5b.1.1.1 This Section presents an assessment of the potential water quality impacts associated with construction and operation of the IWMF at the artificial island near SKC. Recommendations for mitigation measures have been provided, where necessary, to minimize the identified water quality impacts to an acceptable level.
5b.2 Environmental Legislation, Standards and Guidelines
5b.2.1 Environmental Impact Assessment Ordinance (EIAO)
5b.2.1.1 The Technical Memorandum on Environmental Impact Assessment Process (EIAO-TM) is issued by the EPD under Section 16 of the EIAO. It specifies the assessment method and criteria that need to be followed in the EIA. Reference sections in the EIAO-TM provide the details of the assessment criteria and guidelines that are relevant to the water quality impact assessment, including:
· Annex 6 Criteria for Evaluating Water Pollution
· Annex 14 Guidelines for Assessment of Water Pollution
5b.2.2 Water Pollution Control Ordinance (WPCO)
5b.2.2.1
The Water Pollution Control Ordinance (Cap. 358) is the
major legislation relating to the protection and control of water quality in
Table 5b.1
Summary of Water Quality Objectives for
Parameters |
Objectives |
Sub-Zone |
Offensive odour,
tints |
Not to be present |
Whole zone |
Visible foam, oil
scum, litter |
Not to be present |
Whole zone |
Dissolved Oxygen (DO)
within |
Not less than 2.0
mg/l for 90% of samples |
Marine waters |
Depth-averaged
DO |
Not
less than 4.0 mg/l for 90 % of samples |
Marine
waters excepting fish culture subzones |
Not
less than 5.0 mg/l for 90% of samples |
Fish
culture subzones |
|
Not
less than 4.0 mg/l |
Inland
waters |
|
pH |
To be in the range of 6.5 - 8.5, change
due to human activity not to exceed 0.2 |
Marine waters excepting bathing beach
subzones; Mui Wo (A), Mui Wo (B), Mui Wo (C), Mui Wo (E) and Mui Wo (F)
subzones |
To
be in the range of 6.0 – 9.0 |
Mui
Wo (D) sub-zone and other inland waters. |
|
To
be in the range of 6.0 –9.0 for 95% of samples, change due to human activity
not to exceed 0.5 |
Bathing
beach subzones |
|
Salinity |
Change
due to human activity not to exceed 10% of ambient |
Whole
zone |
Temperature |
Change
due to human activity not to exceed |
Whole
zone |
Suspended
Solids (SS) |
Not
to raise the ambient level by 30% caused by human activity |
Marine
waters |
Change
due to waste discharges not to exceed 20 mg/l of annual median |
Mui
Wo (A), Mui Wo (B), Mui Wo (C), Mui Wo (E) and Mui Wo (F) subzones |
|
Change
due to waste discharges not to exceed 25 mg/l of annual median |
Mui
Wo (D) subzone and other inland waters |
|
Unionized
Ammonia (UIA) |
Annual
mean not to exceed 0.021 mg(N)/l as unionized form |
Whole
zone |
Nutrients |
Shall
not cause excessive algal growth |
Marine
waters |
Total
inorganic nitrogen (TIN) |
Annual
mean depth-averaged inorganic nitrogen not to exceed 0.1 mg(N)/l |
Marine
waters |
E. coli |
Not
exceed 610 per 100 ml, calculated as the geometric mean of all samples
collected in one calendar year |
Secondary
contact recreation subzones and fish culture subzones |
Not
exceed 180 per 100 ml, calculated as the geometric mean of all samples
collected from March to October inclusive in 1 calendar year. Samples should
be taken at least 3 times in 1 calendar month at intervals of between 3 and
14 days. |
Bathing
beach subzones |
|
5-Day
Biochemical Oxygen Demand (BOD5) |
Change
due to waste discharges not to exceed 5 mg/l |
Inland
waters |
Chemical
Oxygen Demand (COD) |
Change
due to waste discharges not to exceed 30 mg/l |
Inland
waters |
Dangerous
Substances |
Should
not attain such levels as to produce significant toxic effects in humans,
fish or any other aquatic organisms |
Whole
zone |
Waste
discharges should not cause a risk to any beneficial use of the aquatic
environment |
Whole
zone |
Source: Statement of Water Quality Objectives (Southern Water
Control Zone).
5b.2.3 Sediment Quality Assessment Criteria
5b.2.3.1 Environment, Transport and Works Bureau (ETWB) Technical Circular Works (TCW) No. 34/2002 “Management of Dredged/Excavated Sediment” sets out the procedure for seeking approval to dredge / excavate sediment and the management framework for marine disposal of dredged / excavated sediment. This Technical Circular outlines the requirements to be followed in assessing and classifying the sediment. Sediments are categorized with reference to the Lower Chemical Exceedance Level (LCEL) and Upper Chemical Exceedance Level (UCEL) as follows:
· Category L - Sediment with all contaminant levels not exceeding the LCEL. The material must be dredged, transported and disposed of in a manner that minimizes the loss of contaminants either into solution or by suspension.
· Category M - Sediment with any one or more contaminant levels exceeding the LCEL and none exceeding the UCEL. The material must be dredged and transported with care, and must be effectively isolated from the environment upon final disposal unless appropriate biological tests demonstrate that the material will not adversely affect the marine environment.
· Category H - Sediment with any one or more contaminant levels exceeding the UCEL. The material must be dredged and transported with great care, and must be effectively isolated from the environment upon final disposal.
5b.2.3.2 The sediment quality criteria for the classification of sediment are presented in Table 5b.2.
Table 5b.2 Sediment Quality Criteria for the Classification of Sediment
CONTAMINANTS |
LCEL |
UCEL |
Heavy Metal (mg/kg dry
weight) |
||
Cadmium (Cd) |
1.5 |
4 |
Chromium (Cr) |
80 |
160 |
Copper (Cu) |
65 |
110 |
Mercury (Hg) |
0.5 |
1 |
Nickel (Ni) |
40 |
40 |
Lead (Pb) |
75 |
110 |
Silver (Ag) |
1 |
2 |
Zinc (Zn) |
200 |
270 |
Metalloid (mg/kg dry
weight) |
||
Arsenic |
12 |
42 |
Organic-PAHs (mg/kg dry weight) |
||
PAHs (Low Molecular Weight) |
550 |
3,160 |
PAHs (High Molecular Weight) |
1,700 |
9,600 |
Organic-non-PAHs (mg/kg dry weight) |
|
|
Total PCBs |
23 |
180 |
Source: Appendix A of ETWB TCW No. 34/2002 Management of Dredged /
Excavated Sediment
Note: LCEL – Lower Chemical Exceedance Level
UCEL – Upper Chemical Exceedance Level
5b.3 Description of the Environment
5b.3.1 Existing Baseline Marine Water Quality
5b.3.1.1
Marine water quality monitoring data routinely
collected by EPD were used to establish the baseline condition. The EPD monitoring data collected in 2008 were
summarised in Table
5b.3 for four selected stations close to the Project
site in
Table 5b.3 Baseline
Marine Water Quality Condition for
Parameters |
Lantau Island (South) |
|
WPCO WQO (in marine waters) |
|||
SM12 |
SM13 |
SM17 |
SM6 |
|||
Temperature (℃) |
23.7 (16.7 - 28.6) |
23.9 (17.0 - 29.3) |
23.2 (16.9 - 27.8) |
23.0 (17.2 - 28.2) |
Not more than |
|
Salinity (psu) |
29.9 (23.1 - 33.4) |
29.5 (21.7 - 33.6) |
31.1 (24.9 - 34.0) |
31.8 (28.4 - 34.0) |
Not to cause more than 10% change |
|
Dissolved Oxygen (DO) (mg/L) |
Depth Average |
7.0 (5.4 - 8.6) |
7.2 (5.1 - 8.5) |
6.4 (4.0 - 7.9) |
6.8 (4.6 - 8.2) |
Not less than 4 mg/L for 90% of the samples |
Bottom |
6.5 (4.7 - 7.9) |
6.7 (5.2 - 8.1) |
5.6 (2.1 - 7.8) |
7.2 (6.1 - 8.6) |
Not less than 2 mg/L for 90% of the samples |
|
Dissolved Oxygen (DO) (% Saturation) |
Depth Average |
98 (81 - 129) |
101 (74 - 127) |
89 (59 - 105) |
94 (68 - 106) |
Not Available |
Bottom |
91 (71 - 109) |
94 (76 - 121) |
77 (30 - 102) |
101 (86 - 128) |
Not Available |
|
pH |
8.0 (7.7 - 8.3) |
8.0 (7.7 - 8.4) |
7.9 (7.7 - 8.2) |
8.0 (7.7 - 8.3) |
6.5 - 8.5 (±0.2 from natural range) |
|
Secchi Disc Depth (m) |
2.1 (1.1 - 3.5) |
2.3 (1.5 - 3.6) |
2.5 (1.8 - 4.0) |
2.6 (1.8 - 3.8) |
Not Available |
|
Turbidity (NTU) |
5.0 (1.8 - 9.5) |
4.9 (2.3 - 10.9) |
4.9 (1.9 - 12.8) |
4.1 (0.7 - 11.8) |
Not Available |
|
Suspended Solids (SS) (mg/L) |
6.5 (1.5 - 14.2) |
5.5 (1.7 - 13.3) |
4.6 (1.7 - 11.7) |
4.2 (1.3 - 11.0) |
Not more than 30% increase |
|
5-day Biochemical Oxygen Demand (BOD5)
(mg/L) |
1.1 (0.4 - 2.2) |
1.1 (0.3 - 2.5) |
0.7 (0.5 - 1.3) |
1.1 (0.5 - 2.5) |
Not Available |
|
Ammonia Nitrogen (NH3-N) (mg/L) |
0.070 (0.030 - 0.163) |
0.043 (0.015 - 0.117) |
0.038 (0.010 - 0.140) |
0.028 (0.016 - 0.041) |
Not Available |
|
Unionised Ammonia (UIA) (mg/L) |
0.003 (<0.001 - 0.009) |
0.002 (<0.001 - 0.005) |
0.001 (<0.001 - 0.003) |
0.001 (<0.001 - 0.003) |
Not more than 0.021 mg/L for annual mean |
|
Nitrite Nitrogen (NO2-N) (mg/L) |
0.038 (0.006 - 0.088) |
0.039 (0.006 - 0.091) |
0.029 (<0.002 - 0.074) |
0.024 (0.007 - 0.046) |
Not Available |
|
Nitrate Nitrogen (NO3-N) (mg/L) |
0.178 (0.029 - 0.477) |
0.183 (0.030 - 0.503) |
0.132 (<0.002 - 0.372) |
0.097 (0.005 - 0.300) |
Not Available |
|
Total Inorganic Nitrogen (TIN) (mg/L) |
0.29 (0.07 - 0.61) |
0.27 (0.07 - 0.63) |
0.20 (0.02 - 0.46) |
0.15 (0.03 - 0.36) |
Not more than 0.1 mg/L for
annual mean |
|
Total Kjeldahl Nitrogen TKN) (mg/L) |
0.22 (0.13 - 0.32) |
0.19 (0.12 - 0.33) |
0.17 (0.10 - 0.27) |
0.16 (0.10 - 0.22) |
Not Available |
|
Total Nitrogen (TN) (mg/L) |
0.44 (0.20 - 0.78) |
0.41 (0.21 - 0.78) |
0.33 (0.13 - 0.57) |
0.28 (0.14 - 0.48) |
Not Available |
|
Orthophosphate Phosphorus
(OrthoP) (mg/L) |
0.013 (0.003 - 0.030) |
0.012 (0.003 - 0.039) |
0.012 (0.005 - 0.030) |
0.009 (0.003 - 0.016) |
Not Available |
|
Total Phosphorus (TP) (mg/L) |
0.03 (<0.02 - 0.04) |
0.03 (<0.02 - 0.05) |
0.03 (<0.02 - 0.04) |
0.02 (<0.02 - 0.03) |
Not Available |
|
Silica (as SiO2) (mg/L) |
0.90 (0.21 - 2.00) |
0.87 (0.24 - 2.37) |
0.85 (0.24 - 2.33) |
0.72 (0.17 - 1.67) |
Not Available |
|
Chlorophyll-a (mg/L) |
7.9 (1.3 - 27.3) |
7.0 (1.3 - 31.7) |
3.4 (0.9 - 14.1) |
4.0 (0.5 - 11.3) |
Not Available |
|
E. coli (cfu/100 mL) |
43 (1 - 840) |
2 (<1 - 24) |
1 (<1 - 4) |
2 (<1 - 11) |
Not Available |
|
Faecal Coliforms (cfu/100 mL) |
93 (2 - 2100) |
4 (1 - 55) |
2 (<1 - 5) |
2 (1 - 45) |
Not Available |
Notes: 1. Data
source: Marine Water Quality in
2. Unless otherwise specified, data presented are depth-averaged values calculated by taking the means of three depths: Surface, Mid-depth, Bottom.
3. Data presented are annual arithmetic means of the depth-averaged results except for E. coli and faecal coliforms which are annual geometric means.
4. Data in brackets indicate the ranges.
5b.3.1.2
The Southern WCZ covers an area located to the south of
5b.3.1.3 Although full compliance (100%, including the four selected stations close to the Project site) with the WQOs for unionized ammonia (NH3) and E. coli was achieved in 2010, there was a decrease in TIN compliance rate due to the higher TIN levels recorded mainly in the summer months from June to September. The higher levels of TIN recorded in the southern water during the 2010 summer months could be related to the heavy rainfalls recorded in the Pearl River Delta area in the same period.
5b.3.2 Existing Baseline Marine Sediment Quality
5b.3.2.1
The potential site at SKC and the proposed submarine
cables are located in remote / rural area. The potential for sediment contamination at or
in the vicinity of the Project site is considered low. Sediment quality monitoring data were
routinely collected by EPD in
Table 5b.4 Baseline Marine Sediment Quality Condition
Parameters |
Lantau
Island South |
West
Lamma Channel |
Sediment
Quality Criteria |
||
SM13 |
SM6 |
SM7 |
LCEL |
UCEL |
|
Heavy Metal (mg/kg dry weight) |
|||||
Cadmium (Cd) |
<0.1 (0.1-<0.1) |
<0.1 (0.1-<0.1) |
<0.1 (0.1-<0.1) |
1.5 |
4 |
Chromium (Cr) |
21 (17-23) |
30 (25-37) |
36 (30-47) |
80 |
160 |
Copper (Cu) |
10 (8-13) |
17 (14-20) |
29 (18-38) |
65 |
110 |
Mercury (Hg) |
0.07 (0.05-0.18) |
0.09 (0.07-0.15) |
0.12 (0.08-0.20) |
0.5 |
1 |
Nickel (Ni) |
13 (11-17) |
21 (17-25) |
22 (19-26) |
40 |
40 |
Lead (Pb) |
23 (20-28) |
33 (28-38) |
39 (32-44) |
75 |
110 |
Silver (Ag) |
<0.2 (0.2-<0.2) |
<0.2 (0.2-<0.2) |
0.4 (0.2-0.4) |
1 |
2 |
Zinc (Zn) |
57 (44-73) |
86 (68-100) |
110 (88-140) |
200 |
270 |
Metalloid (mg/kg dry weight) |
|||||
Arsenic |
5.4 (4.4-6.0) |
6.9 (5.8-7.9) |
7.6 (6.1-8.9) |
12 |
42 |
Organic PAHs (mg/kg dry weight) |
|||||
PAHs (Low Molecular Weight) |
90 (90-90) |
93 (90-110) |
92 (90-110) |
550 |
3160 |
PAHs (High Molecular Weight) |
23 (20-29) |
55 (22-120) |
95 (40-160) |
1700 |
9600 |
Organic-non-PAHs (mg/kg dry weight) |
|||||
Total PCBs |
18 (18-18) |
18 (18-18) |
18 (18-18) |
23 |
180 |
Note: LCEL – Lower Chemical Exceedance Level
UCEL – Upper Chemical Exceedance Level
5b.3.2.2 Based on the monitoring data (refer to Table 5b.4), the sediments collected at SM13 (to the northwest of SKC), SM6 and SM7 (in the West Lamma Channel) were uncontaminated materials.
5b.4 Water Sensitive Receivers
5b.4.1.1 Locations of the water sensitive receivers (WSRs) within the Study Area are shown in Figure 5b.1. Water sensitive receivers (WSRs) identified at or in the vicinity of the artificial island near SKC and the submarine cable alignment include:
· Horseshoe Crab;
· Gazetted Beaches;
· Fish Culture Zones; and
· Coral Communities.
5b.4.1.2 According to the recent dive surveys, the seabed along the southwest shoreline of Shek Kwu Chau was found to be mainly composed of bedrocks, boulders, muddy and sandy bottom. Limited marine life was seen except only some coral communities found along the southwest coastline of Shek Kwu Chau. Detailed description and assessment of the impact on marine ecology including coral communities are reported separately under the ecology impact assessment in Section 7b.
5b.4.1.3 No WSD flushing water intake is identified in the vicinity of the Project site.
5b.5.1.1
The
Assessment Area as specified in the EIA Study Brief covers an area within
5b.5.1.2
The
water sensitive receivers that may be affected by various construction
activities for the IWMF were identified.
Potential sources of water quality impact that may arise during the
construction and operation phase of the Project were described. All the identified sources of potential water
quality impact were then evaluated and their impact significance
determined. The need for mitigation
measures to reduce any identified adverse impacts on water quality to
acceptable levels was determined.
5b.6 Identification of Potential Impacts
5b.6.1.1 The major construction works of the Project would be seabed dredging for reclamation and installation of submarine cables, site formation, construction of facilities and construction of the access road. Potential water quality impact during construction phase of the IWMF would be occurred from:
· Drainage and construction site runoff during site formation and foundation piling;
· General construction activities;
· Accidental spillage and accumulation of solid wastes;
· Sewage effluent produced by on-site workforce; and
· Disturbance and re-suspension of seabed sediments for marine works associated with reclamation and installation of submarine cables.
Drainage and Construction Site Runoff
5b.6.1.2 Runoff from the construction works area may contain increased loads of sediments, other suspended solids and contaminants. Potential sources of pollution from site drainage include:
· Runoff and erosion from exposed soil surfaces, earth working areas and stockpiles;
· Release of grounting and cement materials with rain wash;
· Wash water from dust suppression sprays; and
· Fuel and lubricants from maintenance of construction vehicles and mechanical equipment.
5b.6.1.3 Sediment laden runoff during site formation works, if uncontrolled, may carry pollutants (adsorbed onto the particle surfaces) into the nearby coastal waters.
General Construction Activities
5b.6.1.4
Land-based construction works may have the potential to
cause water pollution. Various types of
construction activities would generate wastewater. These include general cleaning and polishing,
dust suppression and utility installation.
These types of wastewater would contain high concentration of suspended
solids. Wastewater would also be generated
from the accumulation of solid waste such as debris, rubbish, plastic package
and construction materials. If
uncontrolled, these would lead to deterioration in water quality.
Accidental Spillage
5b.6.1.5 Variety of chemicals would be used for carrying out construction activities. These chemicals may include petroleum products, spent lubrication oil, grease, mineral oil, solvent and other chemicals. Accidental spillages of chemicals in the works area may contaminate the surface soils. The contaminated soil particles may be washed away by construction site runoff causing water pollution.
Sewage Effluent
5b.6.1.6 Domestic sewage would be generated from the workforce during the construction phase. However, this sewage can be adequately treated by interim sewage treatment facilities, such as portable chemical toilets, which can be installed within the construction site.
Disturbance and Re-suspension of Seabed Sediments
5b.6.1.7 The proposed marine construction works will involve reclamation, construction of breakwaters and anti-scouring protection layer at the southwest coastline of SKC as shown in Figure 5b.2.
5b.6.1.8
The proposed construction method will adopt an approach
where seawalls and breakwaters will first be formed to fully enclose the
reclamation. Containment of fill within
the reclamation area by seawalls and breakwater is proposed, with the seawalls
and breakwaters constructed first (above high water mark) with filling carried
out behind the completed seawalls and breakwater. Under this context, seawall
and breakwater represents the same structure in the construction while differ
only in their functionality. Breakwaters
are built to protect the shore from erosion by wind and wave, thus are built so
that they face outward to the rough sea. In contrast, seawall serves as an
interface between the calm sea and the reclaimed land where coastline
protection is not necessary. Seawalls are built facing sheltered sea where wave
action is expected to be weak. In this sense, the coastline at the northeastern
and northwestern side of the main reclaimed area should be classified as
seawalls, while the rest of the coastline in the site acts as breakwaters. The opening
that needs to be provided for marine access will be shielded by four layers of silt
curtains to control sediment plume dispersion away from the site. The opening
and closing of the silt curtain system would be controlled by the site staff of
the contractor. Barges that need to pass the silt curtain would have to signify
the control staff and wait until permission is granted. The silt curtains would
be fixed at one end on the end of breakwater at marine access opening, while
the movement of the other end would be controlled by appropriate vessels. The
silt curtain system should be kept closed unless passage of vessels or barges
is required. The silt curtain system should
be closed as soon as the barges passes through the marine access opening in
order to minimize the period of curtain opening. The vessels which control the
opening of the silt curtain system would be anchored at the breakwater when the
marine access opening is closed to ensure the silt curtain system could tightly
shield the marine access opening. Filling should only be carried out behind the
silt curtain when the silt curtain is completely closed. This approach was
adopted by the approved Central Reclamation Phase III (CRIII) project as well
as the approved EIA for WDII and CWB as effective measure to control the
dispersion of filling material from the site. The application of silt curtain
at marine access opening is indicatively shown in Appendix 5.5-3.
The silt curtain at marine access opening would be fixed at one end of the end
of the breakwater at the side of the SKC coastline whereas the movement of the
other end at the side of the open sea would be controlled by appropriate vessel(s).
As shown in Table
5b.16, the 90th percentile depth-average
current velocity at the artificial island near SKC would be below
5b.6.1.9
For minimizing the dredging and filling of the overall
reclamation work, the seawall and breakwater at the
perimeter of the reclaimed land is proposed to be in form of a cellular
cofferdam which consists of circular cells connected together to form close
working environment. The use of sheetpiling to aid construction has been
adopted in recent approved EIAs including Hong Kong-Zhuhai-Macao Bridge EIA
and the South Island Line (East) EIA. No significant water quality impact is
expected in this EIA except negligible localized disturbance of bottom
sediment. The disturbed sediment would be highly localized and would be settled
shortly after. To confine any potential loss of fine during the piling work,
floating-type silt curtain would be used to surround the circular cell during the
piling works. Floating-type silt curtain would be applicable at the surrounding
of the circular cells even at the closest location from the coral communities
(which is about
5b.6.1.10 To further minimize the environmental impacts, the overall reclamation works will be implemented in phases as shown in Figure 5b.5 and Plate 5b.4. In phase one, the cofferdam section enclosing the reclamation area would be first constructed. In addition, the section of breakwater preventing the wave from striking the reclamation area directly from the northwest direction would also be constructed. Appropriate measures, such as application of silt curtain around the circular cell when filling, would be also applied to reduce the potential impacts on water quality. Afterwards, the reclamation can be started within the fully formed breakwater and seawall (Phase 2). Appropriate measures, such as silt curtain shielding the marine access opening, would be applied to reduce the potential impacts on water quality. In Phase 3, the remaining breakwater and berth will be constructed while the enclosed area within the cofferdam is being filled and surcharge loading of the reclaimed area is in progress. The detail work phasing at the northwestern seawall of the reclamation area is indicatively shown in Appendix 5.5-3. MSW treatment facilities and the associated supporting facilities will then start to be constructed after the surcharge loading. As described in Section 2, the construction of cofferdam would involve the piling of circular cell of metal sheet into the seabed. No dredging would be required for the installation of these circular cells. As the dredging and filling is bounded behind the sheet piles and cofferdams, it is not likely that suspended solid will be generated and enters the nearby water body. As discussed above, silt curtain would be applied to control the loss of filling material during the filling of circular cell as well as during reclamation. Thus significant loss of fine to the water column is not expected. Band drains may be required to remove excessive water from the reclaimed area. Any surcharge discharge from the band drains, which contains no extra pollutant except suspended fine, could be discharged into the sea via silt removal facilities. Details of the construction phasing are provided in Section 2.
5b.6.1.11
Whereas the installation of submarine cables will
employ subsea burying machine to form narrow cable trench at sea bed up to 5 meter
deep by water jetting and lay the submarine cable spontaneously. The trench will be backfilled at the same time
with the sediments settling to the trench. The trench dimensions will be about
5b.6.1.12
To prevent the tidal action from undermining the breakwaters,
anti-scouring dredging would be required around the breakwaters and
seawalls. A layer of sediment of about
5b.6.2.1
Potential
sources of water quality impacts generated from the operation of the Project
include:
· Wastewater generated from the Waste Treatment Process;
· Sewage generated from floor & vehicle washing;
· Sewage generated from the IWMF staff & visitors;
· Transportation of bottom ash, fly ash and APC residues to WENT Landfill for disposal;
· Discharge of saline water from the proposed desalination plant; and
· Disturbance of seabed due to future maintenance dredging.
5b.6.2.2 Besides, change in coastline configuration as a result of the presence of the proposed breakwaters and reclamation could change of flow regime, water quality and sedimentation pattern in marine water. There will be no impacts to water quality from the operation of the proposed submarine cables.
Wastewater Generated from the Waste Treatment Process
5b.6.2.3 The IWMF Phase I will comprise a 3,000 tpd of moving grate incineration plant and a demonstration scale mechanical treatment plant of about 200 tpd capacity. Desalination plant may also be adopted as a water supply system in the IWMF. Wastewater will be generated from the mechanical treatment plant, the incineration plant and the desalination plant (if adopted) in the IWMF Phase I. No spent cooling water discharge is anticipated from the Project operation.
Mechanical Treatment Plant
5b.6.2.4 In the IWMF, “mechanical treatment + dewatering + post-composting” process is recommended for the mechanical treatment plant. A relatively small amount of wastewater will be generated from the treatment processes.
Incineration Plant
5b.6.2.5 Wastewater will also be generated from various processes throughout the incineration plant including:
5b.6.2.6 Boiler - The practice of continuously removing a small percentage of boiler feed water from the boiler to maintain boiler water chemistry is referred to as boiler blowdown. Although the boiler steam cycle is essentially a closed-loop system, impurities can build up in the boiler which, over time, cause scaling and corrosion of the boiler tubes. These effects eventually lead to boiler tube failure. To reduce such problems, continuous boiler blowdown is employed. The hot blowdown water is passed through a heat exchange to recover heat before becoming a source a plant wastewater. The blowdown water is replaced with contaminant-free feed water make-up.
5b.6.2.7
5b.6.2.8 Boiler Feedwater Treatment System - The purpose of the boiler feedwater treatment system is to provide demineralized water for boiler make-up. Demineralized water is needed in the boiler to prevent scaling and corrosion due to mineral deposits. The treatment system typically involves filtering the feed water to remove suspended solids, and removing metals and minerals in a de-mineralizer. The de-mineralizer contains cation and anion exchangers are periodically regenerated using sulfuric acid and caustic soda respectively. Oxygen is removed from the demineralized water using a deaerator. Processed demineralized water is then stored in tanks and drawn off as needed for boiler feeder water, cooling water and other processes. The operation of the various filters, ion exchangers and deaerators requires periodic back flushing of the system to remove the collected contaminants from the treatment system. This process wastewater is then stored in a neutralization tank where appropriate amounts of acid or caustic are add to adjust the pH.
5b.6.2.10 Miscellaneous Blowdown Sources - Other processes that use process water can be minor sources of process water blowdown, these may include water cooled feed chutes, water cooled bearings, cooling water jacketing etc.
Sewage Generated from Floor & Vehicle Washing
5b.6.2.11
Approximately
Sewage Generated from the IWMF Staff & Visitors
5b.6.2.12
The sewage generated from human activities in the IWMF
would include the sewage from the IWMF staff and visitors, as well as the
sewage generated from the canteen, and community facilities. It is estimated
that approximate
Table 5b.5 Estimated Amount of Sewage Generated from the IWMF Staff & Visitors and the Associated Activities
Items |
No. of Employee or Visitor |
Unit Flow Factor (1) (m3/d/person) |
Flow (m3/d) |
Staff and Visitors |
|||
Staff of incineration plant and MT plant |
200 |
0.08 |
16.00 |
Staff of canteen |
25 |
1.50 |
37.50 |
Staff of Community facilities |
20 |
0.35 |
7.00 |
Visitors |
450 |
0.06 |
27.00 |
Sub-total |
87.50 |
||
10% Contingency |
8.75 |
||
Total |
96.25 |
Note (1): The unit
flow factors adopted to estimate the sewage flow generated from the staff and
visitors in the IWMF are primarily based on the guidelines laid down in EPD’s
Guidelines for Estimating Sewage Flows for Sewage Infrastructure Planning
Version 1.0 (GESF).
Reuse and Treatment of Wastewater Generated from Waste Treatment Process and Sewage from Floor & Vehicle Washing and the IWMF Staff & Visitors
5b.6.2.13 Table 5b.6 shows the estimated quantity and possible characteristic of wastewater generated from treatment process and sewage from floor & vehicle washing and the IWMF staff & visitors.
Table 5b.6 Estimated Quantity and Possible Characteristics of Wastewater Generated from Treatment Process and Sewage from Floor & Vehicle Washing and the IWMF Staff & Visitors
|
Flow (m3/d) |
pH |
BOD (mg/L) |
COD (mg/L) |
SS (mg/L) |
Temp. (℃) |
Chloride (mg/L) |
||
Incineration
Plant |
Miscellaneous Blowdown
Sources |
1 |
6-8 |
50 |
30 |
50 |
20 |
|
|
Boiler Feedwater Treatment
System (Demineralizer Drain) |
30 |
9-11 |
- |
- |
20 |
20 |
3,000 |
||
Pump Leak Water |
10 |
6-8 |
- |
- |
50 |
20 |
50 |
||
Boiler and Evaporative |
80 |
10-12 |
- |
- |
50 |
50 |
50 |
||
Mechanical
Treatment Plant Drain |
5 |
6.5 - 9 |
6,000 – 8,500 |
20,000 - 25,000 |
9,000 - 10,000 |
- |
- |
||
Sewage from
Floor and Vehicle Washing |
Floor Washed Drain |
1 |
7-9 |
50 |
30 |
500 |
20 |
100 |
|
Vehicle Washed Drain |
30 |
6-8 |
300 |
200 |
500 |
20 |
100 |
||
Sewage from the IWMF Staff
& Visitors |
88 |
6-8 |
250 |
520 |
250 |
20 |
- |
||
5b.6.2.14 Generally, wastewater shown in Table 5b.6 can be categorized into two types including high organic loading wastewater and low/nil organic loading wastewater. High organic loading wastewater, such as sewage from floor & vehicle washing and the IWMF staff & visitors, will be treated by secondary wastewater treatment plant provided on-site to remove the organic pollutants for reuse on-site (see Section 5b.6.2.15 below). The bunker leachate / ash leachate from incineration plant (as described in Section 5b.6.2.9) would be highly polluted and would be conveyed to the incineration plant and co-incinerated with MSW and is therefore not included in Table 5b.6. On the other hand, low/nil organic loading wastewater, mainly coming from plant machinery such as dimineralizer drain, contains only trace amount of or no organic pollutants. It only requires simple treatment such as sedimentation or neutralization or even not requires any treatment before being used for flue gas cooling in quench tower or ash quenching. As the artificial island near SKC would be built on reclaimed land in rural area, it is envisioned that the IWMF Phase I would be designed with a net zero discharge of process and sanitary wastewater.
5b.6.2.15 A wastewater treatment plant would be provided on-site to provide treatment to some wastewater generated from the IWMF (such as sewage from the IWMF staff & visitors) for reuse in the incineration plant and the mechanical treatment plant or for washdown and landscape irrigation in the IWMF site following the effluent qualities shown below. The following recommended effluent qualities for reuse purposes are based on the “Guidelines for Water Reuse” published by the USEPA.
· pH : 6 – 8
· BOD : 10 mg/L
· Turbidity : 2 NTU
· Total Coliform/100 mL : non-detectable
· Cl2 residual : 1 mg/L
5b.6.2.16
Because of the compacted area in SKC, membrane
bioreactor (MBR), which requires small footprint, is proposed for the IWMF for
mainly human sewage treatment. Based on
the above effluent standards and wastewater characteristics and quantity shown
in Table
5b.6, the wastewater treatment facilities enclosed by
the reinforced concrete structure under the reception hall of the incineration
plant would occupy an area of about
Desalination Plant
5b.6.2.17
If desalination plant is adopted as a water supply
system in the IWMF, the brine water generated would be either discharged back
to the sea where the seawater is collected for desalination or reused for ash
quenching. The brine water drained from
the desalination plant is concentrated seawater (about 1.7-1.8 time more
concentrated than the raw seawater). The
design flow of the desalination plant, if required will be about
Maintenance Dredging
5b.6.2.18 Maintenance dredging maybe required to provide sufficient clearance between the boat and the seabed for safe marine traffic within the marine embayment. Currently, there is no plan for regular maintenance dredging. The need of maintenance dredging would depend on the exact sedimentation and scouring condition after the completion of the Project. Assessment on the water quality impact has been described in Section 5b.7.
5b.7 Prediction and Evaluation of Environmental Impacts
5b.7.1 Land-based Construction Phase Impact
Drainage and Construction Site Runoff
5b.7.1.1
Runoff from the construction works area may contain
increased loads of sediments, other suspended solids and contaminants. As a good site practice, mitigation measures
should be implemented to control construction site runoff and drainage from the
works areas, and to prevent runoff and drainage water with high levels of suspended
solids from entering the nearby water bodies.
It is estimated that the volume of runoff generated on site would be
about
General Construction Activities
5b.7.1.2 Land-based construction activities may generate wastewater and cause water pollution. Their impacts are likely to be minimal, provided that good construction practices and proper site management would be observed. Effluent discharge from temporary site facilities should be controlled to prevent direct discharge to the neighbouring water environment. It is anticipated that water quality impacts caused by general construction activities would be insignificant with adequate implementation of recommended mitigation measures as described in Sections 5b.8.1.2 and 5b.8.1.3.
Accidental Spillage
5b.7.1.3 Site drainage should be well maintained and good construction practices should be observed to ensure that oil, fuels and solvents are managed, stored and handled properly and do not enter the nearby water streams. No adverse water quality impacts are expected with proper implementation of the recommended mitigation measures (refer to Sections 5b.8.1.4 to 5b.8.1.7).
Sewage Effluent
5b.7.1.4 Domestic sewage would be generated from the workforce during the construction phase. However, this sewage can be adequately treated by interim sewage treatment facilities, such as portable chemical toilets, which can be installed within the construction site. It is unlikely that sewage generated from the site would have a significant water quality impact, provided that sewage is not discharged directly to the water environment, and chemical toilets are used and properly maintained.
5b.7.2 Construction Phase Impact from Cofferdam Construction
5b.7.2.1
The cofferdam construction would be adopted for
construction of breakwater under this Project. The cofferdam construction would
involve piling as well as material filling. Water quality impact may arise
during the piling works as well as the filling of circular cells. With
reference to the recent EIAs such as Hong Kong-Zhuhai-
Ambient and Tolerance Values
5b.7.3.1 The sediment plumes passing over a sensitive receiver will cause the ambient suspended solids concentrations to be elevated. The level of elevation will determine whether the impact is adverse. The determination of the acceptability of elevations in suspended solids (SS) concentrations is based on the Water Quality Objectives (WQO). The WQO for SS is defined as being an allowable elevation of 30% above the background. It is proposed to represent the ambient SS value by the 90th percentile of SS concentrations measured under the EPD routine marine water quality monitoring programme at the stations (namely SM12 and SM13) nearest to the WSRs that would be potentially affected by the localized dredging / filling works (including the horseshoe crab at northern SKC, over 700 coral communities at southwest SKC, the gazetted beaches at southern Lantau Island and east Cheung Chau as well as the fish culture zone at Cheung Sha Wan on the eastern coastline of Lantau Island) as shown in Figure 5b.1. The relevant EPD data and allowable elevations in suspended sediment concentration are summarized in Table 5b.7. The 90th percentile SS values presented in Table 5b.7 were calculated based on the EPD monitoring data collected in the period from 2007 to 2010.
Table 5b.7 Ambient and Tolerance Values for Suspended Solids Concentrations in the Vicinity of Sensitive Receivers
Sensitive Receiver
(Relevant EPD Monitoring Station) |
Dry Season |
Wet Season |
||
90th
Percentile |
30%
Tolerance |
90th
Percentile |
30%
Tolerance |
|
Horseshoe Crab at northern SKC, Coral Communities at Southwest
Coastline of SKC, Gazetted beaches at southern Lantau Island (SM13) |
16.3 mg/L |
4.9 mg/L |
8.2 mg/L |
2.5 mg/L |
Gazetted beaches at east Cheung Chau and Cheung Sha Wan fish culture
zone (SM12) |
20.9 mg/L |
6.3 mg/L |
11.6 mg/L |
3.5 mg/L |
5b.7.3.2 The allowable elevation in SS concentration as defined by the WQO for a particular site corresponds to the 30% tolerance level. The calculated maximum SS concentrations from the filling and dredging have been compared with the 30% tolerance values in the above table to determine the acceptability of the impacts. To represent the long term environmental baseline for suspended solid near the Project site, the SS tolerant values are calculated from data of 2007 to 2010. The use of the data from 2007 to 2010 would be conservative as the minimum tolerant values calculated (2.5 mg/L) would be lower than the value calculated from all data since 1999 (3.1 mg/L).
Discussion of the Potential Sediment Plume Impacts
5b.7.3.3
The potential site at SKC is located in the open water
in the southern waters of
Plate 5b.1 Marine Chart for the Study Area
5b.7.3.4 The WSRs closest to the proposed filling and dredging works at SKC include the horseshoe crab at northern SKC, coral communities at southwestern SKC, the gazetted beaches at southern Lantau Island and east Cheung Chau as well as the fish culture zone at Cheung Sha Wan on the eastern coastline of Lantau Island (as shown in Figure 5b.1). Besides the coral communities identified at the southwest coastline of SKC, all the other WSRs are considered to have a low potential to be significantly influenced by the dredging work as these WSRs are not located along the key transportation paths of any sediment plume formed during the dredging work at SKC. The typical flow patterns around the Study Area, which is obtained from the Delft3D Regional Update model developed under Agreement No. CE42/97, are shown in Plate 5b.2 and Plate 5b.3 below. The arrows shown in Plate 5b.2 and Plate 5b.3 represent the water flow directions during typical ebb and flood tide respectively.
Plate 5b.2
Tidal Current Pattern during Typical Ebb Tide
Plate 5b.3
Tidal Current Pattern during Typical Flood Tide
5b.7.3.5
As shown in Plate
5b.2 and Plate
5b.3, the fish culture zone at Cheung Sha Wan and the
gazetted beaches at east Cheung Chau are largely influenced by the tidal flow
discharged from the Kap Shui Mun near Ma Wan during ebb tides as well as by the
offshore waters from the open
Near Field Sediment Dispersion Modelling
5b.7.3.6
The method of calculation of the near field
concentrations of suspended sediment plumes is the same as that used in the
approved EIA study for Outlying Islands Sewerage Stage 1, Phase II Package J –
Sok Kwu Wan Sewage Collection, Treatment & Disposal Facilities ([1]).
In this method, a simple model is used
to calculate the depth averaged suspended sediment concentrations along the
centreline of a plume by solving the advection-diffusion equation for a continuous
line source ([2]).
This model is considered appropriate for
the calculation of suspended sediment concentrations from the proposed dredging
/ filling work because the equation is based on a continuous line source of
sediment, which is a reasonable approximation of the loss of sediment due to
suspension during grab dredging / filling. It is appropriate for areas where the tidal
current is uni-directional for each phase of the tidal cycle (i.e. the ebb and
flood phases), which is the case at SKC where the tidal current is also
uni-directional for each phase of the tidal cycle. This method is applicable for suspended
sediment plumes of length no greater than the maximum tidal excursion. The sediment plume generated from the proposed
dredging / filling work at SKC would be transported along the tidal flow around
the southwest coastline of SKC which generally runs from the southeast to the
west and from the west to the southeast during flood and ebb tides
respectively. The maximum depth-averaged
tidal current speeds around SKC could be up to
Tidal Excursion = Maximum Tidal
Current Speed * Period * 2 / π
5b.7.3.7
The tidal excursion is thus calculated to be
approximately
where C(x) = concentration at distance x from the source;
q = sediment loss rate (
D = water depth (
x = distance from source;
ω = diffusion velocity (
5b.7.3.8
Any sediment plume generated from the proposed dredging
/ filling works at SKC would be transported by the tidal flow around the SKC. The representative water depth at the reclamation
area of the artificial island near SKC would be over
5b.7.3.9 The use of the above equation is limited to situations where the value of γ, as defined by the following equation, is small and where ω / u is also small.
γ = W *
t / D
where W = settling velocity of suspended sediment;
t = time;
D = water depth (
5b.7.3.10
The sediments suspended by the dredging / filling
operations may be split into a fine fraction and a coarse fraction. The fine fraction is assumed to remain in
suspension indefinitely, which is based on the fact that the settling velocity
for the sediment particles according to Stokes Law is offset by local
turbulence. The value of settling
velocity, W, for the coarse fraction of the sediment (based on the
Stokes Law) would depend on the sediment particle size. The value for t will be taken to be
half of the tidal period, which may be taken to be the time between the ebb and
flood phases of the tidal cycle. In
5b.7.3.11
The average current speed in the vicinity of the
artificial island near SKC is conservatively taken to be
Prediction and Evaluation of Impacts from Filling for the Reclamation Area
Consideration of Mitigation Measures and Sediment Loss Rate
5b.7.3.12
The filling of reclamation area will be conducted at a production
rate of
5b.7.3.13
Based
on the approved EIA for “Further Development of Tseung Kwan O
Feasibility Study” as well as “Tuen Mun – Chek Lap
Kok Link - Investigation”, the loss rate during sand filling would be about
5% of fine content from the filling material. The fine content of sand is
assumed to be 5% of the bulk under the above 2 studies. Since the filling
material used under this Project would be consisted of rock, sand and public
fill, the fine content of the composite filling material could not be simply
determined. As the fine content for the rock (negligible fine content) and sand
(about 5% fine content ([3])) is expected to be lower than that
of public fill (25% fine content ([4])), it would be conservative to
assume the filling material to be consisted of entirely public fill. It is
assumed under this assessment that 5% of the fine content in the public fill
would be lost during the filing operation, which is based on the assumption
adopted under the approved EIA for “Hong
Kong - Zhuhai - Macao Bridge Hong Kong Boundary Crossing Facilities”. The density of public fill adopted
under this assessment is
5b.7.3.15
The current speeds around the Project site are less
than
Model Results
5b.7.3.16 The results of the calculation of suspended solid concentrations under unmitigated filling scenario are given in Table 5b.8.
Table 5b.8 Calculated Suspended Sediment Elevation under Unmitigated Filling Scenario
Distance from Source (m) |
Suspended Sediment Concentration (mg/L) |
50 |
434.2 |
100 |
217.1 |
200 |
108.6 |
250 |
86.8 |
300 |
72.4 |
400 |
54.3 |
500 |
43.4 |
600 |
36.2 |
700 |
31.0 |
800 |
27.1 |
900 |
24.1 |
Note: The assessment result is calculated based on the following assumptions:
1. the filling material would be consisted of 100% public fill;
2.
the fine
content for public fill is assumed to be 25% and 5% of the fine content would
be loss to the water column during the filling process. The density of public
fill is assumed to be
3.
no silt
curtain would be applied at the marine access opening.
Coral Communities
5b.7.3.17
As shown in Table
5b.8, the level of SS elevation due to the reclamation
filling at the closest coral community (minimum distance from the dredging
boundary to the nearest non-translocatable coral community is
5b.7.3.18 Moving the artificial island near SKC further south (to reduce the level of impact from the proposed marine works) was considered not feasible due to marine safety reason as explained in Section 2. For mitigating the impact on SS elevation on coral, it is proposed that a silt curtains system should be applied at the marine access opening to control the dispersion of filling material. As discussed in Section 5b.7.3.14, the assumed silt removal efficiency for silt curtains system at marine access opening adopted under this assessment is 80%. During the initial period of reclamation, this silt-removal efficiency of the silt curtains system at marine access opening shall be verified by examining the results of water quality monitoring points under the EM&A works. The verified silt curtain effectiveness shall be used for future reference only. The detailed methodology for the field trial of silt curtain system at marine access opening for verification of the silt removal efficiency should be submitted to EPD for approval prior to the trial. The need to implement additional mitigation measures shall be determined in accordance with the event/action plan in the EM&A Manual if there is any exceedance of the water quality identified in the monitoring results under the EM&A programme.
5b.7.3.19
To further reduce the elevation in SS due to the
reclamation filling, the use of public fill for any filling works below +2.5
mPD (ie below high watermark) should be limited as the fine content in public
fill would generally be higher than that of sand and rock. The proposed
composition of the filling material used for reclamation filling at different
locations is summarized in Table
5b.9. It is proposed that the use of public fill below
+2.5 mPD should be controlled at maximum
Plate
5b.4
Different Filling Zones and Distance from nearest Coral
Communities
50
m 200 m 200 m
Table 5b.9 Proposed Composition of Filling Material used for Filling at Different Locations
Area |
Area Code |
Filling Rate (m3/d) |
||
Public Fill |
Sand |
Rock |
||
Reclamation area between 250m and 400m away
from the nearest coral community (or between 50m and 200m away from opening
for marine access) (see Plate
5b.4) |
A |
300 |
4,000 |
2,700 |
Reclamation area more than 400m away from the
nearest coral community (or more than 200 away from opening for marine
access) (see Plate
5b.4) |
B |
1,000 |
3,300 |
2,700 |
Filling of the reclamation area which has
been completely confined by cellular cofferdam (i.e. the opening for marine
access has been closed) |
- |
7,000 |
0 |
0 |
Table 5b.10 Calculated Suspended Sediment Elevation under Mitigated Filling Scenario
Distance from Source (m) |
Suspended Sediment Concentration contributed
from: (mg/L) |
||||
Public Fill |
Sand |
Rock |
Total |
||
For Area A |
|
|
|
7000 m3/day |
|
50 |
8.4 |
3.7 |
0.0 |
12.1 |
|
100 |
4.2 |
1.9 |
0.0 |
6.0 |
|
200 |
2.1 |
0.9 |
0.0 |
3.0 |
|
242 |
1.7 |
0.8 |
0.0 |
2.5 |
|
250 |
1.7 |
0.7 |
0.0 |
2.4 |
|
300 |
1.4 |
0.6 |
0.0 |
2.0 |
|
400 |
1.0 |
0.5 |
0.0 |
1.5 |
|
500 |
0.8 |
0.4 |
0.0 |
1.2 |
|
600 |
0.7 |
0.3 |
0.0 |
1.0 |
|
700 |
0.6 |
0.3 |
0.0 |
0.9 |
|
800 |
0.5 |
0.2 |
0.0 |
0.8 |
|
900 |
0.5 |
0.2 |
0.0 |
0.7 |
|
For Area B |
1,000 m3/day |
3,300 m3/day |
2,700 m3/day |
7000 m3/day |
|
50 |
6.9 |
12.4 |
0.0 |
19.3 |
|
100 |
3.4 |
6.2 |
0.0 |
9.7 |
|
200 |
1.7 |
3.1 |
0.0 |
4.8 |
|
250 |
1.4 |
2.5 |
0.0 |
3.9 |
|
300 |
1.1 |
2.1 |
0.0 |
3.2 |
|
400 |
0.9 |
1.6 |
0.0 |
2.4 |
|
500 |
0.7 |
1.2 |
0.0 |
1.9 |
|
600 |
0.6 |
1.0 |
0.0 |
1.6 |
|
700 |
0.5 |
0.9 |
0.0 |
1.4 |
|
800 |
0.4 |
0.8 |
0.0 |
1.2 |
|
900 |
0.4 |
0.7 |
0.0 |
1.1 |
|
Note: The assessment result is calculated based on the following assumptions:
1. the filling material would be composed of public fill, sand and rock in the amount stated in the table;
2.
the fine
content for public fill is assumed to be 25% and 5% of the fine content would
be loss to the water column during the filling process. The density of public
fill is assumed to be
3.
the fine
content for sand is assumed to be 5% and 5% of the fine content would be loss
to the water column during the filling process. The density of sand is assumed
to be
4. the filling of rock do not contribute to any loss of fine during the filling process;
5.
silt
curtain would be applied at the marine access opening and its silt removal
efficiency would be 80%.
5b.7.3.20
As shown in Table
5b.10, the predicted maximum SS elevation at
5b.7.3.21 To ensure the actual efficiency of the silt curtains system would be at least as high as the level assumed in the assessment, a field trial should be carried out at the early stage of the reclamation to determine the silt removal efficiency of the silt curtains system (with total four layers of silt curtains). The silt curtain at the marine access opening is indicatively illustrated at Appendix 5.5-3 and Appendix 5.5-4. The details for the field trial are provided in the EM&A manual.
Prediction and Evaluation of Impacts from Localized Dredging for Anti-Scouring Protection Layer
Consideration of Mitigation Measures and Sediment Loss Rate
5b.7.3.22
The localized dredging works for anti-scouring
protection layer for seawalls and breakwaters at SKC will be conducted at a production
rate of
5b.7.3.23
Based
on the approved EIA for “Construction of
an International Theme Park in Penny's Bay of North Lantau together with its
Essential Associated Infrastructures (Theme Park)”, the sediment loss rate
from grab dredging in areas with significant amount of debris or big boulders
on the seabed would be
5b.7.3.25
The current speeds around the Project site are less
than
Model Result
5b.7.3.26 The results of the calculation of suspended sediment concentrations are given in Table 5b.11.
Table 5b.11 Calculated Suspended Sediment Elevation under Unmitigated Dredging Scenario (Silt Curtain Implemented)
Distance
from Source (m) |
Suspended
Sediment Concentration (mg/L) |
15 |
174 |
50 |
52.2 |
100 |
26.1 |
200 |
13.1 |
300 |
8.7 |
400 |
6.5 |
500 |
5.2 |
600 |
4.4 |
700 |
3.7 |
800 |
3.3 |
900 |
2.9 |
Coral Communities
5b.7.3.27
As shown in Table
5b.11, the level of SS elevation due to the dredging
operation at the closest coral community (minimum distance from the dredging
boundary to the nearest non-translocatable coral community is >
5b.7.3.28
For mitigating the impact on SS elevation on coral, no
dredging operation would be carried out within
Table 5b.12 Calculated Suspended Sediment Elevation under Mitigated Dredging Scenario (Silt Curtain Implemented and Dredging Rate Reduces)
Distance
from the Nearest Coral (m) |
Maximum
Allowable Production Rate (m3/day) |
Maximum
SS Elevation (mg/L) |
Above 100 |
380 |
2.5 |
5b.7.3.29
Table
5b.12 below summarizes the recommended dredging rate
for this Project. The permitted number
of grab per hour is also shown in Table
5b.13, which should be specified in the construction
contract for this Project to be followed by the dredging contractor. It is recommended to employ closed grab with
small capacity of
Table 5b.13 Recommended
Maximum Dredging Rate and Permitted Number of Grab Per Hour
Distance
from the Nearest Coral (m) |
Recommended Dredging Rate (m3 per day) |
Permitted No. of Grab per Hour Using Grab Size of Approx. |
Above 100 |
380 |
No
more than 15 |
5b.7.3.30
As shown in Table
5b.12, after the reduction of dredging rate, the
resulted maximum SS elevation would greatly reduced to 2.5 mg/L, which complied
with the SS criterion of elevation from 30% of the ambient SS. Since the
maximum SS elevation is calculated from the shortest distance (
Sediment Elevation and Sedimentation Impact upon Coral Communities
5b.7.3.31
As shown in Table
5b.8 and Table
5b.12, full compliance with the SS criterion is
predicted at all the identified coral communities. As compared with the baseline SS level at SKC
(11.6 mg/L and 8.6 mg/L for SM12 and SM13 respectively in 2008), the maximum
elevation in SS level (2.5 mg/L) only contribute to a small increment. As shown
in the Appendix 5.1, the baseline sedimentation rate at SKC
would be
5b.7.3.32 As discussed in Section 5b.7.3.2, the maximum SS elevation predicted at the nearest coral community would be 2.5 mg/L, which would fulfills the criteria for SS elevation. It is therefore believed that no adverse SS impact on the coral community at Southwest Coastline of Shek Kwu Chau would be anticipated. Detailed assessment of the effect of SS upon these coral communities and further mitigation measures recommended for protection of these coral communities are separately provided under the ecological impact assessment in Section 7b.
Horseshoe
Crab at
5b.7.3.33
Horseshoe crab was identified at northern SKC about
Other Sensitive Receivers
5b.7.3.34
Besides the coral communities and horseshoe crab, the
nearest WSRs (including the gazetted beaches at southern
Oxygen Depletion from Dredging
5b.7.3.35 The sediment oxygen demand (SOD) of the sediment samples collected for biogas risk investigation has been used to determine the reductions in dissolved oxygen (DO) concentration, based on the predicted increases in suspended sediment concentrations in accordance with the following equation:
where DODEP = Dissolved oxygen (DO) depletion
(mg/l)
C = Predicted maximum suspended
solids (SS) concentration (mg/l)
SOD = Sediment oxygen demand
(mg/kg) measured in the sediment samples collected from marine SI
K = Daily
oxygen uptake factor (set as 1)
5b.7.3.36 In the calculation, the daily oxygen uptake factor, K, was set to be 1, which means instantaneous oxidation of the sediment oxygen demand. This was a conservative prediction of DO depletion since oxygen depletion is not instantaneous. It is worth noting that the above equation does not account for re-aeration which would tend to reduce impacts of the SS on the DO concentrations in the water column.
5b.7.3.37 The calculation was performed using the highest levels of sediment oxygen demand (SOD) measured in the sediment samples collected during the sediment sampling for conservative predictions. The highest SOD level (575 mg/kg) was recorded at Station MI11. Locations of the sampling stations are given in Figure 6b.1.
5b.7.3.38
The
predicted maximum DO depletion during dredging was used to evaluate the water quality impacts. The calculated maximum DO depletion was
subtracted from the measured background DO level to determine the resultant DO
level in marine water. The 10 percentile
values of the measured DO levels were used as the background levels, following
the approach adopted in the approved EIAs for “Dredging Works for Proposed Cruise Terminal at Kai Tak” and “Wan Chai Development Phase II &
Central-Wan Chai Bypass”. The
proposed analysis, which is on the conservative side, will likely overestimate
the impact on DO. The predicted maximum DO depletions are given
in Table 5b.14.
Table 5b.14 Calculation
of the Effects of Increased Suspended Solids
Concentration on Dissolved Oxygen Concentration under Unmitigated Scenario
Maximum
SS Elevation at Important Ecological Sites (mg/L) |
SOD
in Sediment (mg/kg) |
Maximum DO
Depletion (mg/L) |
Background
Depth-averaged DO (mg/L) |
Resultant
DO (mg/L) |
WQO
for Depth-averaged DO |
174 |
575 |
0.1 |
5.90 |
5.8 |
≥4 mg/l |
5b.7.3.39 No significant DO depletion was predicted at the sensitive receivers in SKC even under unmitigated scenario. The dredging activities would cause a maximum DO depletion of 0.1 mg/L at the nearest sensitive receiver. Full compliance with the WQO for depth-averaged DO was predicted in SKC. Hence, no adverse impacts at the sensitive receivers on the DO levels in SKC would be expected from the dredging works.
5b.7.4 Construction Phase Impact from Installation of Submarine Cables
5b.7.4.1 The cable burying machine would include an injector lowered to the seabed. The injector fluidizes a trench using high pressure water jets and the cable is immediately laid within the trench. The sides of the trench slip around the cable, burying it and leaving a small depression in the seabed. The maximum width of the seabed fluidized by the injector is 3 m and the cable is buried to a maximum depth of 5 m. During the jet plough cable laying process the seabed sediments will be disturbed and a small percentage will be lost to suspension in the lower part of the water column in the immediate vicinity of the jet plough. The analysis of the potential transport of fine sediments suspended into the water column during the cable laying process has been conducted based on the assumptions and methodology adopted under past approved studies for similar cable laying works such as the “VSNL Intra Asia Submarine Cable System – Deep Water Bay (EP-294/2007)” and the “Proposed 132kV Submarine Cable Route for Airport "A" to Castle Peak Power Station Cable Circuit, CLP Power (EP 267/2007)”.
5b.7.4.2
During cable laying, the seabed sediment will be
released at the bottom of the water column which will result in high localized
suspended sediment concentrations and high settling velocities. This is because at high concentrations within
a much localized area, suspended sediment will tend to form large aggregations
of sediment particles (the process of flocculation) which have a higher
settling velocity than the individual sediment particles. It is expected that the suspended sediments
will remain within
5b.7.4.3
Typically the settling velocity of SS is determined by
examining the relationship between SS initial concentrations and the cohesive
nature of the sediment being disturbed. Typically,
as SS concentration increases, so will settling velocity as sediment particles
flocculate, gain mass and settle faster. However, this relationship does not hold true
when initial SS concentrations exceed values such as
Settling
Time =
5b.7.4.4 The distance travelled by the sediment will thus be the settling time multiplied by the current velocity.
Distance
Travelled = 200
s x
5b.7.4.5
The above calculation indicated that the sediments
disturbed during laying of the submarine cable will settle onto the seabed
within approximately
5b.7.4.6 As shown in Figure 5b.3, the most of the influence zone of the laying works would be far away from the Shek Kwu Chau site. The laying of submarine cables in this area would not cause cumulative water quality impact with the dredging for sheet piling at the Shek Kwu Chau site. No concurrence works between laying of cables and dredging/reclamation works within the same location is allowed. For works close to each other, the construction program should be arranged so that the dredging/reclamation works within area bounded by the breakwaters and the laying of cables would not operate within a distance of 80m from each other to avoid any accumulative impact on the environment (in case if such tight schedule is necessary).
5b.7.4.7 During submarine cable laying works, a water quality monitoring programme should be conducted to ensure no unacceptable water quality impacts will occur at the nearby WSRs.
5b.7.5 Consideration of Cumulative Construction Phase Impacts
5b.7.5.1 Two projects may be carried out concurrently with the IWMF Project at the artificial island near SKC:
ESB-209/2009 – Outlying
Marine works
5b.7.5.2
The concurrent project would involve the
construction of a Sewage Treatment Works (STW) at San Shek Wan. Although the STW building would be located
outside the study area of the IWMF at Cheung Sha, nevertheless, its
associated submarine outfall, which is approximately
5b.7.5.3
While the tentative construction schedule for the IWMF
project is 2013 to 2018, there may be an overlapping period for the two
projects. As the water quality impacts
generated from the proposed reclamation and submarine cable installation works
are predicted to be localized and will not encroach any WSRs, no significant
cumulative impact on marine ecology due to water quality would be anticipated.
The location of the proposed outfall of the Outlying Islands Sewerage Project
is shown in Figure 5b.7.
The distance of the proposed outfall
from the IWMF site is approximately
ESB-187/2008 – Improvement of Fresh Water Supply to Cheung Chau (2010 – 2013)
Marine works
5b.7.5.4
This concurrent project would construct a submarine
water main across Adamasta Channel, between Northern Channel of Cheung Chau and
5b.7.5.5
According to the tentative schedule of the concurrent
project, which is 2010 to 2013, the submarine water main laying works may
overlap with the construction works for the IWMF Project (2013-2018). Considering that the overlapping time would
be short, and that the water quality impacts generated from the IWMF marine
works are predicted to be localized and will not encroach any WSRs, no
significant cumulative impact on marine ecology due to water quality would be
anticipated. The location of the new water main of the Fresh Water Supply
Improvement Project is shown in Figure 5b.7. The distance
of the proposed submarine water main is at least
5b.7.5.6
Other concurrent projects, such as the Tuen Mun-Chek
Lap Kok Link,
5b.7.6 Operation Phase Impact from Project Effluent
Wastewater from Waste Treatment Process and Sewage from Floor & Vehicle Washing and the IWMF Staff & Visitors
5b.7.6.1 As discussed in Section 5b.6.2.14, the IWMF facilities would be designed with a net zero discharge of process and sanitary wastewater. A wastewater treatment plant would be provided on-site to treat high organic loading wastewater (such as sewage from floor & vehicle washing and the IWMF staff & visitors) for reuse in the incineration plant and mechanical treatment plant or for washdown and landscape irrigation. The bunker leachate / ash leachate from incineration plant (as described in Section 5b.6.2.9) would be highly polluted and would be conveyed to the incineration plant and co-incinerated with MSW. All other wastewater (i.e. low/nil organic loading wastewater mainly coming from plant machinery such as demineralizer drain) only requires simple treatment such as sedimentation or neutralization or even not requires any treatment before being used for flue gas cooling in quench tower or ash quenching. Therefore, no adverse water quality impact would be expected.
Transportation of bottom ash, fly ash and APC residues to WENT Landfill for disposal
5b.7.6.2 As discussed in Section 2, the IWMF would comprise (a) an advanced thermal incineration plant of about 3,000 tpd capacity and (b) a demonstration-scale mechanical treatment plant of about 200tpd or less for mixed MSW. The main waste type to be generated during the operation of the thermal incineration plant would be bottom ash, fly ash and air pollution control (APC) residues. For treating 3,000 tpd of mixed MSW, it is estimated that approximately 660 tpd of bottom ash and 120 tpd of fly ash and APC residues would be generated from the thermal incineration plant.
5b.7.6.3 The bottom ash is considered to be inert provided that the combustion systems in the incinerator are designed and operated correctly, and would be disposed of at landfill. Fly ash and APC residues from the flue gas stream can also be disposed of at landfill after proper treatment. The pollution load in fly ash and APC residues would likely be higher and more readily leachable than that in bottom ash. Cement solidification or chemical stabilization would be adopted to pre-treat the fly ash and APC residue to ensure that they would conform to the proposed Incineration Residue Pollution Control Limits and leachability criteria.
5b.7.6.5 Moreover, as suggested in Section 6, the ashes would be stored in covered container during transportation, the possibility for accidental exposure of the ashes to the environment would be low. Transportation of incineration waste would also be avoided during adverse weather condition to prevent unanticipated risk. It is believed that the risk to the environment concerning the accidental spillage of ashes during marine transportation is low.
5b.7.6.6 As the incineration waste would be stored within covered container which prevent the contact between the incineration waste and marine water, leaching of pollutants from the incineration waste may not occur even if the containers fall into the sea accidentally. Even if spillage occurs under a very remote condition with some of the incineration waste containers damaged, the containers could still be effective in limiting the dispersion of leached contaminants. As discussed in Section 5b.7.6.4, the flying ash and APC residue from the incineration plants would be solidified and the leaching of pollutant would be limited. For the non-solidified bottom ash which passes the leachability criteria of Incineration Residue Pollution Control Limits (IRPCL), the possible level of leaching would be far lower than the contaminant levels listed in the IRPCL. In case of spillage, the dropped containers which contain the incineration waste are expected to fall to the seabed and could be recovered by the cleanup operation following the spillage. This would prevent the prolonged localized impact on water quality in case some of the containers are opened or damaged.
5b.7.6.7 It is expected that the possibility for the spillage of incineration waste would be scant. In case of spillage, the leaching of pollutant from the incinerated waste would be limited and the consequential impact on water quality, ecology and fishery resources would be low. A comprehensive emergency response plan for any accidental spillage should be submitted by the operation contractor to the EPD for agreement before the operation of the facilities. Salvage and cleanup action to recover the spilled incineration waste containers following the spillage should be carried out according to the emergency response plan to mitigate the environmental impact in case of spillage. Further details on the management of waste are given in Section 6.
Discharge of Saline Water from Desalination Plant
5b.7.6.8
Approximately
5b.7.6.9 A comparison of the characteristics of the saline water discharge with the standards for effluents discharged into the inshore waters of Southern Water Control Zone is given in Table 5b.15 below:
Table 5b.15 Comparison of Saline Water Discharge from Desalination Plant with Effluent Discharge Standard
Parameters |
Saline
Water Note 1 |
Discharge
Standard Note 2 & 3 |
Compliance
with Discharge Standard |
pH |
6 – 8 |
6 – 9 |
Yes |
Temperature (℃) |
14 – 29 |
40 |
Yes |
Suspended Solids (mg/L) |
6 – 30 |
30 |
Yes |
BOD (mg/L) |
0.4 – 3.0 |
20 |
Yes |
Total Residual Chorine (mg/L) |
<1 |
<1 |
Yes |
Notes: 1. It is calculated based on the
assumption that the brine water produced is generally 1.7-1.8 times more
concentrated than raw seawater for SS and BOD. There will be no temperature
elevations in the saline water discharge as compared to the ambient water
temperature. The characteristics of the
baseline seawater quality are obtained from “Marine Quality in
2. Discharge standard for flow rate of >1500 and ≤
3. The effluent discharge standards do not specify a standard for salinity.
5b.7.6.11
The density profiles at the monitoring point was extracted
and analyzed on days of spring tide and neap tide in both dry and wet seasons. The model results shown that the vertical
density in dry season was found to be uniformly distributed over the entire
water column while in wet season, a stratification of seawater was observed. Based on the analysis of ambient density
profiles, two sets of density profiles were adopted for near field simulation,
including one set of density profile predicted for dry season (with no sign of
stratification) and one set of density profile predicted for wet season (with the
highest degree of stratification predicted over the entire 15-day wet season
simulation period). The current profiles
extracted at the same monitoring point was also analyzed and calculated as 10
and 90 percentile values (for dry and wet seasons). The near field impact was simulated for
different combinations of vertical density profile and ambient current velocity
using the design effluent flow rate (
Table 5b.16 Density Profile at SKC IWMF Desalination Plant Outfall
Depth from water surface (m) |
Density (kg/m3) |
|
Dry Season (D) |
Wet Season (W) |
|
0.76 |
1.0150 |
1.0067 |
2.27 |
1.0150 |
1.0068 |
3.79 |
1.0150 |
1.0075 |
5.30 |
1.0150 |
1.0101 |
6.82 |
1.0150 |
1.0124 |
8.33 |
1.0150 |
1.0137 |
9.85 |
1.0150 |
1.0145 |
11.36 |
1.0150 |
1.0152 |
12.88 |
1.0150 |
1.0152 |
14.39 |
1.0150 |
1.0152 |
15.15 |
1.0150 |
1.0152 |
Table 5b.17 Current Velocity Profile at SKC IWMF Desalination Plant Outfall
Depth from water surface (m) |
Current Velocity (m/s) |
|||
Dry Season |
Wet Season |
|||
10%ile (dv10) |
90%ile (dv90) |
10%ile (wv10) |
90%ile (wv90) |
|
0.76 |
0.0604 |
0.3369 |
0.1140 |
0.6427 |
2.27 |
0.0555 |
0.3342 |
0.1530 |
0.6734 |
3.79 |
0.0532 |
0.3307 |
0.1746 |
0.6403 |
5.30 |
0.0517 |
0.3260 |
0.1401 |
0.5553 |
6.82 |
0.0506 |
0.3201 |
0.1210 |
0.4561 |
8.33 |
0.0498 |
0.3128 |
0.0897 |
0.3540 |
9.85 |
0.0491 |
0.3038 |
0.0538 |
0.2801 |
11.36 |
0.0484 |
0.2921 |
0.0364 |
0.2365 |
12.88 |
0.0475 |
0.2763 |
0.0316 |
0.2104 |
14.39 |
0.0457 |
0.2514 |
0.0250 |
0.1698 |
15.15 |
0.0457 |
0.2514 |
0.0250 |
0.1698 |
Table 5b.18 Summary of Proposed Model Runs
Model Run No. |
Run ID |
Density Profile |
Ambient Current Velocity |
1 |
D-dv10 |
D |
dv10 |
2 |
D-dv90 |
D |
dv90 |
3 |
W-wv10 |
W |
wv10 |
4 |
W-wv90 |
W |
wv90 |
Prediction and Evaluation of Near
Field Modelling Results
Plate 5b.5 Illustration
of Plume Properties of Surfacing Plume and Trapped Plume
5b.7.6.13 Table 5b.19 summarizes the results from the VISJET simulations. The predicted initial dilutions in Table 5b.20 were corrected with the background concentration build up due to the tidal effects. The basic assumption of any near field model is mixed with clean water. In actuality this is not true, particularly in a tidally mixed environment. The average tracer background build up concentrations were calculated from the far field Update model. The background build up was quantified by performing a conservative tracer run on the effluent. A conservative tracer, i.e. without decay or reaction, was used. The initial concentration of the tracer in the desalination plant seawall discharge outfall was set to be 1000 mg/l. It should be noted that the results from the grid cell into which the tracer is loaded is not representative of the true background build up as this cell will always contain the background build up plus the continuous tracer loading. Therefore, the necessary far field tracer results were taken from a cell located adjacent to the outfall grid cells. The average tracer results were predicted in both dry and wet seasons and were used for the background build up corrections. Table 5b.19 shows an example of the background build up correction (Run ID: D-dv10).
Table 5b.19 Example for Background Build Up Correction
Run ID |
Minimum Initial Dilution 1 |
Initial Tracer Concentration in Effluent 2 (mg/L) |
Average Tracer Concentration 3 (mg/L) |
Corrected Minimum Initial Dilution 4 |
(A) |
(B) |
(C) |
(D) |
|
D-dv10 |
106 |
1000 |
1.68 |
90 |
Note: 1. Minimum initial dilution predicted by VISJET model. This dilution occurred in the dry season (Run ID: D-dv10).
2. Effluent tracer concentration assumed in the far field modelling.
3. Average background build up concentration for dry season predicted by the far field model.
4. The average background build up concentration for dry season was used for the correction in this case as the minimum dilution occurred under the dry season scenario. Corrected Initial Dilution, (D) = (B) ÷ {[1 x (B) + ((A) – 1) x (C)] ÷ (A)}
Table 5b.20 Summary of Initial Dilutions Predicted at the Edge of ZID
Run ID |
Initial Dilution at the Edge of ZID 1 |
Corrected Initial Dilution at the Edge of ZID 2 |
Downstream Distance from Centre of the Outfall at the Edge of ZID
(m) 3 |
D-dv10 |
106 |
90 |
4.5 |
D-dv90 |
4802 |
529 |
60.5 |
W-wv10 |
113 |
103 |
8.4 |
W-wv90 |
321 |
251 |
71.8 |
Note: 1. Initial dilutions at the edge of the ZID calculated by VISJET model.
2. Initial dilutions at the edge of the ZID were corrected using the background build up concentration predicted by the far field Update model.
3. Definition of ZID is provided in Section 5b.7.6.12.
5b.7.6.14
As shown in Table
5b.20, the predicted minimum dilution rate is 90 which
occurred in dry season with the smallest ambient current velocity
(D-dv10). The maximum predicted
downstream distance is about
5b.7.7 Operation Phase Impact due to Maintenance Dredging
5b.7.7.1
Depend on the actual sedimentation and scouring
condition, maintenance dredging within the embayment area maybe needed to
provide safe marine access route for the daily marine transportation. The need
for maintenance dredging should be determined by onsite survey after the
completion of this Project. Currently, no regular maintenance dredging is
proposed under this Project. It is suggested that any maintenance dredging in
the future should not be carried out within
5b.7.8 Operation Phase Impact on Flow Regime
5b.7.8.1
The potential site at SKC is located in the open water
in the deeper southern waters of Hong Kong and is influenced by the
5b.7.8.2
The proposed reclamation is facing the large open
Wanshan Archipelago Cross Section
Plate 5b.6 Location of Cross Section
5b.7.8.3
Taking into account the water depths along this cross
section and the maximum current velocity along this open water (of up to
5b.7.9
Operation Phase Impact on the Channel between
the
5b.7.9.1
As shown in Figure 5b.2, a channel
would be formed between the
Plate
5b.7 Seabed Level at
Plate
5b.8 Typical Localized Current Flow Pattern
at SKC (Extracted from the Delft3D Detailed HATS Model developed under the Approved
EIA for Harbour Area Treatment Scheme (HATS) Stage 2A
5b.7.9.2
In addition, there will be no pollution discharge into
the area within the proposed breakwater and the channel between the SKC shore
and the proposed facilities. As discussed in Section 5b.6.2.14,
the IWMF would be designed with a net zero discharge of process and sanitary
wastewater. The only discharge would be the concentrated saline water from the
desalination plant. Since the proposed outfall discharge locates at the
Southern side of the breakwater (above
5b.7.9.3 Floating refuse may be washed into the channel and pose a risk for deterioration of water quality of the channel. It is suggested that regular collection of refuse within the channel area should be done to avoid deterioration of water quality of the channel. It is expected that water quality impact due to floating refuse would be minimal if measure could be taken to regularly collect refuse within the channel.
5b.7.10 Operation Phase Impact on Dissolved Oxygen within the Breakwaters Sheltered Water
5b.7.10.1
The breakwaters around the SKC would reduce the water
current within the sheltered water for the safety of the vessels in the Project
Site. However, the effect of slowing down the water is limited as the
breakwaters only partially enclose the sheltered area, leaving a huge void for
water to exchange with the surrounding.
The width of the opening of breakwater at the narrowest location is
about
5b.7.11 Operation Phase Impact on Sediment Deposition and Water Quality
5b.7.11.1 Ecological sensitive receivers including benthic and coral communities would be sensitive to sediment deposition. The marine embayment formed by the proposed breakwaters would reduce the local currents and stimulate sediment deposition within the embayment area, which may have an effect on the intertidal and subtidal habitats at or near the Project site. In addition, construction of the breakwaters and reclamation as well as any future maintenance dredging for safe marine access would unavoidably cause a direct loss of intertidal and subtidal habitats at or near the Project site. Detailed ecological impact assessment is provided separately in Section 7b.
5b.7.11.2
The key pollution / sediment source within the Study
Area would be the
5b.7.11.3 No additional pollution loading will be discharged into the area within the proposed breakwaters where the coral communities would be located. Hence no significant water quality impact in terms of sediment deposition upon the coral communities would be expected from the Project. The water quality effects upon the coral communities identified within the Project site due to the saline water discharges from the desalination plant have been assessed by near field dispersion modelling (refer to Sections 5b.7.6.8 to 5b.7.6.14). Detailed impact assessment on the identified coral communities and necessary mitigation measures for protection of these coral communities are given in Section 7b.
5b.7.11.4
All other identified WSRs, such as the beaches at
Lantau South and Cheung Chau East would not be directly influenced by the tidal
flow running across the Project site (refer to Section 5b.7.7) and in view that these WSRs are even far more
away from the Project activities (more than
Drainage and Construction Site Runoff
· At the start of site establishment, perimeter cut-off drains to direct off-site water around the site should be constructed with internal drainage works and erosion and sedimentation control facilities implemented. Channels (both temporary and permanent drainage pipes and culverts), earth bunds or sand bag barriers should be provided on site to direct storm water to silt removal facilities. The design of the temporary on-site drainage system will be undertaken by the contractor prior to the commencement of construction.
· Boundaries of earthworks should be surrounded by dykes or embankments for flood protection, as necessary.
· Sand/silt removal facilities such as sand/silt traps and sediment basins should be provided to remove sand/silt particles from runoff to meet the requirements of the TM-DSS. The design of efficient silt removal facilities should be based on the guidelines in Appendix A1 of ProPECC PN 1/94, which states that the retention time for silt/sand traps should be 5 minutes under maximum flow conditions. The detailed design of the sand/silt traps shall be undertaken by the contractor prior to the commencement of construction. The design of a typical sand trap, extracted from ProPECC PN 1/94, is provided in Appendix 5-3.
· Water pumped out from foundation piles must be discharged into silt removal facilities.
· Measures should be taken to minimize the ingress of site runoff and drainage into excavations. Drainage water pumped out from excavations should be discharged into storm drains via silt removal facilities.
· During rainstorms, exposed slope/soil surfaces should be covered by a tarpaulin or other means, as far as practicable. Other measures that need to be implemented before, during and after rainstorms are summarized in ProPECC PN 1/94.
· Exposed soil areas should be minimized to reduce potential for increased siltation and contamination of runoff.
· Earthwork final surfaces should be well compacted and subsequent permanent work or surface protection should be immediately performed.
· Open stockpiles of construction materials or construction wastes on-site should be covered with tarpaulin or similar fabric during rainstorms.
General Construction Activities
Accidental Spillage
5b.8.1.5 Maintenance of vehicles and equipments involving activities with potential for leakage and spillage should only be undertaken within the areas which appropriately equipped to control these discharges.
5b.8.1.6 Oils and fuels should only be used and stored in designated areas which have pollution prevention facilities. All fuel tanks and storage areas should be sited on sealed areas in order to prevent spillage of fuels and solvents to the nearby watercourses. All waste oils and fuels should be collected in designated tanks prior to disposal.
· Suitable containers should be used to hold the chemical wastes to avoid leakage or spillage during storage, handling and transport.
· Chemical waste containers should be suitably labelled, to notify and warn the personnel who are handling the wastes, to avoid accidents.
· Storage area should be selected at a safe location on site and adequate space should be allocated to the storage area.
Sewage Effluent
Reclamation, Construction of Breakwaters and Localized Dredging for Anti-Scouring Protection Layer
· The proposed dredging and reclamation should be commenced in phases. The breakwaters and seawalls should be constructed using cofferdam method and the reclamation should be started within the enclosed breakwaters after the completion of the breakwater. Silt curtain should be used to surround the circular cell during the filling of the cell to prevent the loss of fine in the filling material
· Water trapped inside the cofferdam, if any, would be pumped out for treatment before discharge.
·
The maximum production rate for dredging for the
anti-scouring protection layer shall not exceed
· Any gap that may need to be provided for marine access will be located at the middle of the North Western seawall, away from the identified coral communities and will be shielded by silt curtains systems to control sediment plume dispersion.
· The silt curtain system at marine access opening should be closed as soon as the barges passes through the marine access opening in order to minimize the period of curtain opening. Filling should only be carried out behind the silt curtain when the silt curtain is completely closed.
·
To enhance the effectiveness of the silt curtain
at the marine access, the northern breakwater would be built before the
commencement of the reclamation to reduce the current velocity towards the
marine access opening.
· The silt curtain system at marine access opening should be regularly checked and maintained to ensure proper functioning.
· Where public fill is proposed for filling below +2.5mPD, the fine content in the public fill will be controlled to 25% which is in line with the CEDD’s General Specification;
· The filling for reclamation should be carried out behind the seawall. The filling material should only consist of public fill, rock and sand. The production rate for each composition at each filling area should follow those delineated in Table 5b.9 and Table 5b.10. The filling above high watermark is not restricted;
· For dredging for anti-scouring protection layer, the contractor should follow the production rate stated in Table 5b.12 and the maximum number of grab per hour stated in Table 5b.13;
·
No dredging should be carried out within
· Daily site audit including full-time on-site monitoring by the ET is recommended during the dredging for anti-scouring protection layer for checking the compliance with the permitted no. of grab to be performed per hour by the dredging contractor as specified in Table 5b.13;
· Closed grab dredger should be used to minimize the loss of sediment during the raising of the loaded grabs through the water column;
· Frame-type silt curtains should be deployed around the dredging operations;
· Floating-type silt curtains should be used to surround the circular cell during the sheetpiling work;
· The descent speed of grabs should be controlled to minimize the seabed impact speed;
· Barges should be loaded carefully to avoid splashing of material;
· All barges used for the transport of dredged materials should be fitted with tight bottom seals in order to prevent leakage of material during loading and transport;
· No concurrence works between laying of submarine cables and dredging/reclamation works within the same location is allowed. For works close to each other, the construction program should be arranged so that the dredging/reclamation works within area bounded by the breakwaters and the laying of cables would not operate within a distance of 80m from each other to avoid any accumulative impact on the environment (in case if such tight schedule is necessary).
· All barges should be filled to a level which ensures that material does not spill over during loading and transport to the disposal site and that adequate freeboard is maintained to ensure that the decks are not washed by wave action.
· Coral which are directly affected should be translocated as much as practicable. (Please refer to Section 7 for the details on coral translocation)
Site Effluent
5b.8.2.1 The Project Site will be equipped with an adequately sized wastewater treatment plant to provide treatment to some wastewater generated from the IWMF (mainly human sewage) for reuse in the incineration plant and mechanical treatment plant or for washdown and landscape irrigation in the IWMF site. A “net zero discharge” scheme will be adopted during the operation of the IWMF.
5b.8.2.2 A small amount of brine water will be discharged into the marine water from the proposed desalination plant via a seawall discharge outfall at the southern boundary of the artificial island near SKC (outside the proposed breakwaters). The potential water quality impact from the brine water discharge has been assessed to be negligible and therefore no mitigation measure specific to the brine water discharge is required.
Surface Runoff
5b.8.2.3 A pipeline drainage system will serve the development area collecting surface runoff from paved areas, roof, etc. Sustainable drainage principle would be adopted in the drainage system design to minimize peak surface runoff, maximize permeable surface and maximize beneficial use of rainwater.
5b.8.2.4 Oil interceptors should be provided in the drainage system of any potentially contaminated areas (such as truck parking area and maintenance workshop) and regularly cleaned to prevent the release of oil products into the storm water drainage system in case of accidental spillages. Accidental spillage should be cleaned up as soon as practicable and all waste oils and fuels should be collected and handled in compliance with the Waste Disposal Ordinance.
Refuse Entrapment
5b.8.2.5 Collection and removal of floating refuse should be performed at regular intervals for keeping the water within the Project site boundary and the neighbouring water free from rubbish.
Transportation of bottom ash, fly ash and APC residues to WENT Landfill for disposal
5b.8.2.6 Mitigation measures for minimizing the risk of spillage during transportation of bottom ash, fly ash and APC residues are given in Section 6b.5 and Section 6b.6 under the Waste Management Implementation section. Covered container should be used in the shipping of the incineration waste to limit the contact between the incineration waste and the marine water. A comprehensive emergency response plan for any accidental spillage should be submitted by the operation contractor to the EPD for agreement before the operation of the facilities. Salvage and cleanup action to recover the spilled incineration waste containers following the spillage should be carried out according to the emergency response plan to mitigate the environmental impact in case of spillage.
Maintenance Dredging
5b.8.2.7 The precautions stated in Section 5b.8.1.9 should be followed to avoid any adverse water quality impact. The dredging extent shown in Figure 5b.8 for the maintenance dredging should be followed to avoid dredging too close to the coral communities.
5b.9 Evaluation of Residual Impacts
5b.9.1.1 With the implementation of the recommended mitigation measures, no unacceptable residual impacts on water quality are expected.
5b.10 Environmental Monitoring and Audit
5b.10.1.1 Water quality monitoring and audit is recommended to be carried out during the proposed marine construction works at SKC to ensure that the recommended mitigation measures are implemented properly. During submarine cable laying works, dredging, installation of cofferdam, reclamation, a water quality monitoring programme should be conducted to ensure no unacceptable water quality impacts will occur at the nearby WSRs. If the water quality monitoring data indicate that the proposed marine works result in unacceptable water quality impacts in the receiving water, appropriate actions should be taken to review the construction method and additional measures such as slowing down, or rescheduling of works should be implemented as necessary. During the reclamation, the opening for marine access would be shielded by silt curtains system to control the dispersion of filling material from the reclamation area. To ensure the actual efficiency of the silt curtains system would be at least as high as the level assumed in the assessment, a field trial to verify the reduction effect of the silt curtain during the EM&A stage is recommended. Daily site audit should be performed during the dredging for anti-scouring protection layer to ensure compliance with the recommended dredging rates as shown in Table 5b.11. Water quality monitoring and audit is also recommended to be carried out during the first four weeks of the commission of brine water discharge at SKC to ensure that no adverse water quality impact would occur due to the discharge of brine. Details of the water quality monitoring programme and the Event and Action Plan will be provided in the stand-alone EM&A Manual. Monitoring of effluent quality is also recommended for operation stage and under the perspective of the WPCO.
5b.11.1.1 The water quality impact during the dredging works for the proposed breakwater construction will not be anticipated due to the phasing and construction method adopted while the water quality impact during the reclamation and dredging for anti-scouring protection layer has been quantitatively assessed using the near field sediment dispersion model. The model results indicated that the water quality impact generated from the reclamation and dredging works under mitigated scenario would be localized and minor and would unlikely contribute any significant water quality impact. Mitigation measures are proposed to ensure that no unacceptable water quality impact would be resulted from the reclamation and dredging works.
5b.11.1.2
During
the operation phase of the Project, wastewater will be generated from the
proposed incineration plant and mechanical treatment plant. An on-site wastewater treatment plant will be
provided. All generated wastewater will
be discharged to the on-site wastewater treatment plant and treated. The treated effluent from the wastewater
treatment plant will be reused in the incineration plant and mechanical
treatment plant or washdown and landscape irrigation in the IWMF site.
There would be no wastewater effluent discharged to the coastal waters
of
5b.11.1.3 The water quality impact for any potential maintenance dredging has been quantitatively assessed using the near field sediment dispersion model. The model results indicated that the water quality impact generated from the dredging works under mitigated scenario would be localized and minor and would unlikely contribute any significant water quality impact.
5b.11.1.4
Saline
water would be discharged from the proposed desalination plant in a low
discharge rate. The saline water has
been quantitatively assessed to be minor and acceptable. Adverse impacts on water quality due to the
proposed saline water discharge would not be expected.