Advisory Council on the Environment

Environmental Impact Assessment Ordinance (Cap. 499)

(ACE-EIA Paper 12/2002)
For Advice

Environmental Impact Assessment Report
Application No.-EIA-077/2002


1. This paper intends to inform members of the ACE EIA Subcommittee the preliminary findings and recommendations on the key issues of the Environmental Impact Assessment (EIA) study for the Shenzhen Western Corridor (SWC) Project. The EIA Report is being finalized and will be submitted under section 6(2) of the Environmental Impact Assessment Ordinance (EIAO) in July. The Highways Department (HyD) and the consultants will make a presentation.

2. Members' views and comments on these and other key issues are welcome. HyD would take into account views and comments from the ACE when finalizing the EIA Report.

Need for the Project

3. Travel between Hong Kong and Mainland China using land transport at the 3 existing boundary crossings can be a long and frustrating business. In the height of the peak touring season or freight traffic periods, long waiting time is not unusual. To relieve the traffic burden and to promote economies, the HKSAR and the Shenzhen governments determined to find a fourth crosslink, which is now known as the SWC.

Description of the Project

4. The SWC is a strategic highway in the Hong Kong Special Administrative Region (HKSAR) and will form a major crosslink between HKSAR and the Mainland. In total it is 5.2-km long and is built to dual-3 lane expressway standards throughout. The transport corridor runs from Ngau Hom Shek in the northwestern part of the New Territories, stretches across the Deep Bay, and enters into the newly reclaimed land in Dongjiaotau at Shekou, and co-locates HKSAR's boundary crossing facilities (as shown in Figure 1 MSWORD). The HKSAR and Mainland Governments will build the SWC in a concerted effort, with each responsible for building its own bridge within the respective boundary to an agreed standard.

5. The SWC is a Designed Project under section A.1, Part I of Schedule 2 of the EIAO.

Public Consultations and Key Issues

6. This project has undergone vigorous public consultations throughout the EIA process to-date. HyD and its consultants first introduced this project to ACE's EIA Sub-committee in September 2001. A discussion paper on alignment options assessment was also submitted to ACE in October 2001. Two rounds of consultations with five Non-Government Organisations (NGOs) have been held in September/October 2001 and February/March 2002 respectively, to keep them abreast of project status, alignment selection and initial EIA findings, and to solicit their views and exchange ideas. Their views and comments were instrumental in shaping the focus of the EIA study. A third round of consultation with these NGOs has been planned for late May/early June 2002.

7. HyD and its consultants have also vigorously consulted local communities. These include the District Councils of Tuen Mun and Yuen Long, the Rural Committees of Tuen Mun and Ha Tsuen. A Liaison Group with the oyster farming representatives was also formed to act as a conduit between government departments and the oyster farmers.

8. Key issues identified during these consultations include:

  • Selection of alignment options particularly bridge vs tunnel
  • Air quality impacts particularly due to the use of lower standard fuel from Mainland by cross-border vehicles
  • Water quality impacts particularly the effects of reclamation and bridge piers on the flushing capacity of Deep Bay, risk of oil/chemical spill on the bridge and sediment plume dispersion during construction
  • Ecological impacts particularly habitat loss, sedimentation and erosion of the intertidal mudflat, and the effect of the bridge on birds including increased bird mortality due to collision with the bridge, bridge lighting effect on birds and barrier effect/physical disturbance to birds caused by the bridge.
  • Environmental Monitoring and Audit (EM&A) particularly joint EM&A efforts with the Mainland

9. These key issues and preliminary EIA findings and recommendations are described below.

Alignment Option Selection

10. The study on potential alignment options for this fourth border-crossing transport corridor in fact began in 1995 with the Crosslinks Study. Two Crosslinks Further Studies (Stage 1 and Stage 2) followed and were completed in early 2001. For the SWC, the Crosslinks Studies spanned 6 years and examined three landing point options in Hong Kong: Ngau Hom Shek, Ha Pak Nai and Tsang Tsui. The Crosslinks Study concluded that the landing of SWC could be between Ngau Hom Shek and Pak Nai. This formed the envelope within which the SWC landing point on Hong Kong side would be selected under this project.

11. Within the above envelope, four landing points were evaluated in this project: Sha Kong Tsuen, Ngau Hom Shek, Ngau Hom Sha and Sheung Pak Nai (as shown in Firgure 2). The Sha Kong Tsuen landing point was discarded mainly because of nearby cultural heritage sites (the Sha Kong Tsuen Archaeological Site and the Sha Kong Miu), being closer to Lau Fau Shan creating potentially higher impacts to oyster farming, and encroachment onto the Hung Shui Kiu New Development Area (NDA).

12. As to the form of the SWC, both bridge and tunnel were considered. Four options were compared and evaluated in this project as follow (as shown in Figure 2 MSWORD):

  • Option A - Bridge Option (Straight Type) Landing at Ngau Hom Shek
  • Option B - Bridge Option (Curved Type) Landing at Ngau Hom Shek
  • Option C - Tunnel Option (Immersed Tube Type) Landing at Ngau Hom Sha
  • Option D - Tunnel Option (Drill and Blast Type) Landing at Sheung Pak Nai

13. The evaluation of these four options and the selection of the preferred option followed a systematic ranking and scoring exercise. In this exercise, weightings were assigned to 21 attributes belonging to 8 categories. Environment was a major category with 9 attributes (such as air quality, noise, ecology, etc), and was assigned 46.9% of the total weighting. After totaling the weighted scores for each option, the bridge options A and B on average scored 62% higher than the tunnel options C and D.

14. Several key factors contributed to the comparatively lower score of the tunnel options. Firstly, these tunnel options would require considerable lengths for them to elevate and connect to Deep Bay Link on the Hong Kong side. For Option C, the landing point would have to be at Ngau Hom Sha traversing through archaeological sites and fishponds in Ngau Hom Shek to allow adequate distance for this climb. Option D would require even a much longer climb since the drill and blast type tunnel would be built much deeper into the bedrock. The landing point for Option D would have to be at Sheung Pak Nai to provide adequate distance to accommodate the elevation needed to connect to Deep Bay Link, traversing through a large area of burial grounds and likely to be objected by local villagers. Secondly, while vehicle emission would be dispersed more or less evenly and continuously along a bridge, such emission in a tunnel would be collected and vented in a concentrated manner at the portal. Both tunnel options would have considerable air quality impacts to residents near the portal. Thirdly, the immersed tube type tunnel in Option C would require extensive dredging and backfilling before and after the placement of the immersed tubes, resulting in adverse impact to the water quality and ecology of Deep Bay. Fourthly, the climb for these tunnels on the Shenzhen side would require extra reclamation from the point where the tunnel emerges from the seabed to the point where it joins the reclamation at Dongjiaotau, forming a finger-like structure extruding into Deep Bay. For the drill and blast type tunnel in Option D, this finger-like extra reclamation could extend up to 1 km into Deep Bay. It would impede the flushing capacity of Deep Bay considerably worse than the reduction in flushing capacity caused by the piers of the bridges in Options A and B. Numerical modeling predicted that the reduction in flushing capacity of the base-case reclamation with this finger-like extra reclamation would be 200% of the reduction in flushing capacity caused by the base-case reclamation with the bridge piers under Options A and B.

15. The results of option evaluation concluded that the bridge options were preferred over the tunnel options. Options A and B had very similar scores. Option B, the curved type bridge, scored slightly higher than Option A, the straight type bridge, for aesthetic and safety reasons. A curved alignment offers varying views to the drivers and passengers and would therefore be more aesthetically pleasing than a straight alignment. Further, experience has shown that driver concentration level tends to drop on long straight roads due to lack of steering required, and a long straight alignment is thus potentially more accident prone compared to a curved alignment of similar length.

16. The Engineering Working Group of this project has endorsed Option B - Bridge Option (Curved Type) Landing at Ngau Hom Shek. Mainland also has accepted the concept of a curved bridge alignment for the SWC. Detailed impact assessment was carried out for Option B in this EIA study.

Air Quality Impact

17. A conservative approach was adopted in assessing vehicle emissions. This approach assumed that all cross-border vehicles travelling on the entire road network to and from Mainland would be fuelled with Mainland fuels containing substantially higher sulphur contents than Hong Kong fuels. Dispersion modeling results showed compliance with the Air Quality Objectives on nitrogen dioxide, respirable suspended particulates, carbon monoxide and sulphur dioxide at all representative sensitive receivers, namely scattered village dwellings, at Ngau Hom Shek.

Water Quality Impact

Flushing Capacity

18. Hydrodynamic and water quality modeling were conducted to assess changes on flushing capacity that would also lead to changes in water quality of Deep Bay due to the reclamation at Dongjiaotou and the bridge piers. The Delft3D suite of models used for simulations incorporated the reclamation at Dongjiaotau and bridge piers. Various scenarios using different sizes and numbers of bridge piers were tested. It may be desirable to have as few bridge piers as possible. Yet, fewer piers would mean larger piers and farther spans between two adjacent piers, and there are engineering constraints as to how big and how far apart they could be, not to mention considerably higher impacts when constructing big piers. Spans of 50m, 75m, 100m and 200m between two adjacent piers were tested. In all cases, the reduction in flushing capacity was found to be less than 1%. The 75m span was found to be most optimal balancing both engineering constraints and environmental benefits and was adopted in the bridge design.

19. Besides the span, other factors that could reduce friction (and therefore minimise reduction in flushing capacity) were also incorporated into the design of the bridge. These included the adoption of submerged (into the mud layer) pile caps and streamline shaped piers. Based on the above design features, reduction in flushing capacity caused by the bridge piers alone was predicted to be 0.76% (seasonal average). The reclamation at Dongjiaotou and the bridge piers together would reduce flushing capacity by approximately 3.3% (seasonal average). Changes in water quality due to these flushing capacity reductions were predicted to range from 0.21% to 2.2%. Such small changes should be deemed insignificant.

20. It should be noted that aerial photographs taken recently over the study area show that the intertidal mudflat in the study area was inundated with oyster beds. These oyster beds obstruct water movement and therefore affect the flushing capacity as well. This project would result in the permanent removal of approximately 10 ha (139m wide and approximately 800m long) of oyster beds along the alignment over the intertidal mudflat. Roughly, this removal of oyster beds could improve the flushing capacity by approximately 0.38%.

Oil and Chemical Spill

21. Oil and chemical spill into Deep Bay due to accidents on the SWC could adversely impact water quality and marine ecology. Car or truck accidents usually involve small quantities of fuel. Gasoline is a volatile product that would be evaporated within short duration. Diesel is more persistent and may require on site cleanup. An emergency response plan would be prepared to aim at reducing the response time and therefore reducing the risk of these petroleum products getting into Deep Bay waters.

Sediment Plume Dispersion during Construction

22. Mitigation measures have been proposed to reduce the dispersion of sediment plumes during the construction phase. A cofferdam would be built to surround the construction area of each bridge pier. Pier construction would be carried out inside the cofferdam, thereby preventing the escape of sediment plumes. Closed grabs would be used to remove sediments inside the cofferdams. Silt curtains would also be deployed as a secondary defense to contain the dispersion of sediment plumes if any.

Ecological Impact

Habitat Loss

23. To understand the ecosystem that could be affected by the SWC, ecological findings from the Crosslinks studies were reviewed to identify data gaps, and to fill these data gaps, ecological surveys under this study commenced in September 2001 and would continue until June 2002. To address some of the above concerns, an extensive search and review of literature on bird collision with structures were conducted, as well as video taping of how birds behave around bridges in Hong Kong and Macau.

24. The loss of intertidal mudflat warrants concern since it is an important feeding ground for birds. Permanent loss due to intertidal mudflat area taken up by 10 pairs of bridge piers was estimated to total 0.00324 ha. Temporary loss during construction was estimated to be approximately 0.7 ha. In the context of the total intertidal mudflat area in Deep Bay on the Hong Kong side, losses of such scale could be deemed insignificant. As mentioned above, aerial photographs taken recently showed that the intertidal mudflat in the study area was inundated with oyster beds. Ecological surveys undertaken in this study found that bird feeding activities on intertidal mudflat with oyster beds were considerably less than similar activities on intertidal mudflat with no oyster bed. Oyster beds therefore impede bird feeding activities. This project would permanently clear all oyster beds underneath the alignment (39m wide) as well as 50m on each side of the alignment over the length of the intertidal mudflat (estimated 800m). This would re-instate over 10 ha of intertidal mudflat that would be more conducive to bird feeding activities compared to the present condition.

25. Other losses would include mangrove and seagrass habitats. Approximately 0.25 ha of mangrove would be lost. This would be compensated by replanting and the loss would therefore be temporary. Seagrasses showed patchy distribution in the study area from the ecological survey data. Loss of patchy seagrass habitat was estimated to be less than 0.001 ha and no replanting was recommended.

Changes in Sedimentation and Erosion Patterns

26. Sedimentation and erosion patterns in Deep Bay due to changes in hydrodynamic conditions with the SWC in place were modeled. Modeling results showed small increases in deposition at Mai Po (+0.5 mm/yr), the Ramsar Site (+0.3 to 0.5 mm/yr) and Tsim Bei Tsui (+0.9 mm/yr) to the northeast, as well as the Pak Nai SSSI (+0.4 mm/yr) to the southwest. The SWC is far from Mai Po and the Ramsar Site in Inner Deep Bay. It is therefore understandable that such increases would be an order of magnitude less than those predicted for the Shenzhen River Regulation Project Stage 1 (+2.5 to 4 mm/yr) and Stage 2 (+4 to 10 mm/yr) at Mai Po/Ramsar Site.

Birds Colliding with the Bridge

27. The perception that birds collide with bridges causing increased mortality was unfounded based on an extensive review of global literature (1,500 references spanning 117 years). Locally and in this study, bird movements in relation to 3 bridges or flyovers in Hong Kong and one in Macau were monitored. Results showed that waterbirds foraged beneath bridge decks and flew either under or over bridge decks. No bird mortality due to collisions with bridges, piers or vehicles using these bridges was recorded. Birds do collide with power lines and guy-wires due to invisibility and die according to literature. However, cables of this cable-stayed bridge would be large diameter structures (e.g. 30cm) that are substantially different and visible at night than the power lines and guy-wires (<1 cm). Bridge cable-stays could also be directly lighted to improve visibililty. Moreover, the cable-stayed portion of the SWC would be over the navigation channels in deep water, away from the intertidal mudflat where high bird activities occur. There will be no cable-stay in the intertidal zone.

Bridge Lighting Effects on Birds

28. Flood lights deployed under adverse weather conditions could confuse migrating birds and increase mortality with certain types of structures other than bridges, according to review of literature on the impacts of various types of lighting on bird mortality. Yet we have adopted precautionary principles and listed design measures to avoid any potential impact. These include no power line to be suspended above the bridge deck, cable-stayed portions to be flood-lit in good weather and to use special lighting such as red-coloured stobe lighting during adverse weather conditions, and to install standard highway lighting on the top of the bridge deck and undersurface of the bridge.

Barrier Effect/Physical Disturbance to Flight and Feeding Behaviour of Birds

29. To investigate whether the presence of a bridge would interfere with or disturb the flight behaviour of birds, 3 vehicular bridges or flyovers in Hong Kong and one in Macau were monitored, including video documentation. These were the Route 3 flyover at Kam Tin River, Shatin Road above Shing Mun River, and Tsing Tsuen Bridge in Hong Kong, plus the Lotus Bridge in Macau. These bridges were chosen because their heights above the water surface (15-17m) are comparable to the SWC. Birds were observed feeding under bridges and flying both beneath and over bridges. Black-faced Spoonbills in Macau were observed to forage and roost near the recently completed Lotus Bridge, and flying over the bridge between intertidal mudflats and freshwater wetlands. These suggest that impacts of bridge operation on disturbing the flight behaviour of birds would be undetectable and should have no impact on the population level.

30. Disturbance to birds feeding on the intertidal mudflat during the construction stage would be a concern. As described above, feeding activities on the intertidal mudflat along the SWC alignment should be low due to the proliferation of oyster beds. Nonetheless, we have proposed to remove Sonneratia, an invading mangrove from the Mainland, from the intertidal mudflat of Deep Bay to restore intertidal mudflat habitat presently occupied by this mangrove.

Environmental Monitoring and Audit (EM&A)

31. An EM&A programme is being developed for both the construction and operational stages of the project. This programme will only cover the EM&A work in Hong Kong. As regards the joint EM&A with Mainland, a communication channel between Mainland and Hong Kong will be established to deal with environmental issues when they arise during the construction stage.

Way Forward

32. We would take into consideration members' views, advice and comments when finalising the EIA Report. We plan to submit the EIA Report under the statutory process in July. Most probably, this Sub-Committee will be consulted under the statutory process in November 2002. We will apply for an environmental permit for the project prior to commencement of construction.

Highways Department
June 2002


Back to topdot_clear.gifTable of Content
User defined date2: 
Friday, 28 April, 2006