); common in the area. Form provides good cover, flowers very attractive to insects and fruit to birds.
Trema tomentosa	Common in the area. Supports diverse bird community and produces abundant small fleshy fruit favoured by birds.
Bambusa eutuldoides	Common bamboo species in Hong Kong. Provides habitat for breeding ardeids, especially Chinese Pond Heron.

Proposed Site Management Measures during Construction

7.9.23 The recommended site management measures are generally good site practices and proper water quality control / waste management measures to be implemented by the contractor for all works near stream courses. These measures include:

§ Construction activities should be restricted to works area that should be clearly demarcated.

§ Excavation works should be carried out during the dry season where stream flow is low. Where adequate space is available, works should be carefully phased such that only one side of the channel is constructed. Temporary diversion should be provided to ensure continuous water flow to the downstream section.

§ The proposed works site inside or in the proximity of natural streams should be temporarily isolated, such as using bunds or sandbag barriers (wrapped with geotextile fabric) or other similar techniques, to prevent adverse impacts on the stream water quality.

§ For the stream section where the existing natural stream bed and bank will be left untouched, no disturbance to the stream bed and bank should be allowed from construction works, equipment or workers. If temporary access track on streambed is unavoidable, this should be kept to the minimum width and length. Temporary stream crossings should be supported on stilts above the stream bed.

§ Adequate temporary drainage measures including sediment and oil/grease traps should be provided to prevent contaminated site run-off entering the water bodies.

§ Stockpiling of construction materials, spoils and waste should be properly covered and located away from water bodies to prevent silty runoff and other pollutants from entering the water bodies during rain storms.

§ Construction effluent, site run-off and sewage should be properly collected, treated and disposed.

§ Supervisory staff of the contractor should be assigned to station on site to closely supervise and monitor the construction works. All workers should be regularly briefed to avoid disturbing the flora and fauna near the works area.

7.9.24 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 be reviewed by the Environmental Team Leader.

7.10 Post-mitigation Acceptability of the Project

7.10.1 Tables 7.30 – 7.33 provide details of significant potential ecological impacts of the Project, summarise the proposed mitigation measures and assess post-mitigation acceptability of the project. All impacts which were considered to be of at least moderate scale and/or low to moderate importance are taken into account. Some potential impacts of small scale and/or low importance are covered in the Tables where either mitigation measures to further reduce or eliminate impacts are considered desirable and/or where these can be accommodated without having any adverse impacts on other project objectives.

Table 7.30

Summary of Significant Potential Ecological Impacts, required Mitigation Measures, and Post-mitigation acceptability of the Project at LMH01

Description of impact	Scale/importance of impact without mitigation	Mitigation measures adopted/required	Residual impact
Disturbance to stream fauna of conservation importance.	Small scale and low importance (but can be reduced further).	Restrict stream working to dry season.	Very small construction phase impact, but no operational phase impact, to ecologically important stream bed. No significant impact to stream banks as permanent loss of 160 m of natural stream banks will be compensated by conversion of 200 m of concrete-lined to gabion-lined banks.

Table 7.31

Summary of Significant Potential Ecological Impacts, required Mitigation Measures, and Post-mitigation acceptability of the Project at MUP04A

Description of impact	Scale/importance of impact without mitigation	Mitigation measures adopted/required	Residual impact
Loss of natural stream course (upper section).	Low to moderate scale and importance.	Upper section of stream will not be channelised.	None.
Reduction in extent/seasonal duration of long-inactive wet agricultural land adjacent to upper stream section.	Low to moderate scale and importance.	Upper section of stream will not be channelised so there will be no changes to water table.	None.

Table 7.32

Summary of Significant Potential Ecological Impacts, required Mitigation Measures and Post-mitigation acceptability of the Project at MUP03 & MUP04B

Description of impact	Scale/importance of impact without mitigation	Mitigation measures adopted/required	Residual impact
(No significant ecological impacts)	(No significant ecological impacts)	Revised channel design (stepped gabion sides and gabion bottom will provide stream and riparian habitat.	Minor due to their low ecological value and small net gain in riparian habitat

Table 7.33

Summary of Significant Potential Ecological Impacts, required Mitigation Measures, and Post-mitigation acceptability of the Project at MUP05

Description of impact	Scale/importance of impact without mitigation	Mitigation measures adopted/required	Residual impact
Loss of natural streambed.	Moderate scale and importance during construction; low to moderate during operation (because some recolonisation would take place even if no mitigation).	Existing natural stream bed to be retained; widened stream channel to be floored with natural rocks to simulate natural conditions; works in immediate vicinity of streambed (within 5 m of channel) to be conducted during dry season only.	Very small residual impacts during construction. No significant permanent impacts; potential net gain as wider streambed will provide more stream habitat.
Loss of natural stream banks.	Moderate scale and importance during construction; moderate during operation.	20% of natural banks to be retained; provision of semi-natural conditions in widened two-stage channel (with bottom of natural rock and fines) will permit recolonisation by streamside flora and fauna.	Small residual impacts during construction. Small residual permanent impacts, although the importance of these will reduce as vegetation recolonises widened stream bed.
Loss of riparian vegetation (especially streamside trees).	Low to moderate scale and importance during construction; low to moderate during operation.	20% of natural banks (and associated riparian vegetation) to be retained. Landscape compensatory planting of appropriate streamside trees and shrubs (plus transplanting of existing trees if possible) in suitable locations.	Unavoidable low impact during construction (with due diligence). No significant permanent impacts; potential net gain as planted areas mature and provide increased area of riparian habitat

7.11 Ecological Environmental Monitoring and Audit (EM&A) Requirements

7.11.1 Ecological monitoring and audit is required during the construction and operation stages of the Project in order to ensure that adverse ecological impacts are no greater than predicted, that ecological mitigation measures are implemented satisfactorily and to permit remedial measures to be taken as quickly as possible.

7.11.2 Ecological monitoring is not required for works at MUP03, 04A or 04B where no significant adverse impacts are predicted. However, it is proposed that there be construction phase (water quality and ecological) monitoring at LMH01 and construction (water quality) and post-construction monitoring at MUP05.

Construction Phase Monitoring and Audit at LMH01

7.11.3 Monitoring and audit should cover the following:

· Ensure that excavation and other engineering works do not cause adverse ecological impacts outside the proposed project area;

· Ensure that works do not result in significant siltation downstream of the works area; and

· Ensure that the recommended mitigation measures are properly implemented by the contractor.

7.11.4 Recommendations for the ecological monitoring programme during construction phase at LMH01 are shown in Table 7.34. Details are presented in the EM&A Manual.

Table 7.34

Recommended Ecological Monitoring and Audit Programme at LMH01

Potential Impacts	Monitoring required	Location of Monitoring	Frequency of Monitoring	Duration of Monitoring
Construction phase impacts to stream conditions*
Soil runoff impacts to stream bed	Water quality of stream (dissolved oxygen, suspended solids, turbidity and pH)	Upstream and downstream of construction site	Three times per week	Four weeks before start of construction (for baseline), then throughout construction phase
Construction phase impacts to fish
Impacts to important fish community and other ecological impacts	Surveys of fish species and general site audit (with emphasis on ecological mitigation measures)	Along stream channel, within 100 m upstream and downstream of construction site	Once per week on fish community as a whole and general site audit Monthly fish survey focusing on the species of conservation importance	Three surveys (once per week) before construction (baseline) then throughout construction phase

Notes:

* As part of the water quality monitoring programme (see Chapter 10).

Construction Phase Monitoring and Audit at MUP05

7.11.5 Monitoring and audit should cover the following:

· Ensure that excavation and other engineering works do not cause adverse ecological impacts outside the project area;

· Ensure that those areas where natural banks are to be retained are protected from adverse effects of engineering works, including impacts to riparian vegetation along these banks;

· Ensure that the existing natural stream channel is protected from adverse effects of engineering works, including potential indirect impacts through increased sedimentation;

· Ensure that rocks/fines used to form the bottom of the widened channel have the appropriate physical characteristics to permit re-establishment of semi-natural stream conditions; and

· Ensure that the recommended mitigation measures are properly implemented by the contractor.

Post-construction Monitoring at MUP05

7.11.6 The proposed design for MUP05 channel follows the advice contained in DSD TC No. 2/2004. This is a novel approach to channel widening in Hong Kong and it is considered that it would be appropriate to undertake monitoring in order to determine the success, in ecological terms, of the proposed design. This will provide valuable data which will assist in informing the design process for future projects. Accordingly, baseline monitoring should be conducted prior to the commencement of construction works and post-construction (operational) monitoring should be undertaken to establish the following:

· Stream characteristics at an appropriate interval after the completion of works to include water depth, stream width, nature of substrate and a variety of water quality parameters such as dissolved oxygen, turbidity and pH;

· Vegetation characteristics at an appropriate interval after the completion of works;

· (Re-) colonization of the widened channel by permanently aquatic species (fishes and stream invertebrates); and

· Use of the widened channel by other wetland-dependent taxa (birds, herpetofauna and dragonflies).

7.11.7 Details of the monitoring programme are presented in the EM&A manual and are summarised in Table 7.35.

Table 7.35

Recommended Ecological Monitoring and Audit Programme at MUP05

Potential Impacts	Monitoring required	Location of Monitoring	Frequency of Monitoring	Duration of Monitoring
Construction phase impacts to stream conditions*
Soil runoff impacts to stream bed	Water quality of stream (dissolved oxygen, suspended solids, turbidity and pH)	Upstream and downstream of construction site	Three times per week	Four weeks before start of construction (for baseline), then throughout construction phase
Potential ecological impacts	General site audit (with emphasis on ecological mitigation measures)	Along stream channel, within 100 m upstream and downstream of construction site	Once per week	One walk over survey before start of construction to obtain updated information (baseline) then throughout construction phase
Post-construction ecological impacts
Stream characteristics	Water depth, stream width, nature of substrate and water quality parameters such as pH, DO and turbidity	At an appropriate interval along the constructed channel	To be undertaken along with other post-construction ecological surveys	One year following completion of construction works
Riparian vegetation	Botanical survey	Works area and reinstated area	Once during wet season and once during dry season	One year following completion of construction works
Birds	Transect survey	Within 100 m of construction site	Monthly	One year following completion of construction works
Reptiles	Transect survey (daytime and night-time)	Within 100 m of construction site	One day and one night survey per month during March - October	One year following completion of construction works
Amphibians	Transect survey (daytime and night-time)	Within 100 m of construction site	One day and one night survey per month during March - October	One year following completion of construction works
Fish	Fish survey	Along stream channel, within 100 m upstream and downstream of construction site	Monthly	One year following completion of construction works
Dragonflies	Transect survey	Within 100 m of construction site	Monthly during March - October	One year following completion of construction works
Butterflies	Transect survey	Within 100 m of construction site	Monthly during March - October	One year following completion of construction works
Stream invertebrates	Kick sampling	At upper and lower end of construction site and 100 m downstream of construction site	Monthly	One year following completion of construction works

Notes:

* As part of the water quality monitoring programme (see Chapter 10).

7.12 Conclusions

7.12.1 The proposed works at MUP03 and MUP04B would have no significant ecological impacts even in the absence of mitigation measures because they are existing roadside concrete lined drainage systems.

7.12.2 Proposed works at LMH01 were predicted to have minor construction phase impacts on streambed fauna of conservation importance in the absence of mitigation. However, these can be fully mitigated by restricting works to the dry season between November and March and implementing site management measures. With these measures in place potential impacts are considered to be very small and acceptable.

7.12.3 Proposed works at MUP04A were predicted to have low to moderate permanent adverse impacts on a stream and adjacent area of long-inactive agricultural land. However, all impacts will now be avoided by not channelising the upper section of this stream.

7.12.4 Proposed works at MUP05 were predicted to have moderate adverse impacts on a natural streamcourse with natural bed, banks and riparian corridor. Substantial design changes in terms of ‘two-stage’ channel with natural stream banks and beds retained which are in accordance with the advice contained in ETWB and DSD Technical Circulars and DSD Practice Note have been adopted. These environmental friendly channel designs along with site management measures and tree planting will significantly reduce the predicted impacts to an acceptable level.

7.12.5 With these avoidance, minimization and mitigation measures in place, all adverse ecological impacts of the Project on MUP05 would be eliminated or reduced to a low and acceptable level. Low adverse impacts on riparian vegetation are unavoidable but would be fully mitigated in time as a result of tree planting and recolonisation by herbaceous vegetation, and are considered acceptable in the overall context of the Project.

References

AFCD. 2003. Rare and Precious Plants in Hong Kong. Friends of Country Parks, Hong Kong.

B&V. 2006. Agreement No. CE 6/2002 (DS) Drainage Improvement in Northern New Territories – Package C – Investigation, Design and Construction. Environmental Study (Final Report).

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. & Young, L. 2001. The Avifauna of Hong Kong, Hong Kong Bird Watching Society, Hong Kong.

Chan, B.P.L. & Dudgeon, D. 2001. Conservation Recommendations for Fish Communities of Lowland Streams in Hong Kong. Unpublished report distributed to relevant departments in the Government of the Hong Kong SAR. The University of Hong Kong, Hong Kong.

DSD. 2004. Protection of Natural Rivers and Streams from Adverse Impacts Arising from Construction Works. Drainage Services Department Technical Circular No. 2/2004, Drainage Services Department, Government of the Hong Kong SAR.

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. & Yu, Y.T. 2002. Wild animals to watch: terrestrial and freshwater fauna of conservation concern in Hong Kong. Memoirs of the Hong Kong Natural History Society (25): 123-160.

Kadoorie Farm & Botanic Garden. 2004. A Pilot Biodiversity Study of the eastern Frontier Closed Area and North East New Territories, Hong Kong, June-December 2003. Kadoorie Farm & Botanic Garden Technical Report No. 1. KFBG, Hong Kong.

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

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.

Lee, V.L.F., Lam, S.K.S., Ng, F.K.Y., Chan, T.K.T. & Young, M.L.C. 2004. Field Guide to the Freshwater Fishes of Hong Kong. Cosmos Books Ltd., Hong Kong.

Lo, P. Y. F. & Hui, W-l. 2004. Hong Kong Butterflies. Cosmos Books Ltd., Hong Kong.

Shek, C.T. 2006. A Guide to the Terrestrial Mammals of Hong Kong. Agriculture, Fisheries and Conservation Department, Hong Kong.

Shek, C.T. 2003. Survey of Hong Kong non-flying terrestrial mammals by camera trapping in 2002. Hong Kong Biodiversity 5: 10-11.

Shek, C.T. & Chan, C.S.M. 2005. Roost census of cave dwelling bats of Hong Kong. Hong Kong Biodiversity 10: 1-8.

Wilson, K. D. P. 2003. Field Guide to the Dragonflies of Hong Kong. Agriculture, Fisheries and Conservation Department, Hong Kong.

Xing, F.W., Ng, S.C. & Chui, L.K.C. 2000. Gymnosperms and angiosperms of Hong Kong. Memoirs of the Hong Kong Natural History Society 23: 21-136.

Young, J.J. & Yiu, V. 2002. Butterfly Watching in Hong Kong. Hong Kong Lepidopterists' Society, Hong Kong.

---

[1] Note that Shek and Chan presented numbers of each species as percentages of the total roost count.