8.5 Habitat Quality

8.5.1 Background

8.5.1.1 In this section, the ecological importance of the habitats identified within the Study Area are evaluated in accordance with criteria stipulated in Annex 8 of the EIAO TM.

8.5.2 Secondary Woodland

8.5.2.1 As discussed in the foregoing Section 8.4, secondary woodland patches within the Study Area can be differentiated into three types according to their topographic location. Although there are differences in many ecological aspects of these patches, the ecological value of secondary woodland as a whole is considered high. An assessment of the secondary woodland in accordance with the criteria stated in Annex 8 of the TMEIA is provided in Table 8.16 below.

Table 8.16 Ecological Evaluation of Secondary Woodland Within the Study Area

Criteria	Secondary Woodland
Naturalness	Secondary woodland dominated by native plant species and suffering only limited human disturbance
Size	Comparatively large in size as a whole
Diversity	Rich in terms of floral diversity
Rarity	None of the plant species recorded are rare in HK. Five rare faunal species were recorded in secondary woodland, including one of very high conservation interest and four of moderate conservation interest
Re-creatability	Possible if adequate resources are available (man-power, land, finance, re-planting material) and in the absence of disturbance, but original habitat characteristics including the community composition and structural complexity may require > 20 years to establish
Fragmentation	Patchily distributed in Study Area. Fragmentation of patches is minimal
Ecological linkage	Native secondary woodland may provide a movement corridor for wildlife within the Study Area and it could serve as a seed source to facilitate the succession process in the surrounding area. Functionally linked to streams passing through this habitat
Potential value	High potential for growth and expansion in terms of size and species diversity
Nursery/ breeding ground	The protected endemic Romer’s Tree Frog was found to be breeding in this habitat – notably at Shek Mun Kap and Lung Tseng Tau
Age	Relatively old (>50 years) with respect to the size of the trees and the structural complexity and community composition
Abundance/ Richness of wildlife	High, 106 species of fauna present; 5 species of conservation interest
Ecological value	High

8.5.3 Plantation Woodland

8.5.3.1 Since the plantation woodlands described in the foregoing Section 8.4.9, are significantly different in terms of habitat characteristics, the ecological evaluation will be conducted separately, with the exception of categories pertaining to fauna. According to the Table 8.17 below, it is considered that the mature and semi-mature plantation woodland along the Cheung Sha catchwater, as well as around Pak Kung Au, have a moderate to high ecological value, whereas the ecological value for other plantation woodland is moderate.

Table 8.17 Ecological Evaluation of Plantation Woodland Within the Study Area

Criteria	Plantation A: along Cheung Sha catchwater	Plantation B: on hill-slope around Pak Kung Au	Plantation C: on hillslope on the east of Tung Chung Road
Naturalness	Man-made habitat planted	Man-made habitat planted	Man-made habitat planted
Size	Relatively large in size as a whole	Relatively large in size	Relatively large in size
Diversity	Rich floral diversity	Moderate floral diversity	Botanically poor
Rarity	None of the plant species recorded are rare in HK. Seven faunal species of moderate conservation interest	None of the plant species recorded are rare in HK. Seven faunal species of moderate conservation interest	None of the plant species recorded are rare in HK. Seven faunal species of moderate conservation interest; one of very high conservation interest
Re-creatability	Readily re-creatable provided adequate resources are available (man-power, land, finance, re-planting material)	Readily re-creatable provided adequate resources are available (man-power, land, finance, re-planting material)	Readily re-creatable provided adequate resources are available (man-power, land, finance, re-planting material)
Fragmentation	Moderately fragmented	Not fragmented	Not fragmented
Ecological linkage	Functionally linked to streams passing through this habitat	Functionally linked to streams passing through this habitat	Functionally linked to streams passing through this habitat
Potential value	Relatively high given the maturity of the plantation	Relatively high given the location of the plantation	Relatively high given the location of the plantation
Nursery/ breeding ground	The protected endemic Romer’s Tree Frog was found to be breeding in this habitat	The protected endemic Romer’s Tree Frog was found to be breeding in this habitat	The protected endemic Romer’s Tree Frog was found to be breeding in this habitat
Age	Mature plantation forest	Relatively young plantation forest	Relatively young plantation forest
Abundance/ Richness of wildlife	High, 100 species of fauna recorded, 4 species of conservation interest	High, 100 species of fauna recorded, 3 species of conservation interest	High, 101 species of fauna recorded, 1 species of conservation interest
Ecological value	Moderate to high	Moderate to high	Moderate

8.5.4 Tall Shrubland

8.5.4.1 Since the habitat structure and species diversity of tall shrubland within the Study Area is complex and rich, and has a species composition comparable to secondary woodland, the ecological value of this habitat type is considered moderate to high. An assessment of the tall shrubland in accordance with the criteria stated in Annex 8 of the TMEIA is provided in Table 8.18 below.

Table 8.18 Ecological Evaluation of Tall Shrubland Within the Study Area

Criteria	Tall Shrubland
Naturalness	Natural habitat
Size	Moderate as a whole for the Study Area
Diversity	Botanically diverse
Rarity	None of the floral species recorded was rare. Thirteen species of fauna of conservation interest were present, including two of very high conservation interest, four of high conservation interest and three of moderate conservation interest
Re-creatability	Readily re-creatable provided that adequate resources are available (man-power, land, finance, re-planting material) and in the absence of disturbance
Fragmentation	Fragmentation is moderate with respect to the distribution pattern of the habitat within the Study Area
Ecological linkage	Functionally linked to streams passing through this habitat
Potential value	High given the rich tree flora within the habitat
Nursery/ breeding ground	The protected endemic Romer’s Tree Frog was found to be breeding in this habitat, at Pak Kung Au
Age	Moderate in terms of the succession pathway
Abundance/ Richness of wildlife	High, 113 species of fauna recorded in this habitat, 12 species of conservation interest
Ecological value	Moderate to high

8.5.5 Shrubland-Grassland Mosaic

8.5.5.1 Shrubland-Grassland Mosaic is species rich and structurally complex as a whole for the mosaic, and is considered to have a moderate ecological value (Table 8.19).

Table 8.19 Ecological Evaluation of Shrubland-Grassland Mosaic Within the Study Area

Criteria	Shrubland-Grassland Mosiac
Naturalness	Natural habitat but may suffer frequent disturbance (hill fire)
Size	Large in size as a whole for the Study Area
Diversity	Botanically diverse as a whole for the habitat mosaic
Rarity	None of the floral species recorded was rare. Three species of avifauna of conservation interest were present
Re-creatability	Readily re-creatable provided that adequate resources are available (man-power, land, finance, re-planting material) and in the absence of disturbance
Fragmentation	Fragmentation is relatively limited with respect to the distribution pattern of the habitat within the Study Area
Ecological linkage	Functionally linked to streams passing through this habitat
Potential value	Moderate as a result of the location and the potential fire disturbance
Nursery/ breeding ground	No significant breeding ground or nursery detected
Age	Young in terms of the succession pathway
Abundance/ Richness of wildlife	Moderate, 41 species of fauna recorded in this habitat; 3 species of conservation interest
Ecological value	Moderate

8.5.6 Streams and Riparian Habitat

8.5.6.1 Riparian vegetation in the Study Area is broadly similar to that of the surrounding habitat, comprising secondary woodland, plantation or tall shrubland. However, stream habitats are of particular value due to the fauna present, either utilizing the riparian zone (terrestrial and aquatic species) or within the stream itself. Several streams in the Study Area are seasonal, or of very low base flow, and these are of lower ecological value than the permanent streams with reliable discharge, upon which fully aquatic fauna are dependent. The following evaluation divides streams into two broad categories, those with generally higher base flow supporting species of conservation interest, and those with generally lower base flow not supporting species of conservation interest, as summarised in Table 8.15. An evaluation of stream and riparian habitat is presented below in Table 8.20.

Table 8.20 Ecological Evaluation of Stream and Riparian Habitat Within the Study Area

Criteria	Streams A (relatively higher base flow): TC, CS, 15, 18, 19, 21, 23, 24, 25, 26, 27, 32, 35, 40, 41, 42, 43	Streams B (relatively lower base flow): 1-14, 16, 17, 20, 22, 28-31, 33, 34, 36-39, 44
Naturalness	Natural habitat, pristine at higher elevations, generally more disturbed at lower elevations	Natural habitat, pristine at higher elevations, generally more disturbed at lower elevations
Size	Small in size as a whole for the Study Area; generally exceeding 500m in length from headwaters to coast. Relatively moderate to high base flow	Small in size as a whole for the Study Area; several less than 500m in length due to interception by catchwater. Relatively low to moderate baseflow
Diversity	Botanically diverse as a whole for the habitat mosaic	Botanically diverse as a whole for the habitat mosaic
Rarity	No rare floral species recorded. Numerous rare/endangered species of fauna were present across these streams as a whole, including eight of very high conservation interest, eight of high conservation interest and four of moderate conservation interest	Chinese cobra present at stream 9 although not an aquatic species
Re-creatability	Re-creatable provided that works conducted in ecologically-sensitive manner and original flow not diverted or polluted, and in the absence of disturbance	Readily re-creatable provided that works conducted in ecologically-sensitive manner and original flow not diverted or polluted, and in the absence of disturbance
Fragmentation	Low – moderate. Generally non-fragmented continuous linear habitat	Moderate - severe. In some cases fragmented by catchwater channel or existing road
Ecological linkage	Functionally linked to surrounding terrestrial habitat(s)	Functionally linked to surrounding terrestrial habitat(s)
Potential value	High as a result of the pristine condition and generally low level of disturbance	Moderate as a result of low/seasonal base flow and/or catchwater interception
Nursery/ breeding ground	Significant breeding grounds of numerous species of conservation interest, including Beijiang Thick-lipped Barb, Romer’s Tree Frog, Short-legged Toad, Hong Kong Newt, several odonates	No significant breeding ground or nursery detected
Age	Ancient geomorphological drainage features	Ancient geomorphological drainage features
Abundance/ Richness of wildlife	Very high, 119 species of non-avian fauna recorded in these habitats as a whole, 21 species of conservation interest (although 5 species non-aquatic)	Moderate, 1 species conservation interest (although non-aquatic)
Ecological value	Very high	Moderate-low, given the value of the surrounding habitat (secondary woodland, plantation woodland or tall shrubland)

8.5.7 Grassland

8.5.7.1 Grassland within the Study Area is species poor and hence considered to have low ecological value (Table 8.21).

Table 8.21 Ecological Evaluation of Grassland Within the Study Area

Criteria	Grassland
Naturalness	Natural habitat but maintained as a fire plagioclimax
Size	Relatively large as a whole for the Study Area
Diversity	Low
Rarity	None of the species recorded are rare in Hong Kong
Re-creatability	Readily re-creatable
Fragmentation	Minimal
Ecological linkage	Not functionally linked to any highly valued habitat in close proximity
Potential value	Limited because of the topographic position of the habitat unless succession is permitted to occur
Nursery/ breeding ground	No significant nursery or breeding ground recorded
Age	Young in terms of the succession pathway
Abundance/ Richness of wildlife	Low, only 16 species of fauna present; no species of conservation interest
Ecological value	Low

8.5.8 Freshwater Marsh (Fong Yuen Marsh)

8.5.8.1 A patch of freshwater marsh is located at Fong Yuen at the northern part of the Study Area near Lung Tseng Tau. Given that this habitat type is limited within the Study Area and relatively rare in Hong Kong as a whole (e.g., Dudgeon and Corlett, 1994) and supports an important fish fauna, its ecological value is considered high (Table 8.22).

Table 8.22 Ecological Evaluation of Fong Yuen Freshwater Marsh within the Study Area

Criteria	Fong Yuen Freshwater Marsh
Naturalness	Semi-natural as the habitat is established on abandoned agricultural fields
Size	Relatively small in size
Diversity	Moderate floral diversity for wetland habitat
Rarity	None of the floral species recorded are rare in Hong Kong. Two rare species of fauna are present, including one species of moderate conservation interest and one of very high conservation interest
Re-creatability	Readily re-creatable provided that adequate resources are available (man-power, land, finance, re-planting material) and in the absence of disturbance
Fragmentation	Not fragmented
Ecological linkage	Functionally linked to streams passing through this habitat
Potential value	Moderate potential value with respect to the planned land-use in surrounding area
Nursery/ breeding ground	The marsh supports a highly significant nursery for the globally endangered and highly restricted Ricefish (Oryzias curvinotus), a species of very high conservation interest
Age	Establishment of the marsh presumably does not pre-date establishment of wet agriculture at the site
Abundance/ Richness of wildlife	Moderate, 59 species of fauna recorded at the marsh; 2 species of conservation interest
Ecological value	High due to presence of Ricefish

8.5.9 Wasteland

8.5.9.1 Wasteland is found scattered around the northern and southern ends of the Study Area and covered mainly by weedy plants with low ecological value (see Table 8.23).

Table 8.23 Ecological Evaluation of Wasteland within the Study Area

Criteria	Wasteland
Naturalness	Semi-natural as the habitat is established on heavily disturbed land and inhabited by weedy species
Size	Relatively small in size
Diversity	Botanically poor
Rarity	None of the species recorded are rare in Hong Kong
Re-creatability	Readily re-creatable
Fragmentation	Moderate
Ecological linkage	Not functionally linked to any high value habitat
Potential value	Low potential value with respect to the planned land-use
Nursery/ breeding ground	No significant nursery or breeding ground recorded
Age	Young in terms of the succession pathway
Abundance/ Richness of wildlife	Very low, ten species of fauna present.
Ecological value	Low

8.5.10 Agricultural Land

8.5.10.1 Agricultural lands within the Study Area are small in size and intensively modified and managed by human activities. The ecological value is hence considered low (Table 8.24).

Table 8.24 Ecological Evaluation of Agricultural Land within the Study Area

Criteria	Agricultural Land
Naturalness	Man-made habitat
Size	Small in size
Diversity	Poor in botanical diversity
Rarity	None of the species recorded are rare
Re-creatability	Readily creatable with adequate resources, i.e. land
Fragmentation	Relatively fragmented given the size of the patches
Ecological linkage	Not functionally linked to any highly valued habitat in close proximity
Potential value	Potentially high depending on the agricultural management practices
Nursery/ breeding ground	No significant nursery or breeding ground recorded
Age	No information
Abundance/ Richness of wildlife	Low, 30 species of fauna recorded; no species of conservation interest
Ecological value	Low

8.5.11 Village Area

8.5.11.1 Rural area usually has intensive human activities with limited ecological resources, and hence has limited ecological value, as shown below in Table 8.25.

Table 8.25 Ecological Evaluation of Village Area Within the Study Area

Criteria	Rural Area
Naturalness	Man-made habitat with intensive human activities
Size	Relatively small in size
Diversity	Botanically poor
Rarity	None of the species recorded are rare in Hong Kong
Re-creatability	Readily re-creatable
Fragmentation	Moderate
Ecological linkage	Not functionally linked to any highly valued habitat in close proximity
Potential value	Low potential value with respect to the planned land-use
Nursery/ breeding ground	No significant nursery or breeding ground recorded
Age	No information
Abundance/ Richness of wildlife	Low, only 17 species of fauna recorded; no species of conservation interest
Ecological value	Low

8.5.12 Vegetation

8.5.12.1 The ecological importance of the species identified within the Study Area is evaluated in accordance with the suggested criteria as stated in Annex 8 of the EIAO TM, namely protection status, distribution and rarity.

8.5.12.2 Accordingly, four plant species of ecological interest were recorded and they were all found in the plantation woodland or secondary woodland around Pak Kung Au, including the orchids Acampe rigida and Liparis viridiflora, the shrub Pavetta hongkongensis and the tree Artocarpus hypargyraea.

8.5.12.3 Although the orchids Acampe rigida and Liparis viridiflora, and the shrub Pavetta hongkongensis are considered common in lowland forest or shrubland habitat in Hong Kong (Siu, 2000; Xing et. al., 2000), they are listed under the Forestry Regulations (Ch. 96, Section 3). Artocarpus hypargyreus is a common tree species in lowland forest but is a Category III nationally protected species in China and listed as vulnerable in the China Plant Red Data Book.

8.5.12.4 Some of these plants are located on the development footprint of the proposed road. These include single individuals of Liparis viridiflora, Acampe rigida, Pavetta hongkongensis and Artocarpus hypargyreus (Figures 8.19a-b).

8.5.13 Marine Intertidal Survey (Pui O Wan)

8.5.13.1 The ecological importance of the predominantly boulder-shore intertidal habitat at Pui O Wan was evaluated in accordance with the suggested criteria stated in Annex 8 of the EIAO TM. The habitat present is composed of a natural boulder shore line with intermittent sand deposits and such natural habitats are generally rated higher in the TM. It should, however, be noted that similar habitat is reasonably common and present in similar locations elsewhere in Hong Kong. The drainage channel will also lead to an insignificant loss of shoreline (the drainage channel will incur the loss of less than 5 m of foreshore, including for the stepped channel itself and a maintenance stairway). The faunal diversity was also low and only twelve common species were recorded along the whole coastline surveyed and eleven faunal species were present in the area directly adjacent to the channel (Appendix H). There was a low abundance of species on the shore although the molluscs were the most abundant group recorded. All species present, apart from the clam Donax sp. are found on hard substrates (often including artificial substrates such as seawalls) throughout Hong Kong. The surf clam (Donax sp.) that was recorded in sands deposited between the boulders is also a common species in Hong Kong although it is distributed in coarse sand rather than hard substrates. The boulder-shore is not fragmented and functionally linked to the adjacent marine habitat (particularly the lower shore). The boulder-shore is not, however, anticipated to represent a significant nursery or breeding ground. Based on the aforementioned, the overall ecological value of the shoreline is therefore considered to be low to moderate. As the intertidal habitat is of low to moderate ecological importance, the area lost to construction of the channel is insignificant, only the potential impacts due to discharges (both construction and operational phases are considered further) and these are discussed below in Section 8.7.7.26 that describes the Southern Section drainage.

8.5.14 Overall Evaluation

8.5.14.1 Habitats within the Study Area may be ranked according to their ecological value as indicated in Table 8.26 below, detailed in ascending order. The stream and riparian habitats with high base flow were the most ecologically valuable habitats present, whereas grassland, rural areas, agricultural land and wasteland were of lowest ecological value.

Table 8.26 Summary of the Ecological Value of Habitats within the Study Area (ascending order of importance)

Habitat	Ecological Value
Stream and Riparian (A)	Very High
Secondary woodland	High
Freshwater Marsh	High
Plantation (A)	Moderate – High
Tall Shrub	Moderate – High
Plantation (B)	Moderate – High
Plantation (C)	Moderate
Shrubland/Grassland Mosaic	Moderate
Stream and Riparian (B)	Low-Moderate
Intertidal Shore (Pui O Wan)	Low-Moderate
Grassland	Low
Village Area	Low
Agricultural Land	Low
Wasteland	Low

Note: (A), (B), (C) refers to group as defined in previous tables

8.6 Ecological Impact Assessment Methodology

8.6.1 The objective of the ecological assessment is to predict the direct, indirect, primary and secondary, on‑site and off‑site impacts of the Improvement to Tung Chung Road between Lung Tseng Tau and Cheung Sha. The significance of ecological impacts have been evaluated based on the criteria specified in Table 1, Annex 8 of the TMEIA (TM), as follows:

¨ habitat quality;

¨ species affected;

¨ size/ abundance of habitats affected;

¨ duration of impacts;

¨ reversibility of impacts; and

¨ magnitude of environmental changes.

8.6.2 Impacts are ranked as “minor”, “moderate” or “severe”, although in a few cases, “insignificant” (less than “minor”) or “extremely severe” may also be given. The ranking of a given impact will vary based on the criteria listed above. For example, an impact might be ranked as “minor” if it affected only common species and habitats, or if it affected only small numbers of individuals or small areas, whereas it might be ranked as “severe” if it affected rare species or habitats, large numbers of individuals or large areas. The major factors giving rise to a ranking of “moderate” or “severe” are spelled out in the text as far as possible. As noted in Annex 16 of the TM, a degree of professional judgement is involved in the evaluation of impacts.

8.6.3 Impacts to species or groups assessed as ‘minor’ are predicted to cause a slight, and/or short term reduction in the local population numbers or geographic distribution of a species or group, but the species or groups are predicted to recover from the perturbation with no long-term adverse impacts. Habitat impacts are considered ‘minor’ when no species of conservation or regulatory concern are found, and when the habitat in question was widely distributed locally.

8.6.4 Impacts to species or groups assessed as ‘moderate’ are predicted to cause local reduction of species or group population numbers. The reductions would be long-term, and probably not recoverable, but the species or groups in question are considered widely distributed or common, and abundant on a local, regional, or global scale. Habitat impacts are judged ‘moderate’ when the habitat in question was of limited local or regional distribution or declining in extent, and when the potential for the habitat to support fauna and flora was considered of conservation or regulatory importance.

8.6.5 Impacts to species or groups are assessed as ‘severe’ when they are judged to adversely affect species or groups which are of conservation or regulatory concern locally, regionally, or globally due to scarcity or declining population or distribution trends. Impacts to habitats are considered ‘severe’ when the habitats are found to be limited or declining in geographic distribution, contain plant species of regulatory or conservation concern, or are generally considered by the scientific community to be of local, regional or global importance to the support of wild fauna.

8.6.6 If ecological impacts are found to be significant (i.e., minor to severe) mitigation needs to be carried out in accordance with the TM. Mitigation measures are not required for insignificant impacts. The policy for mitigating significant impacts on habitats and wildlife is to seek to achieve impact avoidance, impact minimisation and impact compensation in that order of priority. Impact avoidance typically consists of modifications to the project design, but may in extreme cases require abandonment of the project (the “no-go” alternative). Impact minimisation includes any means of reducing the scope or severity of a given impact, e.g. through timing of construction works, modification in design, or ecological restoration of disturbed areas following the completion of works. Impact compensation assumes that an irreversible impact will occur upon a given habitat or species and attempts to compensate for it elsewhere, for example, by enhancement or creation of suitable habitat. Compensation may take place on-site or off-site.

8.6.7 Construction of the alignment will result in permanent loss of habitat. There will be an additional temporary loss resulting from the construction of haulage roads, works areas and temporary work sites. The temporary areas will not be permanently occupied and may, therefore, be left to re-establish naturally or be re-instated once the construction works have been completed.

8.6.8 In calculating the permanent and temporary areas of habitat lost, all areas under the permanent alignment, including all slopes and embankments have been regarded as permanent loss. All areas within the site boundary but not forming a permanent part of the road, namely off-line sections of the haul road and the works areas have been assumed to be temporarily lost.

8.7 Construction Phase Ecological Impact Assessment

8.7.1 General

8.7.1.1 Construction phase impacts have been assessed and evaluated against the major impacts typically associated with road projects, namely habitat loss, habitat fragmentation, and disturbance (English Nature, 1994; Reijnen et al., 1995; Treweek, 1999). Construction phase impacts to fauna are related to direct habitat loss and fragmentation of habitat, the latter applicable to off-line alignment options only as the existing Tung Chung Road can be considered as an in-situ permeable barrier to most fauna). Secondary impacts are related to the effect of habitat loss and subsequent reduction of food resources and breeding sites. The proximity of the project to sensitive habitats renders it imperative that proper control is exerted on handling and disposal of any excavated material.

8.7.1.2 A further significant potential impact is due to disturbance and direct habitat loss during road construction. Although the more mobile species present will avoid the disturbed areas, certain amphibian, fish and damselfly species of high or very high conservation interest in the Study Area have restricted mobility and specific habitat requirements. The off-line sections of the road alignment (mainly in the southern part of the Study Area) will generally have higher potential impact, resulting in greater habitat fragmentation and potential for displacement of territorial species and possible reduction in an ability to successfully find habitat for breeding and/ or nesting. Further details on these aspects are discussed below.

Habitat Loss

8.7.1.3 Habitat loss is a potential threat to many species as it may be linked to direct mortality, species displacement, and is most severe in species that have an inability to translocate to suitable habitat elsewhere (Treweek, 1999). Although species-area relationships have been extensively studied (larger areas of habitat generally provide an area for a greater number of species), it is difficult to estimate the minimum area of habitat required for most animals and this is more difficult with highly mobile species such as birds. Where the new alignment passes through high value habitat such as secondary woodland, the road will be as narrow as possible (consistent with functional objectives), so that habitat loss is no more than is absolutely necessary. Also, retaining walls will be incorporated into the road as a means of minimising the extent of cut slopes.

8.7.1.4 The areas of different habitat loss to the project are presented below in Table 8.27. It should be noted that the works areas at Lung Tseng Tau, close to the Tai Tung Shan header tank and along the Cheung Sha catchwater, as shown in Figures 2.2, 2.7 and 2.10, are on cleared disturbed ground and as such no natural habitats will be lost by the use of these areas. As such, temporary losses are largely restricted to the sections of haul road which go off-line.

8.7.1.5 Permanent habitat loss comprises the area within the footprint of the permanent works plus an average 3 metre wide working space around this. Temporary habitat loss is the width of the haul road plus 3 metres working space.

Table 8.27 Habitat Loss Arising from the Project (ha)

Habitat	Study Area (permanent loss)	Study Area (temporary loss)	Existing Slope Remediation*
Secondary Woodland	2.14	-	0.11
Plantation Woodland	3.98	1.01	0.04
Tall shrub	6.24	0.2	0.08
Shrubland – Grassland Mosaic	0.62	0.05	0.04
Grassland	0	-	0
Freshwater Marsh at Fong Yuen	0	-	0
Agricultural Field	0	-	0
Rural Development (Village Area)	0	-	0
Wasteland	0.23	-	0.01
Stream Course	0.01	-	0

*Obsolete sections of existing Tung Chung Road

Habitat Fragmentation

8.7.1.6 Habitat fragmentation (i.e., the breaking down of existing habitat into smaller areas of habitat) is associated with linear construction projects such as roads. Fragmentation of habitats, such as woodlands, is known to lead to the reduction in numbers of many species such as birds (Treweek, 1999). Increased habitat fragmentation caused by large road (motorway) projects has also been shown to reduce the reproductive success of species such as birds that are intolerant to an increase in edge to habitat ratio (Reijnen et al., 1995).

8.7.1.7 Much of the habitat present in the Study Area is already fragmented, however, due to the presence of the existing Tung Chung Road and other existing developments. In addition, in the southern part of the Study Area, it is notable that the existing catchwater road traverses the hillside from east to west.

Disturbance

8.7.1.8 One of the major construction impacts associated with this project is disturbance to fauna present in the Study Area. Disturbance during the construction phase is likely to be associated with noise and movement from construction traffic and the greater presence of human activities on-site. Short-term disturbance can affect the time species have available for feeding, whilst longer term effects can cause a reduction in the use of a particular area for feeding and/ or breeding (Treweek, 1999). Many animal populations can gradually habituate to low levels of disturbance and transient constructional phase impacts are not predicted to be significant to fauna present, provided that all measures are taken to ensure that disturbance is kept to a minimum. Such mitigation measures are discussed in Section 8.9.

8.7.2 Secondary Woodland

Habitat loss

8.7.2.1 Secondary woodland, when covering an area greater than one hectare, is classed as important in Annex 8 of the TM EIAO. Secondary woodland present in the Study Area was also rated highly in habitat evaluations made for the current project (Table 8.15). Although patchily distributed, a total of 193.6 ha of this habitat is present within the Study Area, of which 2.14 ha will be permanently lost as a consequence of road construction, including the cycle track between Shek Mun Kap Road and Lung Tseng Tau, with a further 0.11 ha loss as a result of slope remediation works on the obsolete sections of the existing Tung Chung Road.

8.7.2.2 Loss of secondary woodland on the on-line section (chainage 1.000 to 3.200, Figures 2.2 to 2.5) will be small, amounting to 0.08 ha, including the area lost to the cycle track. The northern off-line section (chainage 3.200 to 4.500, Figures 2.5 to 2.7) will result in minor loss of 0.28 ha of this habitat, while the southern off-line section (chainage 4.500 to end, Figures 2.7 to 2.11) will incur a further moderate loss of 0.81 ha. The bus bay facilities proposed at San Shek Wan in the extreme south-east of the Study Area will engender a further moderate loss of secondary woodland. Note that no rare or protected flora were recorded in this habitat. Also, although four locally rare species of butterfly were recorded in this habitat as presented in Figures 8.18a-f (Gaudy Baron, Pale Cerulean, Common Rose and Large Branded Swift), none of the known foodplants of these species (e.g., Macrosolon cochinchinensis, Scurrula parasitica, Pongamia pinnata, Derris trifoliata, Aristolochia tagala, A. fordiana) were recorded either in this habitat or anywhere else in the Study Area. Habitat loss is, therefore, not anticipated to have impacts on floral species important for these butterfly species. Loss of habitat will inevitably result in a reduction in available nectar sources for adult butterflies. However, this impact is expected to be insignificant as adult butterflies have moderate to high mobility and should be able to utilise other nectar sources in the vicinity, particularly the large area of secondary woodland remaining. Also, no food plants of butterfly species of conservation interest were recorded.

8.7.2.3 A drainage pipeline for use during both the construction and operational phases will be constructed through secondary woodland to discharge at Pui O Wan (Section 6.3.3.25; Figure 6.2j). The pipeline will be 180m long, and will result in a loss of 0.09ha of woodland. No species of conservation importance were recorded in this area of woodland.

8.7.2.4 Habitat loss could affect the endemic Romer’s Tree Frog inhabiting the secondary woodland patch at Lung Tseng Tau (chainage 1.100 to 1.200, Figure 2.1). During the surveys, the Romer’s Tree Frog was not noted in the secondary woodland area to the west of the road affected by the proposed cycle track. However, the species was noted on the edge of the marsh and so it is very possible that it uses this stretch of woodland. However, loss of woodland in these two areas is limited to small fringe sections close the Tung Chung Road and is anticipated to be insignificant in terms of the secondary woodland area as a whole. Notwithstanding, given the conservation importance of the Romer’s Tree Frog, habitat loss is considered to have a minor impact on the endemic Romer’s Tree Frog.

8.7.2.5 Habitat loss is a threat to birds as it may be linked to direct mortality, species displacement, and is most severe in species that have an inability to translocate to suitable habitat elsewhere (Treweek, 1999). Although species-area relationships have been extensively studied, it is difficult to estimate the minimum area of habitat required for most animals and this is more difficult with highly mobile species such as birds. It has been estimated that the minimum woodland habitat required to support temperate bird species in The Netherlands must be larger than 2 or 3 ha otherwise the characteristic typical ‘core’ woodland species are not supported (Treweek, 1999). A minimum of 50 ha area of woodland has been reported to be required to support viable populations of core (characteristic) woodland species. As the areas of secondary woodland in the Study Area are relatively small they are unlikely to provide habitat that is of high enough quality to sustain large numbers of bird species and this was reflected in the low numbers of species present throughout the Study Area. Although there is potential for some loss of higher quality woodland habitat that is utilised by birds, the overall impact of losing the woodlands present in the Study Area on bird populations is insignificant. Only 16 species of birds, none of conservation interest, were recorded in this habitat.

Fragmentation

8.7.2.6 Fragmentation impacts are generally likely to be insignificant, since this habitat is quite patchily distributed within the Study Area, and the road alignment generally appears to impact only the edges of these patches. However, at chainages 3.200 to 3.400 (Tai Tung Shan) and 6.500 to 6.600 (Figures 2.5 to 2.10), as well as at the site of the proposed bus bay facilities, fragmentation impacts may be moderate due to the road alignment bisecting patches of secondary woodland. The endemic Romer’s Tree Frog, which is a species of very high conservation interest, with low to moderate mobility, was not recorded at these specific locations but was present in other patches of secondary woodland within the Study Area. Several bird species were recorded from this habitat, but these animals are highly mobile and unlikely to be significantly impacted, since the development footprint is narrow and easily negotiable for flying animals.

Disturbance

8.7.2.7 Disturbance impacts are temporary and will be more severe for species with lower avoidance ability. Four butterflies of moderate conservation interest (Gaudy Baron, Pale Cerulean, Common Rose and Large Branded Swift) were recorded in secondary woodland that are moderately to highly mobile as adults. However, no evidence of the relatively sessile larvae was noted, and known larval foodplants were not recorded in this habitat. Since there is no evidence of breeding, disturbance impacts on these species are considered to be insignificant.

8.7.2.8 Disturbance to larger mammals and birds during the construction phase is likely to be associated with noise, construction traffic and the greater presence of human activities on-site. Most larger mammals are nocturnal and will suffer only minor impacts, provided that construction activity ceases during the night. Short-term disturbance to birds can affect the time they have available for feeding, whilst longer term effects can cause a reduction in use of a particular area for feeding and/ or breeding (Hill et al., 1997; Treweek, 1999). Although the effects of disturbance on birds are difficult to quantify, many bird species in Hong Kong (and elsewhere) are known to habituate to disturbance and noise (Melville, 2000; quoted in South China Morning Post Wednesday 9 May 2000). It is, therefore, unlikely that bird species present in the Study Area will suffer significant disturbance effects attributable to the road project and, as such, these construction phase disturbance impacts are considered to be insignificant.

8.7.2.9 The endemic Romer’s Tree Frog, a species of very high conservation interest, was recorded from various secondary woodland locations within the Study Area, some of which (e.g., Lung Tseng Tau, chainage 1.100 to 1.200; Shek Mun Kap, near chainage 1.700 to 1.800, see Figure 2.2) will be potentially disturbed by construction activities such as earth-moving works associated with the project. Locations were the Romer’s Tree Frog was recorded during the surveys is summarised in Figures 8.16a-g. Tadpoles of this species are relatively sessile, while adults are small, have a fairly narrow habitat tolerance and limited dispersal ability. Disturbance impacts to this species are considered moderate.

8.7.3 Plantation Woodland

Habitat loss

8.7.3.1 Plantation woodland occurs primarily in two large areas within the Study Area. One such area at Pak Kung Au (chainage 3.700 to 5.200) has been evaluated as of moderate to high value; while the plantation woodland at the Cheung Sha catchwater (chainage 5.500 to 6.500) has been rated moderate to high in value on the basis of the understorey structure and composition. The plantation woodland on the hillslope to the east of the Tung Chung Road is of moderate value. Overall, the Study Area contains 81.4 ha of this habitat, of which 3.98 ha will be permanently lost as a consequence of road construction, with an additional temporary loss of 1.01 ha for haulage roads. However, this is a re-creatable man-made habitat, containing several naturalised exotics, and impacts from habitat loss are considered insignificant.

8.7.3.2 Note that individuals of four floral species of moderate conservation interest (the tree Artocarpus hypargyreus, the shrub Pavetta hongkongensis and the orchids Acampe rigida and Liparis viridiflora) were recorded in this habitat. An individual of Artocarpus hypargyreus was located in very close proximity to the proposed road (at chainage 5.080, Figure 2.7, near Stream 26), while single individuals of Pavetta hongkongensis and Liparis viridiflora were located close together on the development footprint (chainage 4.380, Figure 2.6, near Stream 22). Loss of these individual specimens is considered a minor impact. All of these species are, however, common in Hong Kong. Also, four locally rare species of butterfly were recorded in this habitat (Black-veined Sergeant, Commander, Pale Cerulean, and Small Grass Yellow), although none of the known foodplants of these species (e.g., Fraxinus retusa, Osmanthus fragrans, Uncaria hirsute, Wendlandia sp., Pongamia pinnata, Derris trifoliata, Chamaecrista mimosoides) were recorded either in this habitat or anywhere else in the Study Area. Therefore, loss of plantation woodland is anticipated to have insignificant impact on floral species important for butterflies of conservation interest.

Fragmentation

8.7.3.3 Fragmentation impacts are generally evaluated as being insignificant, although the proposed road bisects two large areas of plantation woodland. One individual of Romer’s Tree Frog was encountered in plantation woodland adjacent to the existing Tung Chung Road at 190m, close to the intersection with stream No. 27. This species has low to moderate mobility and is of very high conservation interest, but was rarer in this habitat than in secondary woodland, shrubland or riparian habitats within the Study Area.

8.7.3.4 The Mock Viper, which is of moderate conservation interest, was also recorded in plantation woodland (one individual within the patch between chainage 5.500 and 6.500, Figures 2.9 and 2.10). This species has only moderate mobility and habitat fragmentation may impact adversely on availability of suitable foraging areas.

8.7.3.5 Several bird species were recorded from this habitat (various locations throughout the study area) including Greater and Lesser Coucal’s (moderate conservation interest), but these species are highly mobile and unlikely to be adversely impacted, since the development footprint is narrow and easily negotiable for flying animals.

Disturbance

8.7.3.6 Disturbance impacts are temporary and will be more severe for species with lower avoidance ability. Four butterflies of moderate conservation interest (Black-veined Sergeant, Commander, Pale Cerulean, and Small Grass Yellow) were recorded in plantation woodland as moderately to highly mobile adults (Figure 8.18a-f). However, no evidence of the relatively sessile larvae was noted, and known larval foodplants were not recorded in this habitat. Since there is no evidence of breeding, disturbance impacts on these species are considered to be insignificant.

8.7.4 Tall Shrubland

Habitat loss

8.7.4.1 Tall shrubland occurs extensively throughout the Study Area, occupying a total area of 131.7 ha, of which 6.2 ha will be permanently lost due to road construction, with temporary loss of 0.2 ha arising from construction of haulage roads. This habitat was rated as having moderate to high ecological value in the Study Area. Some 3.07 ha of this habitat will be lost along the online section (chainage 1.000 to 3.200, Figures 2.2 to 2.5), 1.52 ha in the northern off-line section (chainage 3.200 to 4.500, Figures 2.5 to 2.7) and a further 1.84 ha in the southern offline section (chainage 4.500 to end, Figures 2.7 to 2.11). In addition, a small area of this habitat (0.007 ha) will be lost at the bus bay facilities at San Shek Wan and 0.08ha to the slope remediation works. Due to the extensiveness of this habitat within the Study Area, however, loss of 6.52 ha is considered as being a minor to moderate impact.

8.7.4.2 No rare or protected flora were recorded along the development footprint. Three locally rare butterfly species of moderate conservation interest (Common Rose, Pale Cerulean, Small Grass Yellow) were recorded in this habitat (Table 8.14; Figures 8.18a-f). However, larval foodplants of these species (Aristolochia tagala, A. fordiana, Pongamia pinnata, Derris trifoliata, Chamaecrista mimosoides) were absent. Therefore, habitat loss is not expected to impact on floral species of conservation significance to butterflies.

8.7.4.3 Three bird species of moderate (Greater and Lesser Coucals) and high (Grey Nightjar) conservation interest were recorded within this habitat. All of these species are highly mobile and impacts are considered insignificant.

8.7.4.4 Romer’s Tree Frog was recorded breeding at several locations in this habitat, particularly on the northern side of the Study Area at Lung Tseng Tau to Pak Kung Au (chainage 1.000 to 4.500) (Table 8.14; Figures 8.16a-g). This species has low mobility and impact of habitat loss is considered as minor to moderate, depending upon whether breeding sites (small seepages) are destroyed. Three dragonfly species of conservation interest, the endemic Melligomphus moluami, the endemic subspecies Leptogomphus elegans hongkongensis and Zygonyx iris insignis were recorded foraging in this habitat (Table 8.14; Figures 8.18a-f). These species are more associated with riparian habitat and loss of tall shrubland habitat at the scale engendered by the proposed road development is considered an insignificant impact.

Fragmentation

8.7.4.5 Tall shrubland habitat is already highly fragmented in the Study Area. In the northern section where losses will be greatest the habitat is fragmented by the existing Tung Chung Road and further fragmentation arising from the proposed road development is considered to be insignificant.

Disturbance

8.7.4.6 Three butterflies of moderate conservation interest (Pale Cerulean, Common Rose and Small Grass Yellow) were recorded in tall shrubland (Table 8.14; Figures 8.18a-f). However, no evidence of the relatively sessile larvae was noted, and known larval foodplants were not recorded in this habitat. Since there is no evidence of breeding, disturbance impacts on these species are considered to be insignificant. Such impacts are also considered to be insignificant for the three dragonfly species of conservation interest (Melligomphus moluami, Leptogomphus elegans hongkongensis and Zygonyx iris insignis) all of which are moderately or highly mobile as adults and do not breed in tall shrubland.

8.7.4.7 Ferret Badger and Chinese Cobra, both recorded from tall shrubland within the Study Area (Table 8.14; Figures 8.17a-b and 8.16a-g respectively), will potentially be impacted by disturbance, since construction works will be continuous along a long front. Although these relatively mobile species will be able to avoid areas of high disturbance, the nature of the development will present a barrier to movement for these species during times of construction activity, effectively reducing available foraging areas. However, the Ferret Badger is nocturnal, and will not be active during times of construction activity and in addition, much undisturbed habitat will be present, even during times of construction activity. It is therefore anticipated that disturbance impacts on these two species of conservation interest will be insignificant.

8.7.5 Shrubland-grassland mosaic

Habitat loss

8.7.5.1 This habitat accounts for a total area of 115.3 ha within the Study Area. An area of 0.62 ha of this habitat, rated as of moderate ecological value, will be permanently lost as a consequence of the new road development. Temporary habitat loss arising from construction of haulage roads will amount to 0.05 ha with a further 0.39ha as a result of the slope remediation works. It should be noted that the moderate ecological value rating is made to some extent on the basis of the combined faunal assessment with the tall shrubland habitat (for reasons explained in Section 8.4 above). The Romer’s Tree Frog was not, however, recorded in shrubland-grassland mosaic.

8.7.5.2 No rare or protected flora were recorded in this habitat along the development footprint. Therefore, habitat loss is not expected to impact on floral species of conservation significance.

8.7.5.3 Three bird species of moderate (Greater and Lesser Coucals) and high (Grey Nightjar) conservation interest were recorded within this habitat (Table 8.13; Figure 8.15). Although of some conservation interest both of the coucal species are common and widespread in the territory. All three of these species are highly mobile and able to disperse into nearby areas of similar habitat. Impacts of habitat loss are therefore, considered, insignificant.

Fragmentation

8.7.5.4 Shrubland-grassland mosaic habitat is already fragmented in the Study Area. Given the fact that the only species of conservation interest recorded in this habitat were birds, with very high mobility, the further, small-scale fragmentation which will be engendered by construction of the new road is considered to be insignificant in terms of impacts on fauna.

Disturbance

8.7.5.5 Three birds of moderate to high conservation interest (Greater Coucal, Lesser Coucal and Grey Nightjar) were recorded in shrubland-grassland mosaic. However, since these species have high dispersal ability and are readily able to avoid disturbed areas, disturbance impacts of the new road are considered to be insignificant. It should also be noted that traffic volumes will be low, and that Grey Nightjar is nocturnal and will therefore be active at times when traffic flow is lowest.

8.7.6 Grassland

8.7.6.1 This habitat accounts for a total area of 43.5 ha within the Study Area. However, the proposed road development will not encroach on this habitat. In view of this, and the fact that grassland in the Study Area was rated as being of low ecological value, any impacts to this habitat, and to the species occupying it, arising from the project are considered to be insignificant. No species of conservation interest were recorded in this habitat.

8.7.7 Stream and riparian A (relatively high base flow): Tung Chung Stream, Cheung Sha Stream, Stream No.’s 15, 18, 19, 21, 23, 24, 25, 26, 27, 32, 35, 40, 41, 42, 43

General

8.7.7.1 Based upon both wet and dry season field surveys, numerous streams in the Study Area are apparently seasonal, or of very low base flow, some influenced by the water gathering grounds and catchwater system, and the surveys have shown these streams to be of lower ecological value than the permanent streams with reliable discharge, upon which fully aquatic fauna are dependent.

8.7.7.2 Based upon the surveys, the streams can be divided into two broad categories comprising those with generally higher base flow supporting species of conservation interest, and those with generally lower base flow not supporting species of conservation interest. A summary of the quality of the two sets of streams is provided in Table 8.20 and a summary of the flow details of each stream is provided in Table 8.15.

Habitat loss

8.7.7.3 It has been necessary to provide a balance between the provision of elevated structures which span the streams and avoid direct impacts and the visual impacts of these structures. However, based upon the fact that loss of stream bed in streams of high ecological value, and its replacement with smooth-walled drainage channels, can incur loss of refugia, foraging and predator avoidance opportunities for aquatic fauna such as fish, amphibians (particularly tadpoles) and odonate larvae, development of the project has sought to avoid impacts on key streams. Thus, new bridge structures have been proposed for as many of the streams crossed by the new road as possible, with emphasis on the streams which has been confirmed by the wet and dry season ecological surveys to be of high ecological value, as discussed in this section. For all but four of these streams of high ecological value, namely Streams 15, 18, 19 and 21 (Figure 8.5), stream modifications, and thus direct impacts and habitat losses, have been avoided, with the new road passing over the water courses on bridge structures. Thus bridges will be provided for streams 23, 24, 25, 26, 27, 32, 35, 40 and 43 for ecological reasons as summarised in Table 8.15. Streams 39, 41 and 42 are not crossed by the road and will not therefore be impacted by the project.

8.7.7.4 The bridge piers/abutments will be constructed at least 5m away from the streams and excavation works for the columns will be protected by the drainage works described in more detail below and in Section 6, in order to protected from any runoff entering the streams.

8.7.7.5 The haul road will generally follow the alignment of the new road, but will bypass the construction site at some stream courses where longer bridges are proposed. However, in order to avoid affecting these streams, the haulage road will be designed to pass over the stream courses on temporary steel bridges of at least twice the width of the stream thus, protecting against direct impacts to the streams and the riparian vegetation.

8.7.7.6 Tung Chung Stream and Cheung Sha Stream are not crossed at any point by the proposed road alignment and thus will not be directly affected by the scheme.

8.7.7.7 Thus, habitat loss on key streams will be restricted to Streams 15 (extension of existing bridge resulting in about 6m loss of stream bed), 18 (widening of existing culvert by 20m), 19 and 21 (30m and 50m culverts, respectively, to be constructed at road crossing). The culverting works on streams 18, 19 and 21 will lead to a loss of the existing stream bed as the stream is modified in such a way that the flow is altered and existing streambed will dry out. The modifications to Streams 15, 18, 19 and 21 cannot be avoided, as discussed in Sections 6.3.4.5 to 6.3.4.9.

8.7.7.8 Impacts may be incurred in the streams as a result of the destruction of the natural stream bed, with potential deleterious impacts (crushing, displacement, entrapment) on sedentary aquatic fauna such as tadpoles of rare amphibian taxa. This may be compounded by destruction of riparian vegetation with deleterious impact on rare amphibians (e.g., Romer's Tree Frog, Short-legged Toad, Lesser Spiny Frog) for which riparian vegetation is a highly important habitat, particularly during the breeding season, which lasts from March to August but peaks during April to June.

8.7.7.9 The impact of habitat loss is minor in stream 15 as even though only about 6m of stream bed will be lost, which is negligible in relation to the entire stream length, the Hong Kong Newt is present. Also based upon, the number of aquatic species of conservation interest present (various amphibians, see Table 8.14) and also the size of habitat lost (20-50m) for streams 18, 19 and 21, the impacts are predicted to be also minor.

Fragmentation

8.7.7.10 Fragmentation is not expected to be an impact for streams over which an elevated structure is placed (stream No.’s 23, 24, 25, 26, 27, 32, 35, 40, 41, 42, 43). Fragmentation may, however, be an impact in the newly culverted streams (19 and 21) and further fragmentation may occur through the extension of the culvert in Stream 18 by 20m in both construction and operational phases if culvert design restricts the access of aquatic fauna such as fish, tadpoles and odonate larvae. Stream 18 supports the Lesser Spiny Frog (high conservation interest), stream 19 supports the Hong Kong Newt (very high conservation interest) and stream 21 supports both the Lesser Spiny Frog and Romer’s Tree Frog (very high conservation interest). However, these fauna were recorded at low densities in these particular streams (see Appendix H) and generally have greater mobility than fish, with adult stages capable of movement in the terrestrial riparian zone, and no fish of conservation interest were recorded in streams 18, 19 and 21.

8.7.7.11 At stream 15, a small population of the very high conservation interest Beijiang Thick-lipped Barb is isolated between the existing bridge culvert and a sheer natural rock face immediately upstream. The protected Hong Kong Newt was also present in Stream 15 although further fragmentation is not considered to be significant impact to the species as it has higher mobility than the Beijiang Thick-lipped Barb. The bridge widening works in Stream 15 will comprise digging a trench for laying of the permanent drainage pipeline and then constructing a concrete base slab as the foundation for the bridge extension. The base slab will take up some 6 metres of streambed. The works proposed at this stream are unlikely to impact on this small population, which is already isolated by the existing topography. This additional loss of streambed is not considered to increase fragmentation of stream 15 significantly. Owing to the conservation status of the Beijiang Thick-lipped Barb it is, however, recommended that, to enhance the population in the catchment (the species is also recorded in the larger Tung Chung Stream of which stream 15 is a tributary) this isolated population be evacuated and released downstream of the bridge, as detailed in Section 8.9.

8.7.7.12 Stream 18 is already culverted under the existing Tung Chung Road, however, this culvert will be lengthened by 20m to accommodate the new road directly adjacent to it. In light of the presence of species of conservation interest, fragmentation impacts to stream 18 are, therefore, considered to be minor. The new culverts to be constructed for Streams 19 (a 30m culvert) and 21 (a 50m culvert) will also cause fragmentation impacts on aquatic fauna of conservation interest present that are anticipated to be minor. It is therefore recommended that the entrance and exits to the culverts are designed in a way so as to not create a barrier to fish or other aquatic fauna such as tadpoles. It is also proposed that the new culverts for Streams 19 and 21 be designed in such a away as to create pools along their length which will aid fish and faunal transfer (see Section 8.9).

Disturbance

Direct Impacts on Streams

8.7.7.13 Unlike the other streams deemed to be of high ecological value which are crossed on bridge, Streams 15, 18, 19 and 21 will be subject to works in the streams themselves and impacts are likely to be linked to some siltation of the water course during construction. Siltation of the water body can clog the gills of fish, tadpoles and odonate larvae, causing death by asphyxiation. Finer silt particles may also clog the feeding apparatus of these organisms, reducing foraging efficiency. Increased water turbidity arising from siltation also impedes foraging success, particularly for predaceous taxa (such as odonate larvae, tadpoles and some fish). These taxa may be further impacted by die-off of prey items associated with increased siltation. Streams 18, 19 and 21 currently support highly sensitive species of amphibian of high or very high conservation interest and steam 15 supports the Beijiang thick-lipped Barb and Hong Kong Newt, also of high conservation interest (Section 8.4).

8.7.7.14 In respect of streams 18, 19 and 21, it is proposed to construct the culvert section prior to its connection with the stream, thus avoiding any direct impacts on the stream course during this time. Thus, the potential for high suspended solids will be limited to the period when the culvert is connected up at either end and stream flow is directed through the culvert. This process will comprise damming the original path of flow, which is not expected to generate large amounts of suspended solids. Also the activity is short term taking a matter of days only. Coarse material will settle out rapidly, probably within the culvert itself and finer material in the water column will be progressively more diluted as it flows down stream. It should be noted that spate (flood) events that commonly occur during the wet season also lead to high suspended particles in the water column and it is likely that the organisms present in streams in the study area are adapted to fine silt particles. The overall ecological impacts associated with disturbance of stream bed sediments in streams 15, 18, 19 and 21 are, therefore considered to be insignificant.

8.7.7.15 As noted above, the bridge widening works in Stream 15 will comprise digging a trench for laying of the permanent drainage pipeline and then constructing a concrete base slab as the foundation for the bridge extension. The base slab will take up some 6 metres of streambed. However, prior to these works commencing, whatever flow is in the stream will be ponded by building a dam across the stream, upstream from the works. A temporary pipeline will then divert the stream water around the works area into the unaffected downstream area. Once the base slab has been completed, the dam will be removed and the stream will flow across this to the existing bridge base slab. The works will be relatively short term, taking approximately 4 weeks to construct the trench and a similar time period for the base slab. However, as all natural stream flow will be diverted, the potential for suspended solids migrating downstream is minimal. The potential is further reduced by the works being undertaken in the dry season. It should be noted that spate (flood) events that commonly occur during the wet season also lead to high suspended particles in the water column and it is likely that the organisms present in streams in the study area are adapted to silt particles. The overall ecological impacts associated with disturbance of stream bed sediments in streams 15, 18, 19 and 21 are, therefore considered to be insignificant.

Construction Run-off

8.7.7.16 Unpolluted stream courses are highly sensitive habitats. While the majority of the streams are crossed on structure, as noted above, and only four of the streams of high ecological value will be subject to works within the stream courses themselves, all the streams are vulnerable to indirect impacts associated with construction site run-off. Siltation may not only impact the streams crossed by the road, but also the Tung Chung Stream since these streams are tributaries of the latter. The Tung Chung Stream is of very high ecological significance, supporting important fish fauna.

8.7.7.17 However, a comprehensive drainage system has been proposed for both the northern and southern sections of the new road. Full details of the system are provided in Section 6. In both cases the temporary drainage system will be in place before any major roadworks such that all construction run-off is collected and treated prior to discharge.

Northern Section Drainage

8.7.7.18 In the northern section, part of the permanent drainage system will be built in advance and a special carrier pipe/drainage channel will collect all run-off from the northern road section from the crest and transport it to a discharge outfall in the “Wong Lung Hang Channel” nullah near Ha Ling Pei. In this way, no construction run-off will be discharged into any streams crossed by the northern section of the alignment and thus impacts on Tung Chung Stream will also be avoided.

8.7.7.19 Key elements of the drainage system to improve its effectiveness include:

¨ construction of the off-line section of the road may only commence when either the pipeline trench or carrier pipe has been put in place up to Tai Tung Shan Reservoir;

¨ during excavation work for the drainage system itself, each stretch of dug channel will be concrete lined before the end of each day to minimised run-off. Control and proper disposal of excavated material in accordance with the Waste Management Plan shall be undertaken;

¨ all temporary drainage channels shall be lined with concrete to prevent erosion;

¨ the new bridge piers/abutments will be constructed at least 5m away from the streams and excavation works for the columns will be protected by the drainage works; and

¨ hinterland drainage shall be collected by the use of perimeter channels and diverted to the nearest stream course to prevent external run off from crossing the site in order to prevent it becoming contaminated and to minimise the amount of polluted runoff to be controlled.

8.7.7.20 Other measures which are proposed during the construction phase to minimise run-off and the effects on streams include:

¨ temporary cut slopes shall be covered with tarpaulin during rainfall when not being worked to prevent erosion;

¨ major stockpiles shall be sited outside the Country Park and away from stream courses;

¨ spoil heaps shall be covered at all times to minimise losses in the wet season and reduce the dust emissions during the dry season to minimise impacts on air quality;

¨ stock piled materials shall be contained in a designated area down gradient from any stream or up gradient with a suitably constructed barrier to reduce loss of materials to the stream; and

¨ no oil or fuel shall be stored within the Country Park or the water gathering grounds.

8.7.7.21 The construction of the haul road will also involve some earthworks and in order to reduce the areas of exposed earthworks during this phase, the contractor will be required to pave the off-line section of haul road and associated slopes within 3 days of the completion of individual sections of the off-line haul road. The temporary drainage system would be constructed concurrently to the formation of the haul road but the contractor would be required to progress the drainage system in advance of the haul road formation to ensure proper runoff control before the road is paved.

8.7.7.22 Based upon the incorporation of the measures outlined above, impacts on the northern streams (including the ecologically rich Tung Chung Stream) due to sediment run-off are considered to be insignificant.

8.7.7.23 High suspended solid loads draining into the Tung Chung Stream would be discharged into the Tung Chung Bay where the San Tau SSSI is located. This protected location contains seagrass beds (Halophila ovata and Zostera japonica) which could potentially be impacted (reduced photosynthesis) by high suspended solids in the water column. A sedimentation tank is, however, also proposed during the construction phase for the treatment of all discharge into the nullah in order to protect the Tung Chung Bay from elevated suspended solids. Based upon the calculation in Section 6.3.3.15, any construction run-off into Tung Chung Bay could be diluted up to 40 times, even without the further dilution effects of the marine water and other catchments on the west valley. It should be noted that the seagrass bed at San Tau SSSI has been subject to impacts associated with the reclamation works for the new airport at Chek Lap Kok and the North Lantau Highway. The seagrass successfully recovered from this disturbance once the works were completed. In addition, in view of the on-going and planned reclamation works in northern Tung Chung, the additional impact of any sedimentation arising from the current project should be limited.

8.7.7.24 Based upon this and the fact that the run-off will be treated before discharge, ecological impacts on Tung Chung Bay and ecological communities present including seagrass beds are predicted to be insignificant.

Southern Section Drainage

8.7.7.25 In the southern section, since any drainage system will need to be carried across the new elevated structures along this section of the route, it will generally not be possible to construct the system ahead of the main roadworks as is proposed in the northern section of the alignment. Thus, the southern alignment is divided into small catchments in order to enhance the control of run-off and a temporary drainage system provided whereby the hinterland drainage is collected separately, as for the northern section, and the run-off is treated via a sedimentation tank prior to discharge into the adjacent streams. Discharge into the streams which ultimately run into the catchwater are also treated via an oil interceptor.

8.7.7.26 Run-off from the works site at the new junction with South Lantau Road will also be treated via a temporary sedimentation tank prior to discharge into the proposed carrier pipe discharging on the foreshore of Pui O Wan (Figure 6.2j). This part of the permanent drainage system will be constructed in advance in order for it to be used during the construction phase. As discussed in Section 8.4.11, there are no species present on the shore at Pui O Wan that are of conservation interest and the boulder shoreline has an overall low to moderate ecological value. The drainage channel will lead to an insignificant loss of some predominantly hard-substrate (mostly boulder) shoreline as the drainage channel will incur the loss of less than 5 m of foreshore, including for the stepped channel itself and a maintenance stairway. Construction-phase impacts including runoff is, therefore considered to represent an insignificant ecological impact to the intertidal species present in Pui O Wan as run-off will be treated via a sedimentation tank. Significant concentrations of contaminants from road runoff are not predicted during the operational-phase (Section 6.4.1). As discussed above for the construction phase, the operational-phase impacts to intertidal organisms attributable to discharges of road runoff are also considered to be insignificant.

8.7.7.27 While discharge into the local streams, catchwater and sea at Pui O Wan cannot be avoided, the proposed construction drainage system minimises impacts by dividing the areas of discharge into small units, thereby reducing the amount of discharge into any one stream and also by treating the run-off prior to discharge. The control measures highlighted above in Section 8.7.7.19 and 8.7.7.20 are also applicable to the southern section.

8.7.7.28 Based on the measures outlined above, impacts on the streams in the southern study area including the Cheung Sha Stream due to run-off and sedimentation are considered to be insignificant.

8.7.7.29 In view of the foregoing, disturbance impacts are rated as minor for streams 19 and 21 and insignificant for streams 15 and 18. Stream No.’s 23, 24, 25, 26, 27, 32, 35, 40 (Cheung Sha Sheung Tsuen Stream) and 43 will be crossed by elevated structures, with bridge supports at least 5m away from the stream bed. Disturbance impacts on these streams will be insignificant. Stream No.’s 39 and 41 and 42 are not crossed by the road alignment. All streams will be protected by the proposed drainage system and impacts are predicted to be insignificant.

8.7.7.30 Based upon the importance of the Tung Chung Stream, Cheung Sha Stream and Cheung Sha Sheung Tsuen stream which have been shown to contain fish species of conservation importance, a summary of the potential impacts is provided in Section 8.7.8 to 8.7.10 below.

8.7.8 Tung Chung Steam

8.7.8.1 A drainage system will be constructed that will lead to no discharges into the Tung Chung Stream thereby protecting the valuable fauna present. The existing road widening will also be to the east of the stream thereby allowing the existing road to act as a buffer between construction works and the watercourse. No ecological impacts are therefore predicted for the Tung Chung Stream.

8.7.9 Cheung Sha Stream

8.7.9.1 The selected alignment was chosen in some respects due to the distance away from the ecologically important Cheung Sha Stream thereby minimising the potential for impacts (only two small tributaries Stream No.’s 25 and 26 are crossed by the road). The drainage system for construction phase activity will treat run-off (sedimentation tank) prior to discharge into streams and no ecological impacts are thereby anticipated. No ecological impacts are therefore predicted for the Cheung Sha Stream.

8.7.10 Cheung Sha Sheung Tsuen Stream (Stream 40)

8.7.10.1 This ecologically important stream (contains species of conservation interest including Philippine Neon Goby and Black-headed Thick-lipped Goby) will be avoided as it is crossed by a bridge. The drainage system for construction phase activity will as for other southern area streams treat run-off (sedimentation tank) prior to discharge and no ecological impacts are thereby anticipated. No ecological impacts are therefore predicted for the Cheung Sha Sheung Tsuen Stream.

8.7.11 Stream and riparian B (relatively low base flow): stream No.’s 1-14, 16, 17, 20, 22, 28-31, 33, 34, 36-39, 44

Habitat loss

8.7.11.1 In order to provide new bridges as far as possible, in addition to the bridges provided for the streams of high ecological importance detailed above, Streams 28, 29, 31, 33, 34, 36, 37, 38, 40 and 41 will also be crossed on structure and notwithstanding their potentially lower ecological value, will not be subject to direct impacts.

8.7.11.2 For steams with existing bridges or culverts, it will be necessary to undertake bridge widening or culvert widening and these works cannot be avoided. However, the streams which require these works are on the whole streams classed as having low ecological value, with the exception of streams 15 and 18 as discussed in Section 8.7.7 above. In addition, in accordance with the proposal to protect streams of high ecological value as far as possible, but balance the visual impacts of bridge structures, new culverts have been proposed for a number of streams but these works have been restricted to streams of low ecological value, with the exception of streams 19 and 21 where it is unavoidable, as discussed above.

8.7.11.3 Streams 10, 11, 12 and 14 will have existing bridges widened, while existing culverts at streams 1-9, 13 and 17 will be extended. The remaining streams will be placed in new culverts. The details of these works is summarised above in Section 8.7.7 and further details provided in Section 6. Overall, less than 0.01ha of stream course will be lost as a result of these works. These streams have been rated as being of moderate conservation value, being characterised by low discharge which does not support any fauna of conservation interest. Also, habitat loss per water course is small, and is not expected to be a significant impact.

Fragmentation

8.7.11.4 Most of these streams are already fragmented to some degree, either by water capture for the Cheung Sha catchwater or by diversion or culverting at the junction with the existing Tung Chung Road. Further fragmentation of these streams is not expected to cause significant impact, since no species of conservation concern have been recorded.

Disturbance

8.7.11.5 Disturbance to the streams is associated with siltation as a result of works within the stream course and in-direct impacts associated with construction run-off. New culverts will be required for streams crossed by the off-line section of the new road (streams 20 and 30) as presented in Table 8.15. In all cases, the culverts will be built separately in advance of any connection with the existing stream, as described for Streams 18, 19 and 21 above. Impacts associated with suspended solids, again, will be limited to the period when the flow is diverted. As the streams being culverted already have a very low base flow and in many cases are dry for much of the year, the opportunity for suspended solids to be transported downstream is negligible. Notwithstanding, any release of suspended solids would be expected to be localised as the material settles out, as the period of works is of short duration, and as the streams do not contain any species of conservation interest, impacts on the ecology are considered to be insignificant.

8.7.11.6 There are four existing small bridges (STR01 to 04 on the figures in Section 2) along the on-line section of the roadworks, crossing streams 10, 11, 12, 14, in addition to STR05 crossing stream 15 as described above, which need to be widened. The details of the widening works are described in Section 8.7.7.15. As discussed, prior to these works commencing, whatever flow is in the stream will be ponded by building a dam across the stream, upstream from the works. A temporary pipeline will then divert the stream water around the works area into the unaffected downstream area. Based upon this and the absence of species of conservation interest in these streams, ecological impacts are considered to be insignificant.

8.7.11.7 Streams 1-9, 13 and 17 will require existing culverts to be extended. As with the requirement for new culverts, the key potential disturbance impacts associated with these works are the downstream run-off of suspended solids. The works will be limited based upon the construction of the culverts in advance of its connection to the stream as described from streams 18, 19 and 21 above. In addition, these streams have a low overall flow, works will be undertaken when the flow is low and since there are no species of conservation interest present, disturbance impacts such as siltation and destruction of existing stream bed are anticipated to be insignificant.

8.7.12 Freshwater Marsh (Fong Yuen)

8.7.12.1 No significant habitat loss or fragmentation will ensue directly from the proposed development, with the proposed permanent drainage pipeline and cycle path passing close to the existing road alignment and being designed to avoid the marsh completely, by a distance of 20m. The road works will be some 60-70m at nearest point from the marsh.

Disturbance

8.7.12.2 Disturbance from earth-moving works associated with construction of a storm water sewer and cycle path above the marsh is not expected to impact adversely on the important population of the Ricefish Oryzias curvinotus present in the marsh, since drainage disruption will be avoided and run-off will be collected in a concrete-lined trench before it can enter the marsh (see Section 6). Thus, no impacts are predicted on the Fong Yuen Marsh.

8.7.13 Village Area

8.7.13.1 This habitat accounts for a total area of 23.0 ha within the Study Area. However, the proposed road development will not encroach on this habitat. In view of this, and the fact that Village Area in the Study Area was rated as being of low ecological value, any impacts to this habitat, and to the species occupying it, arising from the project are considered to be insignificant.

8.7.14 Agricultural Land

8.7.14.1 This habitat accounts for a total area of 3.4 ha within the Study Area. However, the proposed road development will not encroach on this habitat. In view of this, and the fact that agricultural land in the Study Area was rated as being of low ecological value, any impacts to this habitat, and to the species occupying it, arising from the project are considered to be insignificant.

8.7.15 Wasteland

8.7.15.1 This habitat accounts for a total area of 7.3 ha within the Study Area, of which 0.23 ha will be permanently lost as a consequence of development. Owing to the small scale of habitat loss, the highly fragmented nature of the habitat and its low ecological value, any impacts to this habitat, and to the species occupying it, arising from the project are considered to be insignificant.

8.7.16 Impacts on Key Species

8.7.16.1 The composition of the terrestrial animal community is largely dependent on the floral structure and species composition of the habitats present (Dudgeon and Corlett, 1994). The study area contains a large number of faunal species of conservation interest, particularly species associated with freshwater habitats. There is a concomitant requirement to preserve habitat integrity as a necessary prerequisite for species conservation, in particular the narrow belt of riparian vegetation, which is an important wildlife corridor. For individual animal species, impacts (particularly construction phase impacts) generally will be higher for species which have lower avoidance ability, either due to having very specific habitat requirements or low locomotory capability, or both. These factors are summarised in Table 8.28 below for the species of conservation interest recorded in the Study Area. It should be noted that larval stages of amphibians and odonates are potentially more vulnerable than adults due to being confined to aquatic habitats.

Table 8.28 Summary of Habitat Requirements, Mobility and Predicted Impacts on Faunal Species (Adults) of Conservation Interest in the Study Area

Species	Conserv-ation Interest	Remarks and Impact Evaluation (see Tables 8.29 and 8.30)	Locations recorded
Melogale moschata (Ferret Badger)	High	Fairly broad habitat tolerance. Moderate to high mobility. Minor impact from road barrier effect	Dead on Tung Chung Road at 150m, (HE 030643), in shrubland habitat
Centropus sinensis (Greater Coucal)	Moderate	Very broad habitat tolerance. High mobility. Insignificant impact	Plantation woodland at numerous locations along Tung Chung Road, around Pak Kung Au and near Cheung Sha
Centropus bengalensis (Lesser Coucal)	Moderate	Fairly broad habitat tolerance. High mobility. Insignificant impact	Tall shrubland/ shrubland-grassland mosaic at numerous locations along Tung Chung Road, around Pak Kung Au and near Cheung Sha
Caprimulgus indicus (Grey Nightjar)	High	Fairly broad habitat tolerance. High mobility. Insignificant impact	Tall shrubland/ shrubland-grassland mosaic near the AFCD Management Centre
Psammodynastes pulverulentus (Mock Viper, snake)	Moderate	Fairly narrow habitat tolerance. Moderate mobility. Insignificant impact	Plantation woodland beside Cheung Sha catchwater, 120m (grid square 0461)
Naja atra (Chinese Cobra, snake)	High	Fairly broad habitat tolerance. Moderate mobility. Insignificant impact	Shrubland, streams at Shek Mun Kap and Stream Nos. 23 and 40
Ptyas korros (Indochinese Rat Snake)	Moderate	Fairly broad habitat tolerance. Moderate mobility. Insignificant impact	Plantation woodland (catchwater) (previous record)
Paramesotriton hongkongensis (Hong Kong Newt)	Very High	Narrow habitat tolerance. Low mobility. Minor to moderate impact from culverting of stream 19 and road barrier effect	Stream No.’s 15, 19
Megophrys brachykolos (Short-legged Toad)	Very High	Narrow habitat tolerance. Low to moderate mobility. Minor to moderate impact from road barrier effect	Tung Chung Stream, Stream No.’s 24, 26, 32, 35, 40, 42
Rana exilispinosa (Lesser Spiny Frog)	High	Narrow habitat tolerance. Low to moderate mobility. Minor to moderate impact from culverting of stream 21	Tung Chung Stream, Stream No.’s 18, 21, 23, 24, 25, 26, 27
Philautus romeri (Romer’s Tree Frog)	Very High	Fairly narrow habitat tolerance. Low to moderate mobility. Minor to moderate impact on breeding sites from tall shrubland loss between Lung Tseng Tau and Pak Kung Au, and from culverting of stream 21	Secondary woodland at Shek Mun Kap and Lung Tseng Tau, plantation woodland and tall shrub at Pak Kung Au, Tung Chung Stream and stream Nos. 21 and 27
Acrossocheilus wenchowensis beijiangensis (Beijiang Thick-lipped Barb, fish)	Very High	Narrow habitat tolerance. Low mobility. Insignificant impact due to avoidance of major works at streams where present	Tung Chung Stream, Stream 15
Parazacco spilurus (Predaceous Chub)	Low	Narrow habitat tolerance. Low mobility. Insignificant impact due to avoidance of major works at streams where present	Tung Chung Stream
Awaous melanocephalus (Black-headed Thick-lipped Goby, fish)	High	Narrow habitat tolerance. Low mobility. Insignificant impact due to avoidance of major works at streams where present	Cheung Sha Stream, Stream 40
Stiphodon atropurpureus (Philippine Neon Goby, fish)	High	Narrow habitat tolerance. Low mobility. Insignificant impact due to avoidance of major works at streams where present	Tung Chung Stream, Cheung Sha Stream, Stream 40
Oryzias curvinotus (Ricefish)	Very High	Narrow habitat tolerance. Low mobility. Insignificant impact due to avoidance of any works in Fong Yuen Marsh	Fong Yuen Marsh
Agriomorpha fusca (Damselfly)	High	Narrow habitat tolerance. Low to moderate mobility. Insignificant impact due to avoidance of major works at streams where present	Stream No.’s 26, 27
Drepanosticta hongkongensis (Damselfly)	Very High	Narrow habitat tolerance. Low to moderate mobility. Insignificant impact due to avoidance of major works at streams where present	Stream No.’s 26, 27
Leptogomphus elegans hongkongensis (Dragonfly)	High	Fairly narrow habitat tolerance. Moderate to high mobility. Insignificant impact due to avoidance of major works at streams where present	Shrubland, stream No. 23
Melligomphus moluami (Dragonfly)	Very High	Fairly narrow habitat tolerance. Moderate to high mobility. Insignificant impact due to avoidance of major works at streams where present	Shrubland, Tung Chung Stream, stream No.’s 23, 25, 26, 27, 32
Protosticta beaumonti (Damselfly)	Very High	Narrow habitat tolerance. Low to moderate mobility. Insignificant impact due to avoidance of major works at streams where present	Stream No.’s 24, 26, 27, 32, 35, 40, 42
Sinosticta ogatai (Damselfly)	Very High	Narrow habitat tolerance. Low to moderate mobility. Insignificant impact due to avoidance of major works at streams where present	Stream No.’s 26, 27
Stylogomphus chunliuae (Dragonfly)	High	Fairly narrow habitat tolerance. Moderate to high mobility. Insignificant impact due to avoidance of major works at streams where present	Stream No.’s 27, 32
Zygonyx iris insignis (Dragonfly)	High	Fairly narrow habitat tolerance. Moderate to high mobility. Insignificant impact due to avoidance of major works at streams where present	Shrubland, Tung Chung Stream, stream No’s 35, 41
Pelopidas subochracea (Large Branded Swift, butterfly)	Moderate	Fairly broad habitat tolerance. Moderate to high mobility. Insignificant impact as no destruction of larval foodplant	Secondary woodland at stream 40, 50m
Telicota colon (Pale Palm Dart, butterfly)	Moderate	Fairly broad habitat tolerance. Moderate to high mobility. Insignificant impact due to no destruction of major larval foodplants	Fong Yuen Marsh
Pachliopta aristolochiae (Common Rose, butterfly)	Moderate	Fairly broad habitat tolerance. Moderate to high mobility. Insignificant impact due to no destruction of major larval foodplants	Secondary woodland, near stream 8. Secondary woodland Cheung Sha Sheung Tsuen, 50m
Eurema brigitta (Small Grass Yellow, butterfly)	Moderate	Fairly broad habitat tolerance. Moderate to high mobility. Insignificant impact due to no destruction of major larval foodplants	Tall shrubland, 60m, HE 0265. Plantation woodland at catchwater (100m, HE 0462), shrubland (120m, HE 0264)
Jamides celeno (Pale Cerulean, butterfly)	Moderate	Fairly broad habitat tolerance. Moderate to high mobility. Insignificant impact due to no destruction of major larval foodplants	Secondary woodland and tall shrub at Shek Mun Kap, plantation woodland and shrubland at 150m, HE 0362
Athyma ranga (Blackveined Sergeant, butterfly)	Moderate	Fairly broad habitat tolerance. High mobility. Insignificant impact due to no destruction of major larval foodplants	Plantation woodland, stream 27, 360m
Euthalia lubentina (Gaudy Baron, butterfly)	Moderate	Fairly broad habitat tolerance. High mobility. Insignificant impact due to no destruction of major larval foodplants	Secondary woodland at Shek Mun Kap, 30m
Moduza procris (Commander, butterfly)	Moderate	Fairly broad habitat tolerance. High mobility. Insignificant impact as no destruction of larval foodplant	Plantation woodland (340m, HE 0363)

Note : Grid references based on Countryside Series Map of Lantau, 2^nd Edition. Grid references are provided on Figures 8.14 to 8.18.

8.7.17 Slope remediation works on obsolete sections of existing Tung Chung Road

8.7.17.1 The proposed slope upgrading works that may generate only limited ecological impacts associated with minor slope trimming works which could result in the removal of some existing shrub vegetation. However, shotcrete removal, soil nailing and rock slope stabilisation measures for rock slopes could also be a source of dust which could affect local flora and fauna. Noise generated for the duration of the works could also cause disturbance to wildlife.

8.7.17.2 In these cases, as no major slope regrading will be undertaken, the dense vegetation at the top of the slopes will largely be undisturbed although the removal of some vegetation in the form of shrubland may be required, but the mature trees will be retained. In view of the small extent of slope works required, the woodland and tall shrubland habitats on the whole would not be adversely impacted. As these types of habitat are common throughout Hong Kong and the habitats are currently subject to frequent disturbance, the overall short-term ecological impacts of construction is considered to be insignificant.

8.7.17.3 In view of the frequent human disturbances generated by the slope maintenance works and the traffic on Tung Chung Road and no rare wildlife was recorded on these artificial habitats, impacts are insignificant. Also, as the works will be of only short duration and the area to be affected is situated within marginal habitats, only minor disturbance to bird species and other wildlife is expected.  In addition, the fauna is mobile and will have the large remainder of the country parks to go into in order to remove themselves from the noise source in the short term. 

8.7.17.4 In terms of habitat loss, as the area of vegetation to be removed is considered marginal and already disturbed, no significant impacts due to habitat loss are predicted.

8.7.18 Night-time Working Activities

8.7.18.1 It is likely that the proposed works will be undertaken up to 11.00pm and also on public holidays in order to condense the works programme. Although there are relatively few studies that have quantified the effects of disturbance on wildlife, there is evidence to suggest that birds (Hill et al., 1997) and mammals such as deer are prone to disturbance associated with human activity (English Nature, 1994). Disturbance impacts are difficult to quantify as many species appear to habituate to disturbances (Treweek, 1999). Hill et al. (1997), however, discussed gradients of responses to disturbance impacts in birds. They concluded that depending on the intensity, frequency and duration, disturbance impacts were greatest during high-intensity activity (especially high amplitude 'startling' noises associated with activity such as shooting).

8.7.18.2 There is also potential for disturbance impacts due to night working on other faunal groups that are nocturnal such as the herpetofauna. As mentioned above for the more readily studied faunal groups such as birds and mammals, although there does not appear to be any studies that have quantified the effects of nocturnal disruption, there is still the possibility that disturbance could affect species such as frogs and toads. There is for example, the potential for a reduction in the breeding success due to mating calls being drowned out by loud noises during construction from the operation of noisy machinery. There is also the possibility that such disturbance will cause the frogs to delay their breeding activities until after the disturbance has ended, i.e. after 11.00pm. Thus the length of time available for breeding will be curtailed. These are, however, worst-case scenarios and it is likely that noise will be within tolerable levels from 6.00pm to 11.00pm and the disturbance caused by evening working is considered to represent an insignificant impact to herpetofauna.

8.7.18.3 It is considered that continuous work activity would induce the greatest impacts to populations of animals that form metapopulations (sub-populations), as a permanent barrier created by the works area is likely to prevent inter-breeding between groups either side of the construction site, thereby potentially impacting faunal groups at the population level. Disturbance pressures are likely to be greater in nocturnal species (most mammals and certain birds) that are unaccustomed to pressures such as high levels of noise, artificial lighting and human activity whilst foraging at night. Construction impacts based on working hours that have breaks, however, as is proposed, are ecologically less severe as they allow windows where mobile fauna can disperse between adjacent areas of habitat (necessary for important ecological processes such as foraging, migration and breeding) and the periods at dawn and during the middle of the night are likely to be important for such activities. It should also be noted that the area where evening work is likely to occur, due to the need to aviod impacts on local residents, is reasonably restricted (from north of the WSD reservoir on the existing Tung Chung Road to opposite the treatment works at the Catchwater) and this will not lead to a continuous barrier of disturbance for the full length of the construction area. As such, a permanent night-time barrier is not anticipated and insignificant ecological impacts are predicted.

8.7.18.4 In summary, working up to 11.00pm, but ceasing work between 11.00pm and 7.00am, is considered a sufficient window to allow nocturnal species such as mammals, birds and herpetofauna time for dispersion between habitats. Such rest-breaks are particularly important to noise-sensitive fauna such as mammal and birds. There is also likely to be an overall ecological gain from extending the working hours to 11.00pm, as this will shorten the construction period. Based upon this, extending the works up until 11.00pm is not predicted to give rise to additional significant impact to fauna in the study area. The mitigation of impacts from construction activities recommended in the EIA for day-time activities should ensure that the impacts are at an acceptable low level during the evening-night period also.

8.7.19 Summary of Construction Phase Impacts

8.7.19.1 A summary of construction phase unmitigated impacts arising from the project is given in Table 8.29 below. Significant impacts, requiring mitigation, are numbered for easy reference in the ensuing discussions.

Table 8.29 Summary of Construction Phase Unmitigated Impacts

General Impact	Specific Impact	Impact No.	Severity	Mitigation
Habitat Loss	Loss of 2.25 ha of high value secondary woodland Loss of 5 ha of plantation woodland Loss of 6.5 ha of tall shrubland Loss of 1 ha of shrubland-grassland mosaic Loss of 0.1ha of stream course	CPI1 CPI2 CPI3	Moderate Minor Minor-moderate Insignificant Insignificant	Required Required Required Not required Not required
Habitat Loss	Reduction of nectar sources for butterflies Secondary woodland loss at Lung Tseng Tau affecting Romer’s Tree Frog Secondary woodland loss impacting on forest-associated avifauna Flora of conservation interest, including the tree Artocarpus hypargyreus, the shrub Pavetta hongkongensis and the orchids Acampe rigida and Liparis viridiflora directly impacted by loss of plantation & secondary woodland Tall shrubland loss impacting on mammals, shrub-associated avifauna and dragonflies of conservation interest Secondary woodland and tall shrubland loss impacting on Romer’s Tree Frog breeding sites (chainage 1.000 to 4.500) Lung Tseng Tau to Pak Kung Au Loss of streambed and riparian vegetation in streams 15, 18, 19 and 21	CPI4 CPI5 CPI6	Insignificant Insignificant Insignificant Minor Insignificant Minor-moderate Minor	Not required Not required Not required Required Not required Required Required
Fragmentation	Fragmentation of plantation woodland close to stream 27 where Romer’s Tree Frog present Fragmentation of tall shrubland habitat Culverting of streams 18, 19 and 21 Extension of barrier and isolation of Beijiang Thick-lipped barb in stream 15 Stream modification works of other streams	CPI7	Insignificant Insignificant Minor Insignificant Insignificant	Not required Not required Required Enhancement^(1) Not required
Disturbance	Disturbance to butterfly and dragonfly species of conservation interest Disturbance to mammal and reptile species of conservation interest Disturbance (noise, increased human activity) to birds Disturbance (earth moving works) to Romer’s Tree Frog between Lung Tseng Tau and Shek Mun Kap Siltation of streams Siltation of Tung Chung Bay Disturbance of stream bed and riparian vegetation at streams 15, 18, 19 and 21 impacting aquatic fauna Disturbance of stream bed and riparian vegetation at other modified streams Disturbance (earth moving works; drainage disruption) at Fong Yuen Marsh impacting Ricefish	CPI8	Insignificant Insignificant Insignificant Moderate Insignificant Insignificant Insignificant Insignificant Insignificant	Not required Not required Not required Required Not required Not required Not required Not required Not required

Note: ¹The road project will not significantly increase fragmentation of stream 15 and lead to greater isolation of Beijiang Thick-lipped Barb although in order to increase the population viability of the species in the Tung Chung Stream catchment as a whole, enhancement measures are recommended for stream 15 (see Section 8.9).

8.8 Operation Phase Unmitigated Impacts

8.8.1 General

8.8.1.1 Operational phase impacts are related to impacts associated with vehicles and include disturbance (noise and increased human activity), barrier effects, vehicular collision and pollution. These potential impacts are discussed further in the sections below.

8.8.2 Disturbance

8.8.2.1 Disturbance impacts incurred during the operational phase are associated with elevated noise levels, presence of vehicles and overall higher contact with associated human activities. Noise levels associated with vehicles rather than visual impacts (perceived threats from moving vehicles) are, however, considered to be the main cause of impact to wildlife species. Territorial species are also known to be more affected by road projects than species that roam over wide areas (Reijnen et al., 1995).

8.8.2.2 It should be noted, however, that many faunal groups are known to habituate to disturbance including elevated noise. Much of the northern part of the Study Area (chainage 1.000 to 4.500, Figures 2.2 to 2.7) is already occupied by the existing Tung Chung Road, which has been in place for several decades.

8.8.2.3 Noise levels associated with vehicles rather than visual impacts (threats) are considered to be the main cause of reduced breeding populations of woodland bird species adjacent to major road schemes (Reijnen et al., 1995). Some territorial bird species are known to show lower breeding densities adjacent to roads such as motorways. However, other species show higher densities closer to roads, although this may be an indirect effect due to different roadside vegetation (Reijnen et al., 1995). As discussed for the construction phase impacts, although the effects of disturbance on birds are difficult to quantify, many bird species in Hong Kong (and elsewhere) are known to habituate to disturbance including elevated noise. Noise levels due to traffic flow are not considered to represent a significant impact. Overall, passive low-level continuous disturbance is better tolerated than active high-level continuous disturbance and furthermore, birds are known to tolerate vehicles and vehicular movement better than people as a source of disturbance (Hill et al, 1997). Operational phase disturbance impacts are, therefore, predicted to be insignificant to avifauna present.

8.8.2.4 Mammal species such as Barking Deer, Ferret Badger, Civets, and Leopard Cat are sensitive to noise and visual disturbance and generally avoid noisy areas. These animals are, however, primarily active at night, during which time noise associated with the road is likely to be significantly lower due to reduced traffic volume. Visual disturbance from large moving vehicles is also reduced at this time. It is therefore considered that the impact of noise and visual disturbance on the foraging behaviour of larger mammals will be insignificant.

8.8.2.5 Noise is not generally considered to cause disturbance to herpetofauna, which tend to be more reliant upon vision or smell for foraging and communication purposes. However, breeding frogs and toads are highly reliant upon auditory communication (i.e., male breeding vocalisations), and it is possible that noise disturbance from the road will have an impact on breeding success of amphibians such as Romer’s Tree Frog, Short-legged Toad and Lesser Spiny Frog in locations proximal to the road. It should be noted, however, that such breeding behaviour is primarily conducted at night, when traffic noise is lower, and that the road will impinge upon a relatively small proportion of the available breeding habitat (streams; moist shady woodland) in the Study Area. It is therefore anticipated that this impact will be insignificant.

8.8.2.6 Noise disturbance from the road will likely have no impact upon the behaviour of insects or fish. It is possible, however, that the visual perception of moving vehicles will have an impact on foraging or day-time perching behaviour of butterflies and dragonflies. These insects are easily unsettled by movement of large objects such as people or vehicles. Owing to the high mobility of these taxa, however, and the small area (relative to available habitat) occupied by the road, this impact is considered to be insignificant.

8.8.3 Barrier Effects

8.8.3.1 The effect of linear developments such as roads that form barriers to wildlife populations are well documented (English Nature, 1994; Treweek, 1999). Many animals are reluctant to cross structures such as roads and they can act as an effective barrier preventing migration between adjacent habitat. A secondary effect related to road operation is the turbulence caused by moving vehicles as many smaller birds are unable to successfully traverse whilst vehicles are passing at high speed without being killed (Treweek, 1999). Most birds are highly mobile, however and the barrier-effect on most species is probably related to vehicle density and speed. Given the relatively low vehicle numbers and vehicle speeds (50 kph) on the Tung Chung Road, there is, therefore, not expected to be a significant barrier effect on any of the bird species present in the Study Area and the potential road operational impacts are considered to be insignificant.

8.8.3.2 Larger mammal species may find the road a physical barrier to movement, although it is only two lanes wide, as an 0.8m high parapet will be constructed on the west side of the road on all retaining walls, which may themselves reach a maximum height of 8m. Also, at times of high traffic flow the road may effectively operate as a barrier due to noise and visual impacts (see above) which induce deliberate avoidance of the road (thereby cutting off the access to habitats on the other side of the road). These impacts are considered minor since most larger mammal activity occurs nocturnally, when traffic flow on the road will be lower. There will also be numerous elevated structures along the road which would serve as corridors to allow mammals to avoid having to cross the road. Also, it should be noted that very few records of large mammal species were obtained during the survey period (May 2001 to January 2002).

8.8.3.3 Herpetofauna are physically able to cross roads provided that barrier walls have not been constructed. However, as noted above, an 0.8m parapet will be in place along much of the road. Also, roads are highly exposed habitats and it is possible that some cryptic species (e.g. Short-legged Toad, Romer’s Tree Frog) would avoid the road as a consequence. The road may therefore effectively act as a barrier to movement of such species between areas of suitable habitat bisected and fragmented by the road. This impact is considered to be minor to moderate in view of the fact that the road will have numerous bridges which would allow herpetofauna to cross safely beneath the road.

8.8.3.4 Herpetofauna of low mobility such as Short-legged Toad, Hong Kong Newt and Romer’s Tree Frog may be impacted by tumbling into U-shaped concrete drainage channels on and at the base of cut slopes. These channels are typically steep and smooth-sided and therefore present a physical barrier to these species. This impact is considered minor.

8.8.3.5 Although the road itself will not create a barrier effect for freshwater fish, fish access within streams may be impeded by channels/culverts constructed at points where the road intersects with such streams. Many culverts and channels constructed for the existing Tung Chung Road incorporate high vertical drop-offs which present a barrier to upstream movement for fully aquatic fauna, such as fish, located downstream of these structures. This may have knock-on effects for breeding and foraging success, and population viability. Streams of high conservation value which may be impacted in this way include No.’s 19 and 21. However, fish species of conservation interest were not recorded in these streams. This impact is, therefore, considered insignificant. Similarly, existing culverts (concrete lined bridge) will be widened at streams 15 and 18 and these existing structures already exert a barrier effect and the extensions to both structures will not incur a significant additional impact.

8.8.4 Mortality due to Physical Impact with Vehicles

8.8.4.1 There are few examples present in the literature that describe the impacts of vehicles on bird populations due to direct physical impacts. The potential for physical impact is especially high where new road projects bisect routes used for activities such as foraging or pathways used for roosting. Birds of prey particularly owls (Reijnen et al., 1995) and probably hawks appear to be particularly susceptible to collision with fast moving vehicles as they often forage along roadsides that provide areas of suitable habitat for their small mammal prey species. Conversely, although evidence is fragmentary, carrion feeders such as crows and magpies may increase in numbers due to the presence of vehicle-killed birds (Reijnen et al., 1995).

8.8.4.2 It has been estimated that bird casualties on roads, mainly motorways, in Europe can vary from between 5000 to 70 million individuals per year (Treweek, 1999), although it is difficult to determine accurately the exact number killed by physical impacts with vehicles. The data available also requires certain caveats as not all roads have traffic travelling at the same speed or traffic density and it is likely that wider roads such as motorways are more difficult to pass than narrower single-lane roads thereby leading to proportionally lower number of casualties in the latter.

8.8.4.3 Also, it is generally assumed that increases in mortality due to road traffic are too small to cause significant decreases in population size and reduction in woodland bird species present is often related to noise impacts (Reijnen et al., 1995). Although the potential for physical impact is persistent, owing to the high mobility of the bird species present, operational impacts are considered to be insignificant.

8.8.4.4 Many species of herpetofauna are susceptible to impacts with motor vehicles when crossing roads during migration to breeding sites (e.g. Hong Kong Newt, which has been observed crossing roads in the New Territories (G.T. Reels, pers. obs.) or whilst foraging (e.g., snakes and lizards). Many herpetofauna species are slow-moving and slow to initiate an avoidance response to approaching vehicles. Many of these exothermic taxa (notably snakes) also actively seek out exposed and sun-heated road surfaces for basking. It is also the case that many motorists coming across snakes or other herpetofauna on the road either do not notice them or do not try to avoid impacting them. During surveys conducted for the current project, one dead Changeable Lizard and one dead Greater Green Snake were observed on the existing Tung Chung Road. Both had presumably been killed by passing vehicles. However, the new road will replace the old road rather than adding to it, and since it is of approximately equal length, has fewer blind corners and passes through broadly similar habitat, it is likely that the net increase in road mortality impact will be insignificant.

8.8.4.5 Larger mammal species are generally agile, and able to perceive approaching vehicles and make an avoidance response before collision. Such animals are also easier to detect than herpetofauna for the oncoming motorist. Impacts do occur, however, and a dead Ferret Badger was observed on the existing Tung Chung Road in October 2001 (north of Pak Kung Au). It should be noted that this is a rare occurrence in the Study Area, due to the scarcity of large mammals present, although it may be argued that this scarcity makes such road mortality all the more significant. In this context it should also be noted that Ferret Badger is a secretive species, thought to be under-recorded in Hong Kong (Reels, 1996), and may therefore be less scarce in the Study Area than previously thought (it appears likely that the individual found dead on Tung Chung Road was the first Lantau record for this species). In view of this, and because of the current and long-established existence of the Tung Chung Road, additional road mortality impact on larger mammals during operation of the new road is considered to be insignificant.

8.8.4.6 Impacts on dragonflies and butterflies due to road mortality are insignificant. Most species do not utilize this habitat and are at any rate highly mobile and able to avoid direct collisions, although air turbulence from passing vehicles may represent a hazard. The ubiquitous dragonfly Pantala flavescens does occasionally show swarming behaviour along horizontal road surfaces, apparently mistaking them for slow-moving rivers, and may incur massive mortality from passing vehicles in the process (G.T. Reels, pers. obs.), but this species is abundant in Hong Kong and elsewhere in the region.

8.8.5 Pollution Emission from Vehicles

8.8.5.1 Pollution impacts from road traffic can affect the abundance of insect prey available to birds. Vehicle emissions could potentially have impacts on the availability of food and hence reduce breeding bird populations (Reijnen et al., 1995). Vehicle emissions also have the potential to induce toxic responses in bird populations inhabiting roadside habitats. Both reduction in insect populations and presence of pollutants are known to be highly localised impacts, however (Reijnen et al., 1995), and they are not considered to represent a significant operational phase impact to the avifauna of the Study Area. As there is only predicted to be relatively low number of vehicles using the road each day, the emissions from vehicles on bird populations is considered to be insignificant. This comment also applies to other terrestrial fauna in the Study Area.

8.8.5.2 Rainwater run-off which has been contaminated by contact with vehicular deposits (such as engine oil, coolant and petroleum fumes) on the road surface may potentially cause a pollution impact, if it enters streams and other watercourses, on benthic invertebrates and on the taxa (fish, tadpoles, odonate larvae) which predate them (CIRIA, 1994). However, all road run-off from the proposed road will be collected via a specially designed carrier pipe and discharged at either end of the alignment into the Wong Lung Hang nullah and an outfall at Pui O Wan. Thus, no run -off into adjacent streams will occur. In view of the interception of run-off this impact is considered insignificant.

8.8.5.3 Accidental spillage of oil or chemicals from an overturned tanker truck may potentially have a severe impact on stream communities. However, the deletion of hair-pin bends on the road should combine to make this a remote possibility. In addition, the road parapet proposed along large sections of the road will prevent vehicles from leaving the road and the permanent drainage pipeline beneath the road will facilitate containment of any spills and diversion away from the streams. A spill response plan will also be implemented as specified in section 6.

8.8.5.4 Notwithstanding it is possible for some of the spilled material to be transferred along the operational carrier pipe and be discharged either in the nullah in Tung Chung or Pui O Wan. It should be noted, however, that the refined oils carried by tankers in Hong Kong although toxic and can cause localised damage when spilled, are likely to disperse quickly. Owing to the relatively high volatility of light oils, fast flowing waters are known to recover rapidly (weeks to months) from spills (Abel, 1996). However, based upon the fact that, for the northern section of the road, any spill would ultimately end up in Tung Chung Bay which contains the San Tau SSSI seagrasses and mangroves, an oil interceptor will be provided for treatment of the road runoff prior to its discharge into the Wong Lung Hang nullah. This will have the effect of minimising the impacts by containing the oil. The emergency response plan and the provision of an oil interceptor are considered to be sufficient to reduce impacts to acceptable levels.

8.8.6 Summary of Operational Phase Impacts

8.8.6.1 A summary of operation phase unmitigated impacts is given in Table 8.30 below. Significant impacts, for which mitigation is required, are numbered for easy reference in the ensuing discussions.

Table 8.30 Summary of Operation Phase Unmitigated Impacts

General Impact	Specific Impact	Impact No.	Severity	Mitigation
Disturbance	Traffic noise and visual impact affecting foraging and / or breeding behaviour of fauna of conservation interest		Insignificant	Not required
Barrier Effects	Barrier effect of road on avifauna Barrier effect of road on larger mammals Barrier effect of road on herpetofauna Barrier effect of U-shaped drainage channels on herpetofauna Barrier effect of culverts on fish (notably streams 19 and 21)	OPI1 OPI2 OPI3	Insignificant Minor Minor-moderate Minor Insignificant	Not required Required Required Required Not required
Mortality due to impact with vehicles	Mortality incurred by avifauna, larger mammals, herpetofauna and insects as a consequence of collisions with vehicles		Insignificant	Not required
Pollution Emission from vehicles	Toxic effects of vehicle exhaust fumes on arthropods and vertebrates which predate them Run-off of contaminated rainwater from road into streams Spillage of oil or chemicals into streams due to leakage from overturned tanker lorry		Insignificant Insignificant Insignificant	Not required Not required Not required

8.9 Impact Mitigation Measures

8.9.1 Construction Phase Impact Mitigation

General

8.9.1.1 The problem of improper spoil disposal is avoidable. Any spoil should be handled, transported and disposed of in accordance with recommendations in Chapter 7. A ticketing system will be used so that the site supervisor can control and keep track of all movements of spoil.

8.9.1.2 Accidental or other intrusion of construction activities beyond the designated works area can also be avoided. The works area boundary should be defined throughout using lightweight but weather-resistant material, such as a low shadecloth barrier. No activities, including casual disturbance, vehicle parking and temporary storage should be tolerated outside the defined boundary.

Habitat Loss

8.9.1.3 Significant (> 1 ha) habitat losses will be incurred in the following habitats: secondary woodland, plantation woodland, tall shrubland (CPI1-3; Table 8.29). These habitats have been determined to have moderate to high ecological value (Tables 8.16, 8.17 and 8.18). Losses are incurred along the road alignment and on cut slopes on the roadside, as well as at the bus lay-by facility at San Shek Wan, the cycle track at Lung Tseng Tau and the drainage pipe to Pui O Wan. These losses may be mitigated by compensatory planting.

8.9.1.4 All land disturbed during the course of the works, that will not subsequently be occupied by the permanent works, will be planted with either native woodland or scrub species. As a general approach, where these areas lie within or alongside areas of existing woodland then woodland species will be planted. In areas bordered by tall and low scrub, then appropriate scrub species will be established.

8.9.1.5 The steepness of slopes poses a constraint on the species that can be planted. Slopes with a gradient of more than 35 degrees would only be suitable for the establishment of small tree and tall shrub species (not tall tree species). Slopes with a gradient of more than 45 degrees would only be suitable for the establishment of shrub species. Soil cut slopes in excess of 55-60 degrees would not be suitable for the establishment of woody vegetation, but would be hydro-mulched with grass, climbers and groundcover species to provide a green appearance.

8.9.1.6 The limit of disturbance has been contained as far as is practical to the alignment of the new and widened carriageways. This results in only a very limited amount of space for re-planting.

8.9.1.7 As the road is bordered for the most part by scrub or woodland habitats, there are very few opportunities for additional planting in areas outside the limit of disturbance.

8.9.1.8 Additional infill planting is also proposed in several open areas alongside the new or existing road corridors, where existing low scrub vegetation would be enhanced with the addition of native tree species.

8.9.1.9 In determining the quantity of the compensation planting for ecological mitigation, reference has been made to the area of land available for planting, and it's ability to support different types of vegetation (especially restriction of tree planting on steep slopes). Approximately 14.5ha of habitat will be lost as a result of the Tung Chung Road Widening project and a general approach of providing compensation planting more than the area lost has been proposed. Details of the planting are summarised in Table 8.31 below. Details of the species to be planted are provided in the mitigation proposals in Section 9 and Appendix I.

Table 8.31 Summary of Planting Proposals

Planting Type	Area (ha)	Description
Native woodland planting	12.0 ha	Newly formed slopes and reinstated disturbed areas alongside existing areas of secondary woodland, plantation woodland or tall scrub. Grass hydro-seeding with native grass seed mix and groundcover plants, with pit planting of native tree species at 1.5 metre centres
Infill planting	12.5 ha	Planting of native trees in previously disturbed bare ground or areas of existing low shrub adjacent to the road, to mitigate for loss of landscape resource and character, for visual screening and to enhance ecological diversity. Estimated that some 2000 no. new trees per hectare could be planted in this manner
Grass hydroseeding with Groundcover plants, or hydro-mulching	0.55 ha	Rehabilitation of existing steep slopes by grass hydro-seeding with native grass seed mix and groundcover plants or by hydro-mulching
Individual trees	321 no.*	Heavy Standard trees planted in flat ground, or pavement tree pits alongside the road
Amenity shrub planting	0.4 ha	Ornamental shrub planting in planter beds alongside the road.
TOTAL	25.45ha

*not included in area total

8.9.1.10 Existing man-made slopes, such as those requiring remediation on obsolete sections of the existing Tung Chung Road, should be restored using landscape treatment and bio-engineering methods such as grasscrete and terracing following current Hong Kong best practice as described in GEO (2000) to maximise opportunities for the establishment of native woodland vegetation as far as possible.

8.9.1.11 To further offset these losses on-site, cut slopes for the new road should be landscaped using environmentally-sensitive bio-engineering measures as described in GEO (2000) such as incorporating a ‘stepped’ design with soil in the resultant terraces to facilitate establishment of vegetation. Native tree and shrub species should be established, and grass slopes avoided where possible.

8.9.1.12 Where the new road passes close to or over individual plants of protected floral species (as at chainage 4.380, Pavetta hongkongensis, Aceampe rigida and Liparis viridiflora (protected under the Forests and Countryside Ordinance Cap. 96); and chainage 5.080, Artocarpus hypargyreus (a Class III nationally protected species in China) (CPI4), these individuals may be transplanted to safe locations within the same habitat. However, since these species are all common in Hong Kong, the ecological effect of their loss would be minor. Whilst there is not considered to be a strong ecological reason for transplantation, a tree survey has been conducted and will recommend transplantation of the individual plants concerned. A translocation specification is recommended to be prepared during the design phase of the project and a design audit conducted to ensure it meets the requirements of the EIA.

8.9.1.13 Romer’s Tree Frog may be impacted by loss of secondary woodland and tall shrubland habitat between Lung Tseng Tau and Pak Kung Au (CPI5). Breeding sites for this species include small seepages close to streams. This impact can be mitigated by the compensatory planting of woodland and shrubland. In addition, provision of alternative breeding sites, such as water-filled pots placed on the ground at remote locations away from disturbance would mitigate for the loss of habitat until the replacement planting has matured. The main criterion for the pots is that predatory fish are excluded. Such measures have been successfully tried and tested in Hong Kong (Dudgeon and Lau, 1999). A specification for the pots is recommended to be prepared during the design phase of the project and a design audit conducted to ensure it meets the requirements of the EIA.

8.9.1.14 Four streams of high ecological value, Streams 15, 18, 19 and 21, will be disturbed by culverting activities. These will impact on the Hong Kong Newt and Beijiang Thick-lipped Barb at Stream 15, Hong Kong Newt at Stream 19 and Romer’s Tree Frog and Lesser Spiny Frog at Streams 18 and 21. These species have low mobility (Table 8.28) and impacts can be minimised by evacuation and translocation of individuals of these species to appropriate nearby locations. Such procedures should be conducted by an experienced herpetologist in advance of works commencing in the impacted areas (CPI6). In addition, all works in these streams should be avoided between April and June in order to avoid impacts during the key breeding period for the Romer’s Tree Frog and Lesser Spiny Frogs, fish and odonates. This period has been selected on the basis that it covers the key breeding periods for the majority of species of conservation interest recorded. The exception to this is the Hong Kong Newt, which breeds between September and March/April (pers. comm. M Lau), and although the latter month of the Hong Kong Newt’s breeding season will be covered by this recommendation, it is not practicable to limit the works for the whole period September to June. As such the period which would protect the largest number of key species has been selected. Notwithstanding, impacts on the Hong Kong Newt are not expected as this and other species will be translocated from the streams prior to the works.

Fragmentation

8.9.1.15 Habitats within the Study Area are already highly fragmented. Mitigation is, however, recommended where the proposed road development incurs further fragmentation on Streams 18, 19 and 21 (CPI7), since these are ecologically valuable habitats (Table 8.20). Although no fish species of conservation interest were recorded at these streams, the presence, albeit at low density, of other aquatic fauna of conservation interest (Lesser Spiny Frog in Stream 18, Hong Kong Newt at stream 19; Romer’s Tree Frog and Lesser Spiny Frog at Stream 21) means that these streams have a high ecological value. The fragmentation caused by construction of the culverts may be minimised by incorporation of ecologically-sensitive designs, as illustrated in Figure 8.20a-c. The culvert outlets should avoid a steep descending wall which is difficult for aquatic fauna to negotiate and should incorporate a gentle rise with pooled areas. Mitigation for culvert inlets and outlets (Figure 8.20a and 8.20c) is applicable to all three streams. For the new culverts at Streams 19 and 21, the culverts are long (30m and 50m, respectively) and should incorporate pooled areas where aquatic fauna can rest and shelter (Figure 8.20b). No benefit would be achieved by applying this technique to Stream 18 as the existing culvert is present and would not be reconstructed.

8.9.1.16 Stream fragmentation primarily affects obligate aquatic fauna such as fish. Stream 15 is already fragmented by a sheer rock wall upstream and a steep culvert exit downstream and construction of the project will not increase this level of fragmentation significantly. However, as the stream contains a small population of Beijiang Thick-lipped Barb that is isolated from the rest of the Tung Chung Stream and its tributaries (including nearby streams 19 and 21), it is proposed, in order to enhance the viability of the population overall, that these fish in Stream 15 may be captured and translocated downstream so that they can enhance the main population of this restricted species in Tung Chung Stream.

8.9.1.17 This can be achieved by the capture of the Beijiang Thick-lipped Barb in Stream 15 using a hand- net and releasing them downstream of the existing bridge, which will enhance the population in the Tung Chung stream of which Streams 19 and 21 are tributaries.

Disturbance

8.9.1.18 Disturbance is an unavoidable impact of the road construction process. However, it will have insignificant impacts on mobile taxa, or taxa which do not have highly specific habitat requirements, such as birds, large mammals, most reptiles, and butterfly and dragonfly adults, and in these cases disturbance impacts will be avoided by ensuring that the works area is as small as practicable and kept within well-defined limits, with shade-cloth shielding the works area, proper spoil disposal and no intrusion into habitats beyond the works area boundary. These considerations especially apply to streams with known high conservation.

8.9.1.19 The Romer’s Tree Frog was also recorded in good numbers in secondary woodland (e.g., Lung Tseng Tau, chainage 1.100 to 1.200; Shek Mun Kap, near chainage 1.700 to 1.800) and tall shrubland (chainage 1.000 to 4.500) and, being a species of low mobility (Table 8.28) will be impacted by disturbance from construction works (CPI8). Impact avoidance is not possible and thus, the impact may be mitigated by translocation of individual frogs present at breeding sites on or adjacent to the road alignment. This includes areas of woodland and shrubland on the east side of the road but also on the west above the Fong Yuen Marsh. Such measures should be conducted by an experienced herpetologist in advance of works commencing in the impacted areas.

8.9.1.20 A summary of the construction phase mitigation measures is provided in Table 8.32 below.

Table 8.32 Summary of Construction Phase Impacts and Mitigation Measures

Impact	Description	Mitigation
CPI1	Loss of 2.25ha of secondary woodland	Restoration of disturbed areas in accordance with the landscape proposal. Road width kept to minimum to avoid habitat loss. Cut slopes for the new road should be landscaped using environmentally-sensitive bio-engineering measures as described in GEO (2000) such as incorporating a ‘stepped’ design with soil in the resultant terraces to facilitate establishment of vegetation. Intensive planting of native tree species at the bus lay-by facility site at San Shek Wan and the Pui O Wan drainage channel. Intensive and extensive replanting and enhancement (with appropriate native tree species) along works area. Onsite and off-site compensatory planting of about 25.5ha, which is more than the area lost.
CPI2	Loss of 5 ha of plantation woodland	As above
CPI3	Loss of 6.5 ha of tall shrubland	As above
CPI4	Flora of conservation interest, including the tree Artocarpus hypargyreus, the shrub Pavetta hongkongensis and the orchids Acampe rigida and Liparis viridiflora directly impacted by loss of plantation woodland / secondary woodland	Transplantation to appropriate nearby location (to be determined during preparation of the specification during the detailed design)
CPI5	Secondary woodland and tall shrubland loss impacting on Romer’s Tree Frog breeding sites (chainage 1.000 to 4.500)	Provision of water-filled pots in areas lost. Intensive planting of native tree species along roadside between Lung Tseng Tau and Shek Mun Kap in accordance with the landscape proposal as noted above
CPI6	Loss of stream bed and riparian vegetation at streams 15, 18, 19 and 21 impacting aquatic fauna	Evacuation and translocation of Hong Kong Newt, Romer’s Tree Frog and Lesser Spiny Frog to suitable adjacent sites by experienced herpetologist prior to works commencing. In addition, all works in these streams should be avoided between April and June in order to avoid impacts during the key breeding period.
CPI7	Culverting of streams 18, 19 and 21	Culverts portals and culverts designed so as to facilitate access/ passage by aquatic fauna (such as ridged floor and incorporation of stone substrate see Figure 8.20a-b). Note that the incorporation of a ridged floor into the culvert extension in Stream 18 would not serve any benefit as the existing culvert will not undergo any reconstruction
CPI8	Disturbance (earth moving works) to Romer’s Tree Frog between Lung Tseng Tau and Shek Mun Kap	Evacuation of Romer’s Tree Frogs by experienced herpetologist prior to works commencing.
Enhancement Measure	Stream 15	Fragmentation of Beijiang Thick-lipped Barb population isolated at nearby stream 15 alleviated by evacuation of fish and release into Tung Chung Stream (of which 15, 19 and 21 are all tributaries) and incorporation of slope surfaces at step channels for aquatic fauna to move up and down stream at the existing bridge at stream 15 (see Figure 8.20c)

8.9.2 Operation Phase Impact Mitigation

Disturbance

8.9.2.1 No significant disturbance impacts are anticipated in the operation stage and thus no mitigation is required.

Barrier Effects

8.9.2.2 Barrier effects are not significant for flying fauna such as birds, dragonflies and butterflies, and are considered only minor for larger mammals (Table 8.29, OPI1). Although the road will have numerous bridge crossings which will enable mammals to safely negotiate the road and it may be possible for small mammals such as Ferret Badgers to make use of wildlife tunnels. These will also be able to be used by amphibians as discussed below. Wildlife tunnels comprise relatively narrow (0.5m) tunnels that pass under the road.

8.9.2.3 Certain cryptic herpetofauna, including species of high conservation interest such as Romer’s Tree Frog and Short-legged Toad, may suffer minor-moderate impact due to inability to cross because of the exposed nature of the road (OPI2). This impact may be mitigated by construction of “wildlife tunnels” under the road in appropriate locations.

8.9.2.4 It is only possible to provide these tunnels in certain locations. It will not be practical to provide wildlife tunnels under the new road in locations where the road is bounded by a retaining wall on the uphill side of the road due to the level difference between the natural ground behind the retaining wall and the road. Any entrance behind the retaining wall would require a vertical shaft to reach the tunnel under the road. Wildlife would unlikely be able to climb the shaft, which would also attract debris and likely become blocked. However, tunnels can be provided at 3 locations where there are few bridges as shown in Figure 8.21. Based upon these and the provision of the bridges all along the route, mammals and amphibians will have opportunity to traverse the road safely.

8.9.2.5 Reptiles and amphibians are particularly liable to fall into, and then be unable to escape from, smooth, steep-sided concrete channels (such as those required for drainage purposes on cut slopes). Romer’s Tree Frog, Short-legged Toad, Hong Kong Newt and Chinese Cobra may all be impacted in this way (OPI3). Such channels should be provided with regular “escape ramps” and/or natural breaks such as exposed stable rock walls to permit these animals to climb out of the channels. Such elements should be included in the EM&A Manual and should be an integral element of the detailed design stage of the project. A conceptual drawing is given in Figure 8.22.

Mortality due to Physical Impact with Vehicles

8.9.2.6 No significant impacts are anticipated as a consequence of animal mortality due to collision with vehicles.

Pollution Emission from Vehicles

8.9.2.7 No significant impacts are anticipated from vehicle exhaust fumes or from run-off of contaminated rainwater from the road into streams.

8.9.2.8 The risk of pollution impacts from accidental chemical or oil spills from tanker trucks is considered insignificant due to the robust road parapet, the incorporation of a permanent road drainage system and the oil interceptor at Wong Lung Hang (see Section 6).

8.9.2.9 A summary of operational phase impacts and mitigation measures is provided in Table 8.33 below.

Table 8.33 Summary of Operation Phase Impacts and Mitigation Measures

Impact	Description	Mitigation
OPI1	Barrier effect of road on larger mammals	Construct “wildlife tunnels” under the road in appropriate locations (see Figure 8.21)
OPI2	Barrier effect of road on herpetofauna	Construct “wildlife tunnels” under the road in appropriate locations (see Figure 8.21)
OPI3	Barrier effect of U-shaped drainage channels on herpetofauna	Channels should be provided with regular “escape ramps” and/or natural breaks such as exposed stable rock walls to permit these animals to climb out of the channels (see Figure 8.22)

8.10 Residual Impacts

8.10.1 The proposed mitigation measures are considered adequate to fully mitigate all of the identified impacts in the short term with the exception of woodland losses and the barrier effects of the road to mammals and herpetofauna, as discussed below.

8.10.2 Residual impacts after mitigation will remain for loss of habitat (CPI1-3) as this cannot be fully compensated in the short term. Initially residual impacts are expected to be moderate as the trees planted will be young and not capable of sustaining a large number of associated core woodland species. As the planting matures, particularly the planting in the existing plantation woodland which should continue to support faunal species, its value and ability to support species will increase and residual impacts will reduce to minor and then insignificant after a period of 15 to 20 years, when the woodland canopy will be mostly closed and suitable for colonisation by a range of faunal species. Table 8.34 provides an evaluation of the residual environmental impacts.

Table 8.34 Evaluation of Secondary Woodland Residual Environmental Impacts

Residual Impact Evaluation Criteria	Secondary Woodland
Effect on public health and health of biota or risk to life.	Not applicable key sedentary species of conservation interest will be translocated in key areas
Magnitude of Residual Impacts	Small in relation to overall habitat present
Geographic extent of impacts	Small extent
Duration of Impacts	Permanent loss of some habitat during life-time of the project although impacts are moderate to insignificant (after ~ 20 years)
Size of the environment that may be affected	Relatively small in comparison to habitat present although relatively large in comparison to the classification under the TM (maximum loss of 2.25ha; TM classes >1ha as important)
Reversibility of Impacts	Reversible although habitat will be replaced by planting a greater area of woodland in other locations. Although the planted woodland will not function ecologically the same as a mature woodland in the short-term, ecological function and species compliment likely to be similar in the longer-term (ca. 20 years)
Ecological Context	Moderate to high value
International and Regional importance	Not important either regionally or globally
Certainty that impacts will occur	High
Residual Impact	Short-term Moderate, Medium-term Minor and Long- term Insignificant

8.10.3 Minor residual impacts of road barrier effects to mammals and herpetofauna (OPI1-2) will persist since mitigation measures (wildlife tunnels) cannot be applied along the entire length of the road. However, such impacts will be ameliorated to some extent by the presence of numerous elevated structures under which fauna may pass. Table 8.35 provides an evaluation of the residual environmental impacts.

Table 8.35 Evaluation of Barrier Effects of the Road

Residual Impact Evaluation Criteria	Barrier Effects of Road
Effect on public health and health of biota or risk to life.	Not applicable
Magnitude of Residual Impacts	Low
Geographic extent of impacts	6.2 km of road from Lung Tseng Tau to Cheung Sha
Duration of Impacts	Permanent barrier effect during life-time of the project
Size of the environment that may be affected	Not applicable
Reversibility of Impacts	Irreversible
Ecological Context	Species of conservation interest may be isolated from adjacent habitat areas used for foraging or breeding
International and Regional importance	Insignificant to minor
Certainty that impacts will occur	Moderate
Residual Impact	Minor

8.10.4 As discussed above, although some insignificant to minor residual impacts will occur these are considered to be acceptable within the overall benefits of the scheme.

8.10.5 A summary of the proposed mitigation measures and predicted residual impacts is provided in Table 8.36 below.

8.10.6 In addition to those recommended above in Table 8.32 and Sections 8.7.7.19 and 8.7.7.20, the following mitigation measures are applicable to this ecological assessment and were presented in detail in the sections on Water Quality and Waste Management. The further ecological mitigation measures comprise the following:

¨ areas in which works have been completed shall be restored within 3 weeks of being completed in accordance with the Landscape Proposal detailed in Appendix I;

¨ temporary cut slopes shall be covered with tarpaulin during rainfall when not being worked to prevent erosion;

¨ major stockpiles shall be sited outside the Country Park and away from stream courses;

¨ spoil heaps shall be covered at all times to minimise losses in the wet season and reduce the dust emissions during the dry season to minimise impacts on air quality;

¨ stock piled materials shall be contained in a designated area down gradient from any stream or up gradient with a suitably constructed barrier to reduce loss of materials to the stream; and

¨ sedimentation tank, oil interceptors and channels used during the construction phase should be inspected every day rainfall is recorded and weekly thereafter. The sedimentation tanks and channels should be cleaned out when the volume of settled sediments amounts to 10% of the total volume of the tank and the oil interceptors should be cleaned when the volume of oil amounts to 50% of the total volume of the oil interceptor.

Table 8.36 Summary of Proposed Mitigation Measures and Predicated Residual Impacts

Potential impact	Potential mitigation measures				Predicted residual impact
	Avoidance	Minimising	Compensation
			On-site	Off-site
CPI1-3 Habitat loss impacting on secondary woodland, plantation woodland and tall shrubland	Much of the northern section of the proposed road (chainage 1.000 to 3.200) is online with the existing Tung Chung Road, thereby substantially avoiding habitat loss along this section. The road also incorporates a number of elevated structures, particularly over streams of very high conservation importance.	Restoration of disturbed areas in accordance with the landscape proposal. Road width along its length kept to minimum to minimise habitat loss.	Cut slopes for the new road should be landscaped using environmentally-sensitive bio-engineering measures as described in GEO (2000) such as incorporating a ‘stepped’ design with soil in the resultant terraces to facilitate establishment of vegetation. Intensive planting of native tree species at the bus lay-by facilities site at San Shek Wan and the Pui O Wan drainage channel.	Intensive and extensive replanting (with appropriate native tree species) along works area. On-site and off-site compensatory planting of about 25.5ha.	Short term low residual impact on habitats unavoidable since not all habitat loss can be mitigated in the short term. However, loss of high ecological value habitat minimised. No residual impacts on Species of Conservation Interest No significant residual impacts from direct habitat loss on other habitats.
CPI4 Four plants of conservation interest directly impacted by loss of plant-ation wood- land / secondary woodland	Unavoidable	Transplantation to appropriate nearby location	Not applicable	Re-planting at appropriate nearby location	No residual impact
CPI5 Secondary woodland and tall shrubland loss impacting on Romer’s Tree Frog (chainage 1.00 to 4.500)	Unavoidable	Not applicable	Intensive planting of native tree species along roadside between Lung Tseng Tau and Shek Mun Kap. Provision of water filled pots as habitats for the Romer's Tree Frog.	Not applicable	No residual impacts
CPI6 Loss of streambed impacts on aquatic fauna by culverting activities at streams 15, 18, 19 and 21	Scheduling of the works in streams 15, 18, 19 and 21 to avoid the period April to June which is the key breeding period for fish, amphibians and odonates.	Evacuation and translocation of Hong Kong Newt, Romer’s Tree Frog and Lesser Spiny Frog to suitable adjacent sites by experienced herpetologist prior to works commencing	Not applicable	Not applicable	No residual impact
CPI7 Fragmentation of streams 18, 19 and 21 due to culverting	Unavoidable	Culverts and culvert exits designed so as to facilitate access by aquatic fauna	Not applicable	Fragmentation of Beijiang Thick-lipped Barb population isolated at nearby stream 15 alleviated by evacuation of fish and release into Tung Chung Stream (of which 15, 19 and 21 are all tributaries)	No residual impacts.
CPI8 Disturbance to Romer’s Tree Frog between Lung Tseng Tau and Shek Mun Kap	Unavoidable	Evacuation of Romer’s Tree Frogs by experienced herpetologist prior to works commencing.	Not applicable	Not applicable	No residual impact
OPI1-2 Barrier effect of road on larger mammals and herpetofauna	Large number of new bridge structures proposed which will provide access.	Construct “wildlife tunnels” under the road in appropriate locations.	Not applicable	Not applicable	Minor residual impact as mitigation cannot be applied along all stretches of road.
OPI3 Drainage channels traversing and at foot of cut slopes isolate herpetofauna of conservation interest which fall into them	Channels should be provided with regular “escape ramps” and/or natural breaks such as exposed stable rock walls to permit these animals to climb out of the channels	Not applicable	Not applicable	Not applicable	No residual impact

8.11 Environmental Monitoring and Audit

8.11.1 It is stipulated that EM&A for ecology is undertaken during both the design, construction and operational phases of the project to ensure that all mitigation measures are fully complied with. In particular, the objectives of the design audit shall be to ensure that the design process clearly implements the design ecology mitigation specified in the mitigation Section 8.9 and to ensure that such designs are ecologically feasible and effective. The construction and operational EM&A objectives shall be to ensure that the ecological contract works and construction mitigation procedures recommended in Section 8.9 are carried out as specified and are effective. The construction and operational phase EM&A will be carried out as part of the site audit programme. Further details are provided in the EM&A Manual.

8.12 Conclusions

8.12.1 The Study Area traverses a large area of Country Park and contains a variety of habitats, of which the most ecologically valuable are streams, secondary woodland, mature plantation, tall shrubland and freshwater marsh. The Study Area is also floristically diverse and supports numerous faunal species of moderate, high, and very high conservation interest, many of which have a narrow habitat tolerance and are sensitive to disturbance. Overall, the Study Area is of high ecological value.

8.12.2 The species of conservation interest recorded along the alignment were also observed in other locations in the broader survey area (Figures 8.14 to 8.19). The species of conservation interest would, therefore, appear to be widely distributed in the wider survey area and do not appear confined to any single location. It is interesting to note that the majority of species of conservation concern were also recorded along the existing Tung Chung Road (15 faunal species of conservation interest recorded) indicating that the conservation requirements of these species are being met and the operation of the existing road have not seriously compromised the species present.

8.12.3 Construction of the proposed road should, accordingly, be conducted in an appropriately ecologically sensitive manner, with particular attention given to avoidance or minimisation of disturbance through appropriate works scheduling and responsible works area maintenance. Many streams in the Study Area are of particularly high ecological value due to a previous low level of anthropogenic disturbance, and it is crucial that impacts to these highly sensitive habitats are avoided or minimised wherever possible. Loss of ecologically valuable habitat should be mitigated both on-site and off-site to the fullest possible extent.

8.12.4 Based upon this, in addition to the measures integrated into the design of the road such as reduction of the road footprint as far as possible to minimise habitat loss and a comprehensive drainage system to avoid and minimise impacts on the stream courses, a series of ecological mitigation measures have been proposed to reduce impacts to acceptable levels.

8.12.5 The effectiveness of impact mitigation measures will need to be assessed by the proposed environmental monitoring and audit programme. As discussed above, some residual impacts will occur but these are considered to be acceptable within the overall benefits of the scheme.

8.13 References

Abel, P.D. (1996). Water Pollution Biology. Taylor and Francis, London.

Bascombe, M.J. (1995). Check list of the butterflies of South China. Memoirs of the Hong Kong Natural History Society 20: 1-206.

Bascombe, M.J., Johnston, G. and Bascombe, F.S. (1999). The Butterflies of Hong Kong. Academic Press, London.

Bibby, C.C., Burgess, N.D. and Hill, D.A. (1992). Bird Census Techniques. Academic Press, London.

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, T. and Young, L. (2001). The Avifauna of Hong Kong. Hong Kong Bird Watching Society.

Chan, B. (1998). Sightings of three freshwater fish. Porcupine! 18: 9.

Chan, B. (1999). Hong Kong’s Freshwater Fish: Who Cares?!? Porcupine! 19: 15-16.

Chan, B. P. L. (2001). Sustainability and Biodiversity: The Impact, Alternative Design and Prospects of Restoration of Channelized Lowland Streams in Hong Kong. Ph.D. thesis. University of Hong Kong.

Chong, D.H. and Dudgeon, D. (1992). Hong Kong stream fishes: An annotated checklist with remarks on conservation status. Memoirs of the Hong Kong Natural History Society 19: 79-112.

Chou, I. (1994). Monographia Rhopalocerorum Sinensium. Henan Scientific and Technological Publishing House, Zhengzhou.

CIRIA (1994). Control of pollution from highway drainage discharges. Report No. 142, CIRIA, London.

Dudgeon, D. (1999). Conservation evaluation of freshwater wetlands: A proposed metric for Hong Kong. Memoirs of the Hong Kong Natural History Society 22: 113-126.

Dudgeon, D. and Corlett, R. T. (1994). Hills and Streams: An Ecology of Hong Kong. Hong Kong University Press, Hong Kong.

Dudgeon, D. and Lau, M. (1999). Philautus romeri in Hong Kong: a frog re-introduction to a degraded tropical landscape. Porcupine! 19: 18-19.

English Nature (1994). Roads and Nature Conservation: Guidance on impacts, mitigation and enhancement.

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. (in press). Wild animals to watch: terrestrial and freshwater fauna of conservation concern in Hong Kong. Memoirs of the Hong Kong Natural History Society.

GEO. 2000. Technical Guidelines on Landscape Treatment and Bio-engineering for Man-made Slopes and Retaining Walls. Hong Kong: Geotechnical Engineering Office, Civil Engineering Department, The Government of the Hong Kong Special Administrative Region.

Gibbons, D.W., Hill, D. and Sutherland, W.J. (1996). Birds. In: Ecological Census Techniques. W.J. Sutherland (Ed.). Cambridge University Press, Cambridge. pp. 227-259.

Goodyer, N.J. (1992). Notes on the land mammals of Hong Kong. Memoirs of the Hong Kong Natural History Society 19: 71-78.

Hill, D.S. and Phillipps, K. (1981). Hong Kong Animals. Government Printer, Hong Kong.

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

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

Kwok, H.K. 1996. The Seasonality of Forest Birds in Hong Kong. PhD Thesis. Hong Kong: University of 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.

Leven, E.P. and Leven, M.R. (2000). Designing Bird-friendly Landscapes. YuanLin 1999/ 2000: 39-45.

Leven, M.R. (2001). Shrubland Birds in Hong Kong: Community Structure, Seasonality and Diet. PhD Thesis. Hong Kong: University of Hong Kong.

Lynch, P. (2001). Wild Corner. Porcupine! 23: 27.

MacKinnon J. and Phillipps, K. (2000). A Field Guide to Birds of China. Oxford University Press, New York.

Man, S.H. and Hodgkiss, I.J. (1981). Hong Kong Freshwater Fishes. Urban Council, Hong Kong.

Mouchel (1998). Widening of Tung Chung Road Feasibility Study. Prepared for Highways Department.

Reels, G.T. (1996). Distribution of large mammals in Hong Kong – a summary of sightings made since mid-1992. Porcupine! 15: 36-38.

Reels, G.T. (2001). Two Hong Kong ‘endemics’ sunk at Wutongshan. Porcupine! 23: 5.

Reels, G.T. and Crow, P. (1999). Hong Kong’s Barking Deer – What is it? Porcupine! 19: 21.

Reels, G.T. and Walthew, G. (1998). Status of Hong Kong butterflies – an update. Porcupine! 17: 11.

Reijnen R., Foppen R., Ter Braak C. and Thissen J. (1995). The effect of car traffic on breeding bird populations in woodland. III. Reduction of density in relation to the proximity of main roads. Journal of Applied Ecology 32: 187-202.

Siu, G.L-p. (2000). Orchidaceae of Hong Kong. Memoirs of the Hong Kong Natural History Society 23: 137-148.

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

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

Wang S. et al. (Ed.). (1998). China Red Data Book of Endangered Animals: Aves. Science Press, Beijing. (In both English and Chinese).

Walthew, G. (1997). The status and flight periods of Hong Kong butterflies. Porcupine! 16: 34-37.

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

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

Wilson, K.D.P. and Reels, G.T. (2001). Odonata of Hainan, China. Odonatologica 30(2): 145-208.

Wu, S-h. and Lee, W.T-c. (2000). Pteridophytes of Hong Kong. Memoirs of the Hong Kong Natural History Society 23: 5-20.

Xing, F., Ng, S-c. & Chau, L.K.C. (2000). Gymnosperms and Angiosperms of Hong Kong. Memoirs of the Hong Kong Natural History Society 23: 21-136.

Yen C.Y. et al. (1996). A Field Guide to the Birds of China. Kingfisher Publisher, Taiwan. (In Chinese).

Young, J.J. and Reels, G.T. (1998). A brief note on the distribution and conservation of Birdwing butterflies in Hong Kong. Porcupine! 17: 10.