6                                            WATER QUALITY ASSESSMENT

6.1                                      Introduction

The construction, operation, restoration and aftercare of the Extension have the potential to cause adverse water quality impacts if not properly managed.  This section examines the potential impacts on the nearby water resources due to discharge of construction runoff into the watercourses and marine waters, the potential discharge of leachate into the surface and groundwater systems.  The impacts are evaluated through a review of the surface water and leachate management systems for the Extension. 

6.2                                      Legislation and Standards

The regulatory requirements and standards to protect water quality are as follows:

·       Water Pollution Control Ordinance (WPCO) (Cap. 358);

·       Environmental Impact Assessment Ordinance (Cap. 499. S.16), Technical Memorandum on Environmental Impact Assessment Process (EIAO-TM), Annexes 6 and 14;

·       Technical Memorandum Standards for Effluents Discharged into Drainage and Sewerage Systems, Inland and Inshore Waters (TM);

·       Practice Note for Professional Persons on Construction Site Drainage (Prop PECC PN 1/94); and,

·       Hong Kong Planning Standards and Guidelines (HKPSG).

6.2.1                                Water Pollution Control Ordinance (WPCO)

The WPCO is the legislation for the control of water pollution and water quality in Hong Kong.  Under the WPCO, Hong Kong waters are divided into 10 Water Control Zones (WCZs).  Each WCZ has a designated set of statutory Water Quality Objectives (WQOs).  The WQOs set limits for different parameters that should be achieved in order to maintain the water quality within the WCZs.  Corresponding statements of WQO are stipulated for different water regimes, i.e. marine waters, inland waters, bathing beaches subzones, secondary contact recreation subzones and fish culture subzones, in the WCZ based on their beneficial uses.

The assessment area (thereafter referred to as the Study Area) is defined in the EIA Study Brief (No. ESB-119/2004) as all areas within 500m from the boundary of the Extension Site (see Figure 6.2a).  In accordance with the WPCO, the Study Area is located inside the Junk Bay WCZ and is in close proximity to the Eastern Buffer WCZ.    

The WQOs for the Junk Bay WCZ and Eastern Buffer WCZ, which are presented in Tables 6.2a and 6.2b, respectively, are applicable as evaluation criteria for assessing compliance of any effects from the discharges of the Project.

Table 6.2a      Water Quality Objectives for Junk Bay Water Control Zone

Water Quality Objectives

Junk Bay WCZ

A.      AESTHETIC APPEARANCE

 

(a)    Waste discharges shall cause no objectionable odours or discolouration of the water.

Whole Zone

(b)    Tarry residues, floating wood, articles made of glass, plastic, rubber or of any other substance should be absent.

Whole Zone

(c)     Mineral oil should not be visible on the surface. Surfactants should not give rise to a lasting foam.

Whole Zone

(d)    There should be no recognisable sewage-derived debris.

Whole Zone

(e)    Floating, submerged and semi-submerged objects of a size likely to interfere with the free movement of vessels, or cause damage to vessels, should be absent.

Whole Zone

(f)      Waste discharges shall not cause the water to Whole Zone contain substances which settle to form objectionable deposits.

Whole Zone

B.      BACTERIA

 

(a)    The level of Escherichia coli should 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 (L.N. 451 of 1991)

(b)    (Repealed L.N. 451 of 1991)

-

(c)     The level of Escherichia coli should not exceed 1000 per 100 ml, calculated as the running median of the most recent 5 consecutive samples taken at intervals of between 7 and 21 days.

Inland waters

C.     COLOUR

 

Waste discharges shall not cause the colour of water to exceed 50 Hazen units.

Inland waters

D.     DISSOLVED OXYGEN

 

(a)    Waste discharges shall not cause the level of dissolved oxygen to fall below 4 mg L-1 for 90% of the sampling occasions during the year; values should be calculated as the water column average (arithmetic mean of at least 3 measurements at 1 m below surface, mid-depth and 1 m above seabed). In addition, the concentration of dissolved oxygen should not be less than 2 mg L-1 within 2 m of the seabed for 90% of the sampling occasions during the year.

Marine waters excepting Fish Culture Subzones

(b)    The dissolved oxygen level should not be less than 5 mg L-1 for 90% of the sampling occasions during the year; values should be calculated as water column average (arithmetic mean of at least 3 measurements at 1 m below surface, mid-depth and 1 m above seabed). In addition, the concentration of dissolved oxygen should not be less than 2 mg L-1 within 2 m of the seabed for 90% of the sampling occasions during the year.

Fish Culture Subzones

(c)     Waste discharges shall not cause the level of dissolved oxygen to be less than 4 mg L-1.

Inland waters

E.         pH

 

(a)    The pH of the water should be within the range of 6.5-8.5 units. In addition, waste discharges shall not cause the natural pH range to be extended by more than 0.2 units.     

Marine waters (L.N. 451 of 1991)

(b)    (Repealed L.N. 451 of 1991)

-

(c)     The pH of the water should be within the range of 6.0-9.0 units.

Inland waters

F.      TEMPERATURE

 

Waste discharges shall not cause the natural daily temperature range to change by more than 2.0oC.

Whole Zone

G.     SALINITY

 

Waste discharges shall not cause the natural ambient salinity level to change by more than 10%.

Whole Zone

H.     SUSPENDED SOLIDS

 

(a)    Waste discharges shall neither cause the natural ambient level to be raised by 30% nor give rise to accumulation of suspended solids which may adversely affect aquatic communities.

Marine waters

(b)    Waste discharges shall not cause the annual median of suspended solids to exceed 25 mg L-1.

Inland waters

I.       AMMONIA

 

The ammonia nitrogen level should not be more than 0.021 mg L-1, calculated as the annual average (arithmetic mean), as unionized form.

Whole Zone

J.   NUTRIENTS

 

(a)  Nutrients shall not be present in quantities sufficient to cause excessive or nuisance growth of algae or other aquatic plants.

Marine waters

(b)  Without limiting the generality of objective (a) above, the level of inorganic nitrogen should not exceed 0.3 mg L-1, expressed as annual water column average (arithmetic mean of at least 3 measurements at 1 m below surface, mid-depth and 1 m above seabed).

Marine waters

K.   5-DAY BIOCHEMICAL OXYGEN DEMAND

 

Waste discharges shall not cause the 5-day biochemical oxygen demand to exceed 5 mg L-1.

Inland waters

L.  CHEMICAL OXYGEN DEMAND

 

Waste discharges shall not cause the chemical oxygen demand to exceed 30 mg L-1.

Inland waters

M. DANGEROUS SUBSTANCES

 

(a)  Waste discharges shall not cause the concentrations of dangerous substances in the water to attain such levels as to produce significant toxic effects in humans, fish or any other aquatic organisms, with due regard to biologically cumulative effects in food chains and to toxicant interactions with each other.

Whole Zone

(b)  Waste discharges of dangerous substances shall not put a risk to any beneficial uses of the aquatic environment.

Whole Zone

N-O (Repealed L.N. 451 of 1991)

Whole Zone

Table 6.2b      Water Quality Objectives for Eastern Buffer Water Control Zone

Water Quality Objectives

Eastern Buffer WCZ

A.      AESTHETIC APPEARANCE

 

(a)    There should be no objectionable odours or discolouration of the water.

Whole Zone

(b)    Tarry residues, floating wood, articles made of glass, plastic, rubber or of any other substances should be absent.

Whole Zone

(c)     Mineral oil should not be visible on the surface. Surfactants should not give rise to a lasting foam.

Whole Zone

(d)    There should be no recognisable sewage-derived debris.

Whole Zone

(e)    Floating, submerged and semi-submerged objects of a size likely to interfere with the free movement of vessels, or cause damage to vessels, should be absent.

Whole Zone

(f)      The water should not contain substances which settle to form objectionable deposits.

Whole Zone

B.      BACTERIA

 

(a)    The level of Escherichia coli should not exceed 610 per 100 mL, calculated as the geometric mean of all samples collected in a calendar year.

Fish Culture Subzones

(b)    The level of Escherichia coli should be less than 1 per 100 mL, calculated as the geometric mean of the most recent 5 consecutive samples taken at intervals of between 7 and 21 days.

Water Gathering Ground Subzones

(c)     The level of Escherichia coli should not exceed 1000 per 100 mL, calculated as the geometric mean of the most recent 5 consecutive samples taken at intervals of between 7 and 21 days.

Other inland waters

C.     COLOUR

 

(a)   Human activity should not cause the colour of water to exceed 30 Hazen units.

Water Gathering Ground Subzones

(b)   Human activity should not cause the colour of water to exceed 50 Hazen units.

Other inland waters

D.     DISSOLVED OXYGEN

 

(a)    The level of dissolved oxygen should not fall below 4 mg L-1 for 90% of the sampling occasions during the whole year; values should be calculated as water column average (arithmetic mean of at least 3 measurements at 1 m below surface, mid-depth and 1 m above seabed). In addition, the concentration of dissolved oxygen should not be less than 2 mg L-1 within 2 m of the seabed for 90% of the sampling occasions during the whole year.

Marine waters excepting Fish Culture Subzones

(b)    The level of dissolved oxygen should not be less than 5 mg L-1 for 90% of the sampling occasions during the year; values should be calculated as water column average (arithmetic mean of at least 3 measurements at 1 m below surface, mid-depth and 1 m above seabed). In addition, the concentration of dissolved oxygen should not be less than 2 mg L-1 within 2 m of the seabed for 90% of the sampling occasions during the whole year.

Fish Culture Subzones

(c)     The level of dissolved oxygen should not be less than 4 mg L-1.

Water Gathering Ground Subzones and other inland waters

E.         pH

 

(a)    The pH of the water should be within the range of 6.5-8.5 units. In addition, human activity should not cause the natural pH range to be extended by more than 0.2 units.

Marine waters

(b)    Human activity should not cause the pH of the water to exceed the range of 6.5-8.5 units.

Water Gathering Ground Subzones

(c)     Human activity should not cause the pH of the water to exceed the range of 6.0-9.0 units.

Other inland waters

F.      TEMPERATURE

 

Human activity should not cause the natural daily temperature range to change by more than 2.0oC.

Whole Zone

G.     SALINITY

 

Human activity should not cause the natural ambient salinity level to change by more than 10%.

Whole Zone

H.     SUSPENDED SOLIDS

 

(a)    Human activity should neither cause the natural ambient level to be raise by more than 30 % nor give rise to accumulation of suspended solids which may adversely affect aquatic communities.

Marine waters

(b)    Human activity should not cause the annual median of suspended solids to exceed 20 mg L-1.

Water Gathering Ground Subzones

(c)  Human activity should not cause the annual median of suspended solids to exceed 25 mg L-1.

Other inland waters

I.       AMMONIA

 

The un-ionized ammoniacal nitrogen level should not be more than 0.021 mg L-1, calculated as the annual average (arithmetic mean).

Whole Zone

J.   NUTRIENTS

 

(a)  Nutrients should not be present in quantities sufficient to cause excessive or nuisance growth of algae or other aquatic plants.

Marine waters

(b)  Without limiting the generality of objective (a) above, the level of inorganic nitrogen should not exceed 0.4  mg L-1, expressed as annual water column average (arithmetic mean of at least 3 measurements at 1 m below surface, mid-depth and 1 m above seabed).

Marine waters

K.   5-DAY BIOCHEMICAL OXYGEN DEMAND

 

(a)  The 5-day biochemical oxygen demand should not exceed 3 mg L-1.

Water Gathering Ground Subzones

(b)  The 5-day biochemical oxygen demand should not exceed 5 mg L-1.

Other inland waters

L.  CHEMICAL OXYGEN DEMAND

 

(a)  The chemical oxygen demand should not exceed 15 mg L-1.

Water Gathering Ground Subzones

(b)  The chemical oxygen demand should not exceed 30 mg per litre.

Other inland waters

M. TOXIC SUBSTANCES

 

(a)  Toxic substances in the water should not attain such levels as to produce significant toxic, carcinogenic, mutagenic or teratogenic effects in humans, fish or any other aquatic organisms with due regard to biologically cumulative effects in food chains and to interactions of toxic substances with each other.

Whole Zone

(b)  Human activity should not cause a risk to any beneficial use of the aquatic environment.

Whole Zone

6.2.2                                Technical Memorandum for Effluent Discharges into Drainage and Sewerage Systems, Inland and Inshore Waters (TM)

All discharges from the Project are required to comply with the Technical Memorandum for Effluents Discharged into Drainage and Sewerage Systems, Inland and Inshore Waters (TM) issued under Section 21 of the WPCO.  The TM defines discharge limits for different types of receiving waters.  Under the TM, effluents discharged into the drainage and sewerage systems, inshore and inshore waters of the WCZs are subject to pollutant concentration standards for particular discharge volumes.  Any new discharges within a WCZ are subject to licence conditions and the TM acts as a guideline for setting discharge standards for inclusion in the licence.  Any sewage from the proposed construction and operational activities should comply with the standards for effluent discharged into the foul sewers, inshore waters or marine waters of the Junk Bay and Eastern Buffer WCZs, shown in Tables 1, 10a and 10b of the TM, respectively.

Currently, the treated effluent from the existing leachate treatment plant of the existing SENT Landfill (thereafter referred to as the Bioplant) is discharged to the Tseung Kwan O Sewage Treatment Works (TKO STW).  The quantity and composition of any effluent discharged from the landfill shall not exceed any of the regulatory limits as stipulated in the existing discharge license.

6.2.3                                Environmental Impact Assessment Ordinance (Cap. 499. S.16), Technical Memorandum on Environmental Impact Assessment Process (EIAO-TM)

Annexes 6 and 14 of the Environmental Impact Assessment Ordinance (Cap. 499. S.16), Technical Memorandum on Environmental Impact Assessment Process (EIAO-TM) provide general guidelines and criteria to be used in assessing water quality issues.

6.2.4                                Practice Note for Professional Persons on Construction Site Drainage (ProPECC PN 1/94)

The ProPECC PN 1/94 issued by the EPD provides some basic environmental guidelines for the handling and disposal of construction site discharges to prevent or minimise construction impacts on water quality.

Whilst the technical circulars are non-statutory, they are generally accepted as best guidelines in Hong Kong and have been adopted as relevant for this assessment.

6.2.5                                Hong Kong Planning Standards and Guidelines (HKPSG)

Chapter 9 of the Hong Kong Planning Standards and Guidelines (HKPSG) provides guidance for including environmental considerations in the planning of both public and private developments.  It applies both to the planning of permanent or temporary uses which will have potential to cause significant changes to the biophysical environment or which are sensitive to environmental impacts.  Section 5 in Chapter 9 of the HKPSG provides additional information on regulatory guidelines against water pollution for sensitive uses such as aquaculture and fisheries zones, bathing waters and other contact recreational waters.

6.3                                      Assessment Methodology

The construction method and sequence described in Section 3 were reviewed to assess the remoteness of the construction works from existing and committed Water Sensitive Receivers (WSRs).  The WSRs were identified according to guidance provided in the EIAO-TM and HKPSG.

The design of the Extension, construction sequence, duration and activities, and the operation, restoration and aftercare activities were reviewed to identify activities with the potential to impact upon identified WSRs and other water courses. 

Following the identification of WSRs and potential water quality impacts, the scale, extent and severity of potential net (ie unmitigated) construction, operation/restoration and aftercare impacts were evaluated, taking into account all potential cumulative effects including those of adjacent projects, with reference to the WPCO criteria.

Where net water quality impacts exceed the appropriate WPCO criteria, practical water pollution control measures/mitigation proposals were identified to ensure compliance with reference to the WPCO criteria.  Water quality monitoring and audit requirements were developed, if necessary, to ensure the effectiveness of the water pollution control and mitigation measures.

6.4                                      Water Quality Sensitive Receivers and Baseline Conditions

6.4.1                                Existing Conditions

The south-western part of the Extension Site will be located on formed land at TKO Area 137 (see Figure 6.4a).  It is currently occupied by a temporary fill bank, where the stormwater drainage channels have been well established.  The potential water quality impacts arising from decommissioning the fill bank has been assessed under the EIA for the Fill Bank ([1]) which concluded that no unacceptable residual water quality impacts were expected during the decommissioning period. 

The northern part of the Extension Site will be piggybacked onto the southern slopes of the existing SENT Landfill and the infrastructure area which includes the site office, the Bioplant and the landfill gas treatment plant (see Figure 6.4a).  These facilities will be demolished after those for the Extension have been constructed and commissioned ([2]).  

The eastern part of the Extension will occupy a small part of the Clear Water Bay Country Park (CWBCP) (see Figure 6.4a).  

Two seasonal streams, namely S1 and S2, were recorded within the Study Area.  As shown in Figure 6.4a, S1 and S2 are located at Ha Shan Tuk and Hin Ha Au respectively.  They are small in size, S1 and S2 are approximately 56 m and 98 m in length respectively.  Both of them are classified as seasonal streams because they were found to have limited water flows during the wet season and no water flows during the dry season.  Photographic records of the streams are illustrated in Figure 9.7d in Section 9 – Ecology Impact Assessment and it is concluded in Section 9 that the ecological significance of these two seasonal streams is considered to be low. 

6.4.2                                Existing Landfill Liner System

The existing SENT Landfill has been designed and constructed, as a secure containment facility incorporating a leachate containment system and a leachate collection system covering the entire waste boundary of the landfill.  As the site is lined, leachate within the landfill is collected and treated to ensure that there will be no off-site migration of leachate from the landfill to the environment.  The leachate containment and collection system comprises, from the bottom to the top, a layer of geocomposite groundwater drainage layer, a 1.5 mm of textured High Density Polyethylene (HDPE) Secondary liner, a 6 mm of geocomposite clay liner (GCL), a 2 mm of textured HDPE primary liner, a layer of non-woven geotextile cushion, a 200 mm granular leachate drainage layer, and a layer of woven geotextile filter.

6.4.3                                Existing Leachate/Wastewater Treatment Facilities

At present, the leachate/wastewater generated from the SENT Landfill is collected and delivered to the Bioplant for treatment prior to discharge to a sewer connecting to the TKO STW.  Landfill leachate is the predominant load whereas wastewater from the administrative office as well as laboratory and maintenance building is also collected and treated at the Bioplant.  The Bioplant comprises an equalization tank, a metals precipitation system, a leachate heater and heat exchanger, an ammonia stripping system, a pH adjustment system, a sequencing batch reactor (SBR) and sludge handling system.  Wastewater is pumped into the Bioplant and stored in the equalization tank.  The pH of the effluent from the equalization tank is elevated by adding lime slurry in order to precipitate out heavy metals.  After gravity clarification and filtration, the wastewater is heated and sprayed through the ammonia strippers.  In the ammonia strippers, hot air is blown through and it separates the ammonia from the wastewater.  The thermal catalytic unit will completely oxidise the off-gas that contains ammonia.  The pH of the wastewater is then adjusted and consequently pumped to the SBR to remove the remaining organic pollutants and ammonia.  The treated effluent is stored in an effluent holding tank and then discharged to the sewer at a rate not exceeding 210 m3 hr-1.

6.4.4                                Water Sensitive Receivers (WSRs)

In order to evaluate the water quality impacts resulting from the construction and operation/restoration and aftercare of the Extension, the WSRs have been identified in accordance with the EIAO-TM and HKPSG.

The WSRs in the Study Area are identified and presented below: 

·       Inshore waters in Junk Bay and Joss House Bay;

·       Clear Water Bay Country Park;

·       Surface water including two seasonal streams S1 and S2; and

·       Groundwater.

 

6.4.5                                Baseline Groundwater Conditions

 

Groundwater samples were collected from the locations GW1, GW2, GW3, and GW4 of the SENT landfill as shown in Figure 6.4a where GW1 and GW2 are up-gradient, and GW3 and GW4 are down-gradient.  Water samples from each location were collected once per week for a total of four consecutive weeks between June and July 2007.  The results of the groundwater baseline monitoring are presented in Annex F.

 

By comparing the groundwater quality of the up-gradient and the down-gradient monitoring wells, it is evident that the groundwater quality of the down-gradient wells is influenced by the influx of seawater.

6.5                                      Construction Phase Impact Assessment

6.5.1                                Potential Impacts

Potential sources of impacts to water quality from the construction activities are:

·       construction runoff;

·       wastewater generated from construction activities; and

·       sewage generated from the workforce.

6.5.2                                Construction Runoff

 

Construction runoff from site areas may contain high loading of SS and contaminants.  Potential water pollution sources from construction site runoff include:

·      runoff and erosion from site surfaces, earth working areas and stockpiles;

·      demolition of existing infrastructure for the SENT Landfill; and

·      tunnel excavation for the twin drainage tunnel.

 

Construction runoff may cause physical, biological and chemical effects.  Its physical effect can cause blockage of drainage channels due to the deposits of increasing SS from the site.  Chemical and biological effects are however highly dependent on its chemical and nutritional contents.  Runoff containing significant amount of concrete and cement-derived materials would lead to increasing turbidity and discoloration, elevation in pH, and accretion of pH solids. 

 

During the first year of construction, works including site formation and construction of site office buildings, workshops, landfill gas and leachate treatment plants will be carried out.  Excavation is necessary for the construction of the new infrastructure.   A perimeter cut-off channel will be constructed around the Extension Site to divert water from outside the site boundary before commencement of site formation works.  In addition, intercepting channels will be provided, for example along the edge of excavation to prevent stormwater runoff from washing across exposed soil surfaces.  The construction runoff will be discharged off site after passing through a sedimentation tank or silt traps.  It is anticipated that, with the implementation of good construction practice, as stated in ProPECC PN1/4, and appropriate mitigation measures (see Section 6.8), contamination of construction runoff will be minimal and there will be no unacceptable water quality impacts to the receiving water bodies, ie surface water including two seasonal streams S1 and S2, inshore waters in Junk Bay and Joss House Bay, and Clear Water Bay Country Park.

Modification of the landfill gas wells of the existing SENT Landfill is required for the accommodation of the new basal liners of the Extension.  No leachate leak will occur as the works will be carried out above the impermeable liner of the capping system. 

 

In the second year, the demolition of the existing infrastructure at the existing SENT Landfill will be carried out.  There will be no wastewater generated by the demolition of existing facilities.  As a preventive measure, all sewers and drains will be sealed to prevent building debris, soil and etc from entering public sewers/drains before commencing any demolition works. 

 

The fuel and waste lubricant oil from the on-site maintenance of machinery and equipment will be collected by a licensed chemical waste collector.  The runoff containing oil and grease will pass through the oil interceptor before being discharged off-site.

6.5.3                                Wastewater Generated from Construction Activities

In order to drain the surface water collected at the south-eastern corner of the Extension to the side slope near TKO Area 137, a 2 m-diameter twin drainage tunnel will be constructed near the side slope of the landfill, separated from the side slope liner system of the Extension by a considerable thickness of in-situ rock (see Figure 3.3i).  A micro-tunnel boring machine (TBM) will be used for the main tunnel excavation. 

There will be no wastewater generated from tunnel excavation, except the recycle water and bentonite slurry required for the cooling of the cutter head during boring rocks and soil respectively.  The recycle water will be conveyed to the sedimentation tanks for treatment and most of the treated water will be reused in the boring operations.  Similarly, the bentonite slurry will be recirculated, wherever practicable, following settlement of cuttings.  Only limited amount of excess water will be disposed to the surface drains in TKO Area 137 after proper treatment.  The disposal of the treated water in compliance with the discharge license granted at the later stage will be required and hence no adverse impact to the nearby water bodies is expected.  Prior to tunnel excavation, ground treatment works will be carried out and hence the tunnel excavation is unlikely to cause any unacceptable variation of the groundwater table.

 

Used bentonite slurries will be reconditioned and reused on-site as far as possible.  The residual bentonite slurry will be mixed with dry excavated material for disposal at the designated public filling facilities.  In accordance with ProPECC PN 1/94, if the used bentonite slurry is intended to be disposed of through the public drainage system, it should be treated to the respective effluent standards applicable to foul sewers, storm drains or the receiving waters as set out in the TM under the WPCO.

6.5.4                                Sewage Generated from the Workforce

 

Sewage will arise from the sanitary facilities provided for the on-site workforce.  The characteristics of sewage would include high levels of 5-day Biochemical Oxygen Demand (BOD5), ammonia and E.coli counts.  It is estimated maximum of 170 workers will be working simultaneously at the construction site during construction phase.  Sufficient chemical toilets will be provided for use by the workforce.  In addition, no sewage will be allowed to discharge directly into the surrounding water body without treatment.  With this regard, adverse impacts to water quality as a result of handling and disposal of sewage generated by the workforce are not expected.

6.6                                      Operation/ Restoration Phase Impact Assessment

6.6.1                                Potential Impacts

During the operation/restoration phase of the Extension, solid wastes deposited in the landfill will decompose by a combination of chemical, physical and biological processes through which solid, liquid and gaseous by-products are produced and all of them would be of concern in the overall management of a landfill.  The liquid by-product is referred to as leachate and is the main concern for the water quality impact of the Extension. 

Figure 6.6a shows the leachate generation processes.  There are two sources of water in the landfill, ie, the water present in the waste when landfilled (primary leachate) and the water added to the landfill from rainfall and groundwater inputs (secondary leachate).  During rainy days, the primary leachate is soon overshadowed by secondary leachate, which will control the long-term leachate generation.  Secondary leachate arises from infiltration of rainwater through the active tipping face and daily cover area.  As the landfill will be fully lined, no leachate will be generated from groundwater infiltration. 

Other potential impacts may include the wastewater produced during the daily operation of the office buildings and associated facilities.

To summarise, the potential sources of impacts to water quality from the operation/restoration activities include:

·      uncontrolled discharge of leachate from the active tipping area into surface water;

·      sub-surface off-site migration of leachate into groundwater and marine water due to potential pin holes and defected seams in the liner;

·      discharge of improper treated effluent leachate from the LTP; and

·      wastewater generated from workforce.

To evaluate the above potential impacts, it is necessary to examine in considerable detail of the surface water, groundwater and leachate management systems proposed for the Extension.  This will also facilitate the design of a monitoring programme which could determine the degree to which the Extension and any associated containment system is functioning in accordance with design objectives and in compliance with the legislative criteria/standards.

The potential impact to groundwater quality due to leakage of leachate is discussed as a whole for operation/restoration, and aftercare phases in Section 6.7. 

6.6.2                                Surface Water Management

 

As discussed in Section 6.6.1, leachate generated from rainfall infiltration will control the long-term leachate generation.  Surface water management, which relates to the infiltration of rainfall through the landfill surface, is discussed in this section. 

 

The overall design objectives for the surface water management system are to:

·       avoid any surface water runoff from outside the Extension (including runoff from the natural slopes of CWBCP and from restored slopes of the existing SENT Landfill) from entering the waste boundary;

·       ensure all runoff from the Extension site drains to Junk Bay (to the west), rather than streams S1 and S2, CWBCP and Clear Water Bay (to the east); and

·       ensure segregation between clean rainwater, and water which has come into contact with waste and therefore will be treated as leachate.

 

The following design features have been incorporated in the outline design of the surface water management to minimise leachate generation and control the discharge of leachate into surface water channels.

 

·       Clean surface water runoff will be separated from contact with waste by use of temporary bunds, diversion channels and cut-off drains; 

 

·       Areas that have been filled with waste, but not yet reached final grade, will be covered by intermediate cover (with an impermeable liner) to minimise rainwater infiltration into the waste and prevent erosion of the intermediate cover soil; 

 

·       The final cap (see Figure 3.3c) will include the following main features or similar materials to minimise rainwater infiltration into the waste:

(a)          A layer of CDV and topsoil mix - reduces infiltration into the waste and wind erosion and provides temporary moisture retention;

(b)          A layer of compacted fill - minimise infiltration into the waste through the cover;

(c)          A layer of geocomposite drainage layer - provides a lateral path for water to exit rapidly;

(d)          A layer of HDPE liner - an impermeable membrane effectively minimises infiltration into the waste and greatly reduces the volume of leachate to be generated from restored areas and seeping of leachate from waste slopes into surface water channels; and

(e)          A layer of non-woven geotextile - separates soil grading layer from HDPE liner.

 

·       Placement of the final capping system will be implemented in phases throughout the life of the Extension. 

 

Detailed design of site drainage will be based on the appropriate Hong Kong Government codes, including the DSD Stormwater Drainage Manual (1994).

A Drainage Impact Assessment (DIA) has been carried out for the Extension.  The DIA has concluded that the existing and planned surface water drainage infrastructure in TKO Area 137 and the surrounding area is adequate to convey surface water flows from the Extension and surrounding catchments to the existing and planned discharge points.  The estimated daily flow rates under normal and extreme conditions at different operational phases (refer to Section 3.6 for the details of each phase) of the Extension are summarised in Table 6.6a.

Table 6.6a      Predicted Daily Flow Rates During Operational Phases

Phase

Under Normal Conditions

Under Extreme Conditions

 

m3 d-1

m3 d-1

1

732

1,058

2

1,354

1,952

3

1,602

2,306

4

1,893

2,722

5

2,108

3,027

6

2,366

3,390

 

Surface Runoff from Clear Water Bay Country Park

In order to avoid surface runoff from CWBCP from entering the Extension Site, a permanent cut-off channel will be constructed along the crest of the eastern side slope.  The southern part of this cut-off channel will drain by gravity to the south-eastern corner of the Extension.  The northern part of this cut-off channel falls to the north, where it will meet up with the cut-off channel for the existing SENT Landfill.

 

At present, the existing SENT Landfill cut-off channel traverses the eastern edge of the landfill, and then turns to the west, towards the existing SENT Landfill infrastructure area.  As part of the Extension development, this portion of the channel will be covered by waste and could not be used during the operation/restoration phase of the Extension.  A 2 m-diameter twin drainage tunnel will be constructed near the side slope of the landfill, separated from the side slope liner system of the Extension by a considerable thickness of in-situ rock (see Figure 3.3i).  This twin tunnel will drain water collected at the south-eastern corner of the Extension to the side slope near TKO Area 137 where it joins the eastern boundary channel (see Figure 3.3i). 

 

Run-off from Existing SENT Landfill

Runoff from the restored slopes of the existing SENT Landfill will be uncontaminated but should be prevented from entering the Extension Site.  A perimeter cut-off channel will be constructed around the Extension Site, and it will be connected to the surface water drainage system to be incorporated into the existing SENT Landfill restoration design.  Following completion of the Extension, an additional channel (see Figure 3.3i) will be constructed around the eastern flank of the Extension and then to the west, to convey flows directly to the western boundary of the Extension Site avoiding the flow to the east to the Clear Water Bay.  Prior to completion of the Extension, collection and pumping of surface water will be required as part of the surface water management plan, to avoid any discharge of stormwater eastwards into CWBCP.

 

Rainfall within Extension Site

Rainfall within the Extension Site will be segregated depending on whether it has been in contact with waste (in which case it will be treated as leachate), or it is uncontaminated (in which case it will be dealt with as clean surface water).

A series of cut-off channels will be formed in the side slopes and on the southern waste slopes of the existing SENT Landfill that lie within the Extension Site boundary.  These channels will intercept rainwater falling on areas above the current level of waste placement, and divert it to the perimeter cut-off channels.

Areas outside the active tipping faces and daily cover area will be covered with an intermediate cover.  In order to minimise leachate generation and control seepage of leachate from waste slopes into the surface water drainage channels, the intermediate cover will include a layer of impermeable geomembrane.  Surface water management will be implemented to collect clean rainwater falling onto intermediate cover area, and divert it to the perimeter cut-off channels.

Rainwater falling onto the restored slopes of the Extension will be collected by surface water channels on the slopes, and drained to the perimeter of the site.  Rain falling onto the active tipping and daily cover areas will infiltrate through the waste and be collected by the leachate collection system, for treatment and discharge (see Section 6.6.5 for the details of leachate collection system).

 

Sediment Traps and Oil Separation

All surface water drainage channels that discharge either directly or indirectly to surface watercourses or to the sea will be provided with sediment traps, silting basins and oil separators (where necessary) to minimise the potential for contamination. 

To conclude, the design of the Extension has comprehensively considered minimising the infiltration of surface water into the landfill and avoiding seepage of leachate from the waste slopes into the surface water drainage system. 

6.6.3                                Groundwater Management

 

Generation of Leachate due to Groundwater Infiltration

As discussed in Section 3.3.1, the Extension will be designed and constructed as a containment facility incorporating a multi-layer composite liner system covering the entire land formation (compacted soil) of the Extension Site where waste will be deposited.  This will not only prevent infiltration of groundwater into the waste and hence minimising leachate generation, and also prevent off-site migration of leachate and contamination of the groundwater.  Construction quality assurance/control procedures will be implemented to ensure that the liner system is proper constructed (ie avoiding puncture of the impermeable HDPE liner by construction equipment during installation, and proper seaming of the joints, etc).  It is hence expected that the groundwater will be isolated from the Extension Site and as a result leachate generation from groundwater infiltration will be negligible.

 

Groundwater Contamination due to Leachate Seepage

 

A geocomposite groundwater drainage layer (as shown in Figure 3.3c) will be constructed underneath the basal lining system.  The compacted soil underneath the groundwater drainage layer will inhibit the downward infiltration of leachate into the groundwater and hence the drainage layer (with an adequate gradient) could allow the collected groundwater to flow horizontally by gravity.  Since the groundwater drainage layer of the Extension will be connected to groundwater diversion pipe trenches, the groundwater flows will be diverted to a series of groundwater collection sumps along the western boundary of the Extension adjacent to the leachate collection sumps.  The groundwater collection sumps (see Figure 3.3j) will be fitted with overflows to soakaways, and also with submersible pumps. 

The following measures will be implemented to avoid any groundwater contamination:

 

·       At present, groundwater monitoring is carried out at the monitoring wells and discharge manholes.  In order to much closely monitor the groundwater quality, it is proposed that groundwater quality at both the groundwater collection sumps and groundwater monitoring wells will be regularly monitored to check for contamination due to leakage of leachate from the Extension (details refer to Section 11 – EM&A).

 

·       If the monitoring data at the collection sumps show that there are no exceedences of the trigger levels, the groundwater retained in the sump will be discharged of from the sump to the soakaway and hence will percolate back into the groundwater.  Similar procedures are currently implemented in the existing SENT landfill for which the groundwater collected in the discharge manholes (if any) is pumped to the surface drains for disposal.  In accordance with the contractor of the existing SENT Landfill, only small amount of groundwater was found in the manhole over the past operational years and no overflows have ever been occurred.  In this regard, the monitoring frequency on a month basis is reasonably sufficient to determine the groundwater quality prior to the discharge to the soakaway.

 

·       In the event that the trigger levels are exceeded, the submersible pumps will pump groundwater into the leachate collection sumps, from where it will be transferred to the leachate treatment plant along with the leachate collected from the landfill.  Again, a similar mechanism is currently utilised in the existing SENT landfill. 

In the presence of these proactive prevention measures in place, the operation/restoration of the Extension would not impact the groundwater quality. 

6.6.4                                Leachate Management

As discussed above, the generation of leachate is mainly from the moisture content of the waste and rainwater infiltration.  As discussed in Sections 6.6.2 and 6.6.3, effective measures and facilities will be provided in the Extension to control surface water and groundwater entering the Extension and hence the leachate production will be reduced to minimal level.  This section assesses the effectiveness of the proposed leachate management system and the potential water quality impacts due to the handling, treatment and disposal of leachate.

The design objectives of the leachate management system are:

·       to contain all leachate within the waste boundary by the use of engineered barriers;

·       to collect and drain leachate for treatment and disposal; and

·       to facilitate the control of leachate levels within the Extension.

The leachate management system comprises the following components:

·       a leachate collection system;

·       a leachate extraction system; and

·       a leachate treatment system.

Each of these components is discussed below.

6.6.5                                Leachate Collection System

A low permeability composite liner system will be placed at the base of the Extension to reduce the discharge to the underlying hydrogeologic environment.  The liner system will be designed as a barrier to intercept leachate so that the contained leachate can be abstracted for treatment prior to discharge from the Extension Site. 

 

Basal Lining System

The basal lining system of the Extension will consist of the following features or similar materials (from top to bottom) (see Figure 3.3c):

·       a layer of filter geotextile;

·       a layer of leachate collection layer;

·       a layer of cushion geotextile;

·       an impermeable, such as the HDPE liner;

·       a geosynthetic clay liner (GCL);

·       a HDPE liner; and

·       a geocomposite groundwater drainage layer. 

The leachate collection layer will be designed to effectively collect and drain leachate which percolates downwards from the waste.  This is important as to reduce the leachate head above the liner system.  In order to fulfill these objectives, the leachate collection layer should:

·       have adequate hydraulic conductivity;

·       be resistant to physical and chemical damage;

·       have a sufficient gradient to allow drainage; and

·       contain pipework with appropriate spacing to facilitate removal of leachate.

The leachate collection layer will comprise a minimum depth of 500 mm crushed non-calcareous aggregate (10–20 mm size) of sufficient physical strength to withstand the likely loadings from the overlying waste (as determined by soaked 10% fines value).  The aggregate used will be rounded to minimise pressure on and damage to the liner system.  A layer of cushion geotextile will be placed between the leachate collection layer and the top of the impermeable liner.  A geotextile filter layer will be placed above the leachate collection layer to prevent downwards migration of fines from the waste.

The leachate collection layer will have a hydraulic conductivity of at least 1x10-4 m s-1, and a minimum gradient (vertical to horizontal) of 1:50.  The leachate collection layer will be placed on the basal liner with care, using a hydraulic excavator, to ensure that no damage is caused to the basal liner.   The pipework will be of sufficient physical strength to limit deflection to no more than 5%.  The thickness of drainage stone above the pipe will be at least equal to the diameter of the pipe.  Pipework will be jointed by butt-fusion welding to prevent leakage.  Access points will be maintained to enable jetting of the pipework to maintain its flow characteristics throughout the life of the Extension.

Drainage pipework will be installed within the leachate collection layer.  The pipework will be manufactured from either HDPE, u-PVC or polypropylene, and will be perforated (with slots or holes) except for the lower 120° of the pipe cross-section, which should be solid to allow for flow of leachate.  The pipe diameter will be determined based on the predicted flow (a minimum diameter of 200 mm is recommended to minimise clogging and allow for inspection and cleaning).

The leachate drainage pipework will be designed such that the maximum head of leachate does not exceed 1m.  In order to control the leachate head below this level, the maximum spacing of the collection pipes should be about 50m.  Otherwise, the leachate level may increase which may cause seepage of leachate through the side slopes and contamination of surface water.  Minimising the leachate head above the basal liner will also reduce the potential for leachate seepage through any potential pin holes/defective seams on the basal liner and hence reduce the potential for groundwater contamination.

 

Piggyback and Side Slope Lining System

At the piggyback and side slope areas, the leachate collection layer would comprise a geosynthetic drainage layer rather than crushed stone, and pipework would not be required.  Leachate collected at the geosynthetic drainage layer will flow down by gravity to the basal lining system, as described above, where leachate will be collected by the pipework. 

6.6.6                                Leachate Extraction System

Leachate will be extracted from the Extension via a series of four collection sumps around the western and southern perimeters of the Extension Site.

The leachate collection sumps will be constructed of pre-cast concrete and will be equipped with submersible pumps to enable leachate to be pumped from the base of the landfill to the leachate collection main, which will transfer leachate to the leachate treatment plant in the infrastructure area.

The leachate collection sumps will be accessed by upslope risers along the toe bund of the Extension, and therefore will not be prone to damage due to movements of the waste mass.

6.6.7                                Leachate Treatment System

Leachate Quantity

As discussed in Section 6.4.3, a Bioplant is currently operated at the existing SENT Landfill to treat the leachate as well as other wastewater generated from the SENT Landfill.  Before the commencement of the Extension, a new LTP will be constructed to handle the leachate and wastewater generated from the existing SENT Landfill.  The Bioplant will be demolished after all the leachate and wastewater from the existing SENT Landfill are diverted to the new LTP.  This LTP has a maximum design flow rate of 1,500 m3 d-1, coupled with a buffer storage capacity of 22,000 m3.  This design capacity is able to cope with the anticipated peak leachate treatment requirement during the last year of operation of the existing SENT Landfill.  The LTP is also capable of treating leachate to comply with the discharge standard stipulated in the discharge license of the existing SENT Landfill. 

Following full restoration and closure of the existing SENT Landfill, the leachate generation from the Extension will reduce significantly.  The buffer storage capacity could be reduced, subject to further review, as the leachate generation from the Extension is smaller.  It is estimated that the averaged combined leachate flow from the restored SENT Landfill and the operating Extension will be approximately 355 m3 d-1.  The peak treated effluent flow will be limited to 1,000 m3 d-1.  The treated effluent from the new LTP will be discharged to a foul sewer leading to TKO STW and the effluent should comply with the discharge standards stipulated in EPD’s Technical Memorandum Standards for Effluents Discharged into Drainage and Sewage Systems, Inland and Coastal Waters. 

Leachate Quality

The quality of leachate has been estimated based on the known composition of leachate at the existing Hong Kong landfills as well as the implementation of Sludge Treatment Facilities (STF) and the Integrated Waste Management Facilities (IWMF).  It is expected that the STF will commence operation in 2012 (before the commissioning of the Extension) whilst the IWMF is planned to be in operation by 2014.  It is understood that the residues (ie the incineration ash) of the STF and IWMF may be disposed of at other landfills in the future depending on the location of the STF and the IWMF which is under planning.  In order to provide a conservative impact evaluation, it is assumed that the Extension will receive the residues from STF and the IWMF in addition to the currently received waste, ie municipal solid waste and construction waste. 

It is not expected that the quality of leachate will be significantly affected by the implementation of STF and IWMF for the following reasons:

·       At present the existing SENT Landfill among the three strategic landfills in Hong Kong is the only one to accept stabilised incineration residues from thermal treatment facilities.  The existing SENT Landfill has proved capable of taking these residues without compromising its performance. 

·       Similar to the existing SENT Landfill Contract, the stabilised incineration residues have to meet the landfill disposal criteria before disposal to the landfill is allowed.  The criteria are set primary in terms of the Toxicity Characteristic Leaching Procedure (TCLP) limits as presented in Table E1 of the EPD”s Guidance Notes for Investigation and Remediation of Contaminated Sites of Petrol Filling Stations, Boatyards and Car Repair/Dismantling Workshops.  Although the volume the stabilised residues received by the Extension may be increased once the STF and IWMF are put into operation, the residues will have to pass the TCLP prior to disposal at the Extension.  The TCLP aims at stabilising the residues and minimising the leaching potential of heavy metals and other potentially toxic substances.  The potential of heavy metals leaching from the stabilised residues are expected to be low and hence will not adversely affect the leachate quality. 

·       Although the stabilised residues would produce inorganic leachate, the residues only form a portion of the total waste expected to be received by the Extension.  As a significant portion of the waste disposed of at the Extension will be MSW, it is expected that the leachate will consist of high levels of COD, BOD and ammoniacal-nitrogen similar to the leachate generated from the existing SENT Landfill.

 

Based on the available information, it is concluded that treatment requirements will be dictated by the removal of COD and nitrogen.  With reference to the performance of the existing SENT Landfill, when these parameters are properly treated, others such as heavy metals are usually found to be satisfactory in the effluent.  The predicted concentrations of the main design parameters of the raw leachate are shown in Table 6.6b.

Table 6.6b      Predicted Concentrations of the Main Design Parameters

Parameter

Unit

Mean

Maximum

Minimum

Influent NH4-N

mg L-1

2,500

4,500

1,500

Influent COD

mg L-1

3,000

4,500

2,000

Hard COD

mg L-1

1,000

1,500

650

Hard TKN

mg L-1

75

125

40

The treated effluent from the LTP will be discharged to the foul sewer leading to the TKO STW.  Effluent quality will be governed by the discharge standards stipulated in the TM.  The applicable limits for the averaged predicted flow of 355 m3 d-1 and the peak flow of 1,000 m3 d-1 are shown in Table 6.6c.


Table 6.6c      Effluent Discharge Standards Stipulated in the TM

Parameter

Unit

Flow rate, m3 d-1

 

 

>200 to ≤400

>400 to ≤600

>600 to ≤800

>800 to ≤1000

pH

-

6 - 10

6 - 10

6 - 10

6 - 10

Temperature

°C

43

43

43

43

Suspended solids

mg L-1

800

800

800

800

Settleable solids

mg L-1

100

100

100

100

BOD

mg L-1

800

800

800

800

COD

mg L-1

2,000

2,000

2,000

2,000

Oil & Grease

mg L-1

50

50

40

30

Iron

mg L-1

25

15

12.5

10

Boron

mg L-1

5

4

3

2.4

Mercury

mg L-1

0.1

0.001

0.001

0.001

Cadmium

mg L-1

0.1

0.001

0.001

0.001

Copper

mg L-1

3

1.5

1.5

1

Nickel

mg L-1

2

1.5

1.5

1

Chromium

mg L-1

2

1

0.7

0.6

Zinc

mg L-1

3

1.5

1.5

1

Silver

mg L-1

2

1.5

1.5

1

Other toxic metals individually

mg L-1

1.5

1

0.7

0.6

Total toxic metals

mg L-1

7

3

2

2

Cyanide

mg L-1

1

0.7

0.5

0.4

Phenols

mg L-1

1

0.7

0.5

0.4

Sulphide

mg L-1

10

5

5

4

Sulphate

mg L-1

1,000

1,000

1,000

1,000

Total nitrogen

mg L-1

200

200

200

200

Total phosphorus

mg L-1

50

50

50

50

Surfactants (total)

mg L-1

40

30

25

25

Source:  Table 1 - Standards for effluent discharged into foul sewers leading into Government sewage treatment plants, Technical Memorandum Standards for Effluents Discharged into Drainage and Sewerage Systems, Inland and Inshore waters

Leachate Treatment Options

Technical feasibility, space requirements as well as the implementation of STF and IWMF were considered to decide the treatment option.  Based on these considerations, it is proposed to treat leachate using a metal precipitation system, ammonia stripping towers (to remove the majority of ammoniacal nitrogen), followed by a sequencing batch reactor (SBR) operating in a “pre-denitrification” mode (for nitrification of the remaining ammoniacal nitrogen and subsequent COD removal and denitrification).

Buffer storage tanks prior to the metal precipitation system and ammonia stripping.  Stripped effluent will be stored in a separate holding tank from where it is fed into the SBR tanks.  Effluent from the SBR tanks will be stored in a final effluent holding tanks, from where it will be discharged to foul sewer.

As mentioned above, the leachate characteristics are not expected to be substantially changed during the operation of the Extension.  The stabilised incineration residues to be disposed of at the Extension are expected to be complied with TCLP limits ([3]) before disposal to the Extension and hence will not adversely affect the leachate quality.  Nevertheless, it is recommended, as a precaution, that a lysimeter study ([4]) be undertaken to confirm the metals concentrations that may occur at the proposed rates of disposal of stabilised incineration residues.  The aim of the lysimeter study is to study the change of leachate quality due to co-disposal of the IWMF residues at the SENT Extension.  If leachate from the lysimeter study does contain increased metals, it should be subjected to treatability trials to confirm if additional treatment process would be required (eg a metal precipitation system) to meet the TM effluent standards.  The metal precipitation system could be easily installed and there is available space at the LTP to install such system, if required.

In addition, the quality of LTP influents will be continuously monitored by the DBO Contractor to capture any change in the characteristics of the raw leachate.  Concurrently, the effluent quality will be monitored by the ET and the monitoring results will be sent to the DBO Contractor.  The DBO Contractor will review all the monitoring data to determine the removal efficiency of the treatment process and will decide whether modifications to the leachate treatment process are needed. 

Leachate Disposal

Treated leachate will be disposed of to the foul sewer leading to the TKO STW.  A Sewerage Impact Assessment (SIA) has been carried out as part of the Feasibility Study to confirm the capacity of the existing and planned sewage collection and treatment infrastructure in the surrounding area.  The SIA has confirmed that the existing and planned infrastructure is adequate for the predicted flows.

The disposal of treatment effluent, which meets the discharge standards stipulated in the TM, from the LTP into the foul sewer leading to the TKO STW will not cause adverse water quality impacts to the identified WSRs and the operations of the TKO STW.

6.6.8                                Wastewater Generated from the Workforce

Similar to the existing SENT Landfill, the wastewater from the administrative office as well as laboratory and maintenance buildings will be collected (about 22.5 m3) and treated together with leachate at the new LTP prior to disposal at the TKO STW.  Details of the treatment of the wastewater are presented in Section 6.6.7. 

It is anticipated that no adverse impacts on the surrounding aquatic environment due to the wastewater will arise. 

6.6.9                                Potential Risk Associated with Leakage of Leachate

 

A hydrogeological assessment has been undertaken as part of the Feasibility Study to evaluate the potential impact on groundwater quality and coastal water quality due to potential off-site migration of leachate from the Extension during the operation/restoration and aftercare phases.  The hydrogeological assessment takes into account the risk associated with leakage of leachate throughout the project lifetime.  As with all groundwater risk assessments for landfills, it is expected that a stringent Construction Quality Assurance Programme will be adopted during the installation of the liner system but for conservative assessment, it is assumed that there is still some degree of leakage through the geomembrane due to installation and manufacturing defects.  This is represented by a probability density function in the LandSim model representing numbers of pinholes, tears and holes in the geomembrane, initially starting at a minimum value and gradually increasing over time. 

 

The operation phase of the Extension (i.e. whilst it is still receiving waste) was included in the model.  Based on the modelling results for a double liner system, it will take considerable time for any leachate leakage, due to manufacturing defects and installation defect, to migrate through the engineered barrier layers and the unsaturated zone.  Risks to groundwater quality generally only occur during the post-closure period (this is discussed in Sections 6.7.3 and 6.7.4).  This approach is accepted by the UK Environment Agency for meeting the requirements of the European Union Groundwater Directive and Landfill Directive.

 

Rather than model the potential impacts of all possible contaminants in landfill leachate, UK Environment Agency Guidance ([5]) recommends modelling of representative parameters only to assess the worst case. Typically, these modelled parameters are present in the highest concentrations in leachate and/or are most mobile in the subsurface.  In this EIA study, the choice of contaminants to be modelled was referenced to the available data on leachate quality at SENT Landfill.  The modelled parameters were chosen to be representative of the key contaminants in leachate, which are widely accepted as being ammoniacal nitrogen, chloride, COD and toxic metals and they are listed below:

 

·      inorganic cations (ammoniacal nitrogen);

 

·      inorganic anions (chloride);

 

·      highly mobile metallic ions (zinc);

 

·      less mobile metallic ions (cadmium, mercury); and

 

·      representative of organic species in leachate (COD).

 

For Cadimum and Mercury, where no existing information is available, the concentrations of these contaminants in leachate are taken from the UK default leachate inventory.

 

The background concentrations of contaminants are taken from the down-gradient groundwater monitoring undertaken in June and July 2007 (see Annex F), and the concentrations in leachate taken from leachate monitoring at existing SENT Landfill. 

 

Model Assumptions for LandSim Model are summarised in Table 6.6d.

Table 6.6d      Model Assumptions for LandSim Model

Leachate Composition and Groundwater Concentrations

 

Parameter

Leachate Composition (mg L-1) (a)

Mean Groundwater Concentrations (mg L-1) (b)

Ammonia as N

1,788 – 2,460

2.15

Cadmium

0.0019 - 0.105 (d)

<0.1

Chloride

1,971 – 2,558

14,588

Mercury

0.00004 – 0.00195 (d)

- (c)

Zinc

0.34 – 3.83

<0.1

COD

2,420 – 3,201

27

Sources:

(a)    With reference to the leachate composition at the existing SENT Landfill.

(b)    The mean value of groundwater data taken at the down-gradient groundwater stations (GW3 and GW4 as shown in Table 6.4b) during June – July 2007.

(c)    The background concentrations for mercury are not currently available.

(d)    The leachate concentrations of cadmium and mercury are currently not available and hence the UK default leachate inventory was used.

 

Defects at Barrier (a)

 

Defects (b)

 

 Nnumber per hectare

Upper Layer (design thickness of 0.002m)

Pin Holes

Minimum 0, Maximum 25

Holes

Minimum 0, Maximum 5

Tears

Minimum 0, Most Likely 0.1, Maximum 5

Lower Layer (design thickness of 0.002m)

Pin Holes

Minimum 0, Maximum 25

Holes

Minimum 0, Maximum 5

Tears

Minimum 0, Most Likely 0.1, Maximum 2

Notes:

(a)     The actual proposed lining is not one of the default systems in the LandSim Model and hence it was necessary to simulate a double composite lining system, which is considered to be comparable to the actual proposed design in view of a similar total thickness of GCL and two layers of HDPE.

(b)     The defects include manufacturing defects and installation defects.

 

The modelled flow rates of leakage through the basal liners and the flow characteristics of the aquifer (including contaminant transport) are calculated by the LandSim model on a probabilistic basis.  Since the parameters are represented by probability density functions rather than single values, it is not appropriate to refer to single values for leakage through the basal liners or aquifer flow.  Rather, results are expressed as a percentile (usually 95th) of the output distribution at a particular time.  In addition, certain parameters (including those influencing leakage through the engineered barrier system) change with time.  Based on the model results, the flow rates of leakage through the basal liners for the 1st through to the 100th year after the operation of the Extension commenced were predicted to be negligibly small (ie in a range of 0 to 1.32E-278 L day-1).  The negligible leakage flow rate over this period reflects the leachate level within the landfill.  As mentioned in Section 6.6.5, the leachate heads in the landfill will be maintained at a minimum level (below 1 m) by leachate extraction and treatment during the Extension contract ([6]) and consequently this will minimise the leakage flow rate.

The groundwater within the fill deposits in the TKO Area 137 will flow westerly and eventually enter the inshore waters in Junk Bay, as shown in Figure 6.6a.  Any contaminated groundwater discharged from the Extension may potentially cause a water quality impact in the Junk Bay.  The development of the Extension will not have adverse impact to the groundwater flow to Joss House Bay which is located to the east of the Extension Site.

The LandSim model was used to evaluate the potential water quality impacts of the leachate leakage from the Extension through the groundwater to the Junk Bay.  Concentrations of the key pollutants at the Junk Bay inshore waters 1 year and 5 years after commencement of the Extension, were modelled without retardation during transport (see Table 6.6e).  The results are compared against the TM standards for effluents discharged into the inshore waters of Southern, Mirs Bay, Junk Bay, North Western, Eastern Buffer and Western Buffer Water Control Zones, with an estimated flow rate in the aquifer at TKO Area 137 of 500 m3 d-1.

 

Table 6.6e      Predicted Contaminant Concentrations at the Inshore Waters in Junk Bay during Operation Phase

Parameter

Concentration after 1 year (mg L-1)

Concentration after 5 years (mg L-1)

Water quality standard (b) (mg L-1)

 

Without Groundwater Background

With Groundwater Background

Without Groundwater Background

With Groundwater Background

 

Ammonia as N

0 (a)

2.15 (c)

0 (a)

2.15 (c)

80 (as total N)

Cadmium

0 (a)

<0.1 (c) (e)

0 (a)

<0.1 (c) (e)

0.001

Chloride

0 (a)

14,588 (c)

0 (a)

14,588 (c)

N/A

Zinc

0 (a)

<0.1 (c)

0 (a)

<0.1 (c)

N/A

Mercury

0 (a)

- (d) (e)

0 (a)

- (d) (e)

0.001

COD

0 (a)

27 (c)

0 (a)

27 (c)

80

Notes:

(a)      The LandSim model does not predict the presence of any contaminants at the Junk Bay inshore water (see Figure 6.4a) within the timeframe modelled, hence the concentrations are expressed as zero rather than as being below a certain detection limit.

(b)      It is based on the predicted groundwater flow rate of 500 m3 d-1 in the aquifer at TKO Area 137 and in accordance with Table 10a - Standards for effluent discharged into inshore waters of Southern, Mirs Bay, Junk Bay, North Western, Eastern Buffer and Western Buffer Water Control Zones.

(c)    The mean value of groundwater data taken at the down-gradient groundwater stations (GW3 and GW4 as shown in Annex F) during June – July 2007.

(d)    The background concentrations for mercury are currently not available.

(e)    For the EM&A for the Extension, the detection limit of cadmium and mercury will be revised in order to allow a direct comparison with the TM standards.

 

Table 6.6e shows that no pollutants released from the Extension will be observed at the Junk Bay inshore waters after 1 and 5 years operation, ie during the operation phase of the Extension.  It indicates that the operation of the Extension is not expected to impact the surrounding water bodies.  It also indicates that the environmental risks associated with leakage of leachate posed to the aquatic environment are very low. 

 

Contingency Plan for Accidental Leakage of Leachate

 

As discussed above, the modelling results show that the environmental risks due to leachate leakage, under the predictable situation such as degradation of the cap or basal liners, are very low.  For the accidental leakages due to, for example, rupture of leachate pipelines, failure of pipe joint sealing and damage of geomembrane, their impacts on the groundwater could be substantially reduced if the contingency plan is well-developed before the operation of the Extension and followed efficiently by the DBO Contractor during the operation. 

 

Monitoring for surface water, groundwater, leachate levels and treated effluent will be implemented throughout the operation/restoration phase.  The objective of the monitoring programme is to continuously check the performance of the Extension and the effectiveness of mitigation measures.  The monitoring programme will also effectively provide an early indication should any accidental leakage of leachate occur.  The contingency plan will be implemented once the monitoring results indicate any exceedances of pre-defined trigger levels.  Details of the determination of the trigger levels should refer to the EM&A Manual. 

 

A comprehensive contingency plan has been established for the existing SENT Landfill.  Wherever applicable, the contingency plan is recommended to adopt the existing contingency plan as the basis for the Extension’s.  The contingency procedures include:

 

·      To establish a Special Environmental Monitoring Plan (SEMP) to determine the likely cause or reason for exceedances or non-compliances, any alterations and modifications to the works, operations and aftercare to reduce the likelihood of the violations, the anticipated outcome of any corrective action programme;

 

·      To identify the source that causes the exceedances and implement a corrective action programme should the Extension cause the exceedance;

 

·      To notify in writing all relevant parties and persons including those are being affected by the incidents.

 

The following modifications are, however, recommended in order to developing a contingency plan particularly suitable for the Extension:

 

·       At present, groundwater monitoring is carried out at the monitoring wells and discharge manholes.  In order to much closely monitor the groundwater quality, it is proposed that groundwater quality at both the groundwater collection sumps and groundwater monitoring wells will be regularly monitored to check for contamination due to leakage of leachate from the Extension.

 

·      Groundwater monitoring wells will be installed at up-gradient and down-gradient of the Extension.

 

·       In the event that the trigger levels are exceeded, the submersible pumps will pump groundwater into the leachate collection sumps, from where it will be transferred to the LTP along with the leachate collected from the landfill. 

 

·      Surface water monitoring stations will be located at three discharge points at western side of the Extension.

 

With the prompt and effective implementation of the contingency plan, it is not expected that adverse impact on groundwater and hence coastal water will arise from the Extension operation.

6.7                                      Aftercare Phase Impact Assessment

6.7.1                                Potential Impacts

Upon completion of final filling and site restoration, the period of aftercare will begin and last for 30 years.  During this period, leachate will continue to be generated.  The established leachate control measures and treatment will continue to operate throughout the aftercare period.

In the previous sections, it has been mentioned that the components of the leachate management system will prevent leachate from seeping from the side slopes to the surface drainage channels and off-site migration from the basal and side slope containment systems.  In addition, proper site maintenance will be undertaken during the aftercare period to ensure that the capping system, leachate collection system and treatment system will be performed to comply with the design requirements.  Surface water, groundwater and effluent quality monitoring will also be undertaken during the aftercare period in accordance to the monitoring plan. 

With the presence of the muti-layer capping and basal liner systems, proper site maintenance and regular monitoring, the probability of the leachate leakage from the capping system and containment system is expected to be very low.  Nevertheless, the water quality impacts of potential leakage of leachate have been assessed and are discussed below.

A hydrogeological model (using the latest version of the LandSim model, version 2.2.15, Environment Agency of England and Wales, 2004) was used to assess the potential impacts of the Extension on surface water and groundwater quality using a number of very conservative assumptions, based on Hong Kong and overseas experience. 

6.7.2                                Evaluation of Potential Impacts on Surface Water

The latest version of the LandSim model allows for long-term degradation of the performance of capping systems and for ultimate cessation of active leachate control measures.  In the Hong Kong context, these factors can be negated by ensuring adequate aftercare, and following expiry of the aftercare period, by regular monitoring and maintenance of the capping system and extraction and treatment of leachate from the Extension until the landfill is stabilised.  The LandSim model has also used a conservative figure for estimating initial infiltration through the landfill cap, based on the requirements of groundwater modelling in the UK.  The application of these conservative approaches in this case ensures a robust and conservative assessment.

 

The hydrogeological assessment concludes that, whilst the cap remains intact (for more than 100 years) and leachate control is maintained, there are no significant impacts on surface water quality. 

The surface breakout could be avoided by mitigation measures such as necessary maintenance or replacement of the HDPE cap to prevent degradation and by on-going active management of leachate to control the leachate head at a maximum of 1 m.

Based on the above, the potential surface breakwater is unlikely to occur in the presence of the active leachate management and mitigation measures.  

6.7.3                                Evaluation of Potential Impacts on Groundwater

Potential risks to groundwater quality will be associated with leakage of leachate from the Extension Site through the basal and side slope lining systems into the underlying fill deposits of the TKO Area 137 (see Figure 6.6a). 

Based on the geology and topography of the Study Area, it is likely that almost all groundwater flow will be occurring within the TKO Area 137 fill rather than in the underlying saturated marine or alluvial deposits.  The base of the landfill has been kept above the groundwater level to minimise the consequences of any leakage from the lining system.  With addition of the multi-layer basal liner system on top of compacted soil isolating the Site from the groundwater, the likelihood of the leachate leakage to the groundwater system is considered to be very low.

In addition, the down-gradient groundwater within the fill deposits is not considered a resource and it is not generally used as a potable water supply.  In the vicinity of the Extension, potable water supplies are not extracted from the groundwater.  Water is supplied to the villages at Shek Miu Wan via a pipeline from Clear Water Bay and natural stream flows are used at the villages of Po Toi O.  Hence in the unlikely event of any leachate leakage to groundwater, the severity of such impacts is considered to be very low.

The leachate leakage to the groundwater will be prevented by effective leachate management (see Sections 6.6.4 to 6.6.7) as well as full implementation of a monitoring programme.  Routine monitoring of the groundwater quality is recommended to detect any leachate leakage and if it is the case, appropriate and adequate remedial measures should be implemented. 

6.7.4                                Evaluation of Potential Impacts on Coastal Waters

The groundwater within the fill deposits in the TKO Area 137 will flow westerly and eventually enter the inshore waters in Junk Bay, as shown in Figure 6.6a.  Any contaminated groundwater discharged from the Extension may cause a water quality impact in the Junk Bay.  The development of the Extension will not have adverse impact to the groundwater flow to Joss House Bay which is located to the east of the Extension Site.

A hydrogeology assessment using the LandSim model has been carried out to evaluate the potential water quality impacts of the leachate leakage from the Extension through the groundwater to the Junk Bay.   

The quality of leachate from the existing SENT Landfill was used as a basis for the assessment.  The baseline groundwater condition was determined from the groundwater baseline monitoring results for the Extension, as presented in Section 6.4.5, by taking the mean of the groundwater data at two down-gradient stations.  Both leachate composition and groundwater concentrations were used for the LandSim model as the model assumptions and are summarised in Table 6.6d.   Leachate monitoring at the existing SENT Landfill does not include toxic metals (other than zinc), which are included in the TM effluent discharge standards.  The concentration of cadmium is hence referred to the UK commonly used leachate inventory, ie 0.0019 to 0.105 mg L-1.

Concentrations of the key pollutants at the Junk Bay inshore waters 10 years and 50 years after commencement of the Extension, were modelled without retardation during transport (see Table 6.7a).  The results are compared against the TM standards for effluents discharged into the inshore waters of Southern, Mirs Bay, Junk Bay, North Western, Eastern Buffer and Western Buffer Water Control Zones, with an estimated flow rate in the aquifer at TKO Area 137 of 500 m3 d-1.

Table 6.7a shows that no pollutants released from the Extension will be observed at the Junk Bay inshore waters after 10 and 50 years operation.  The hydrogeological assessment concludes that, whilst the cap remains intact and leachate control is maintained, there are no significant impacts on groundwater quality.  Even in the very long term (on a timescale of several hundred years), and assuming cap degradation and cessation of active leachate control, the impacts on groundwater are still predicted to be slight, and groundwater discharges to Junk Bay would still remain within the effluent discharge standards stipulated in the Water Pollution Control Ordinance for the relevant estimated flow rates.

Table 6.7a      Predicted Contaminant Concentrations at the Inshore Waters in Junk Bay after 10 year and 50 years since the commencement of the Extension

Parameter

Concentration after 10 years (mg L-1)

Concentration after 50 years (mg L-1)

Water quality standard (b) (mg L-1)

 

Without Groundwater Background

With Groundwater Background

Without Groundwater Background

With Groundwater Background

 

Ammonia as N

0 (a)

2.15 (c)

0 (a)

2.15 (c)

80 (as total N)

Cadmium

0 (a)

<0.1 (c) (e)

0 (a)

<0.1 (c) (e)

0.001

Chloride

0 (a)

14,588 (c)

0 (a)

14,588 (c)

N/A

Zinc

0 (a)

<0.1 (c)

0 (a)

<0.1 (c)

N/A

Mercury

0 (a)

- (d) (e)

0 (a)

- (d) (e)

0.001

COD

0 (a)

27 (c)

0 (a)

27 (c)

80

Notes:

(a)        The LandSim model does not predict the presence of any contaminants at the Junk Bay inshore waters (see Figure 6.4a) within the timeframe modelled, hence the concentrations are expressed as zero rather than as being below a certain detection limit.

(b)       It is based on the predicted groundwater flow rate of 500 m3 d-1 in the aquifer at TKO Area 137 and in accordance with Table 10a - Standards for effluent discharged into inshore waters of Southern, Mirs Bay, Junk Bay, North Western, Eastern Buffer and Western Buffer Water Control Zones.

(c)    The mean value of groundwater data taken at the down-gradient groundwater stations (GW3 and GW4 as shown in Annex F) during June – July 2007.

(d)    Mercury is not included in the table since the background concentrations for mercury are not available.

(e)    For the EM&A for the Extension, the detection limit of cadmium and mercury will be revised in order to allow a direct comparison with the TM standards.

6.8                                      Construction Phase Mitigation Measures

6.8.1                                Construction Runoff

Exposed soil areas will be minimised to reduce the contamination of runoff and erosion.  As mentioned in Section 6.5.2, site formation and excavation for the new infrastructures is required.  To prevent stormwater runoff from washing across exposed soil surfaces, perimeter channels will be constructed in advance of site formation works and earthworks and intercepting channels will be provided for example along the edge of excavation.  Silt removal facilities, channels and manholes should be maintained and the deposited silt and grit should be removed regularly to ensure they are functioning properly at all times.  Temporary covers such as tarpaulin will also be provided to minimise the generation of high SS runoff.  The surface runoff contained any oil and grease will pass through the oil interceptors.

 

In the second year, the demolition of the existing infrastructure at the existing SENT Landfill will be carried out.  There will be no wastewater generated by the demolition of existing facilities.  As a preventive measure, all sewer and drains will be sealed to prevent building debris, soil and etc from entering public sewers/drains before commencing any demolition works. 

During the excavation works for the twin drainage tunnels, the recycle water for cooling the cutter head of the TBM will be conveyed to the sedimentation tanks for treatment and most of the treated water will be reused, where applicable and as much as possible, in the boring operations.  The disposal of the treated water in compliance with the discharge license granted at the later stage will be required.

The fuel and waste lubricant oil from the on-site maintenance of machinery and equipment will be collected by a licensed chemical waste collector. 

The runoff contained oil and grease will pass through the oil interceptor before being discharged off-site. In addition, control measures, including implementation of excavation schedules, lining and covering of excavated stockpiles will be implemented to minimise contaminated stormwater run-off from the Extension site.

6.8.2                                Sewage Effluents

Sufficient chemical toilets should be provided for the construction workforce.  Untreated sewage should not be allowed to discharge into the surrounding water body.  A licensed waste collector should be employed to clean the chemical toilets on a regular basis. 

6.9                                      Operation/Restoration Phase Mitigation Measures

6.9.1                                Surface Water Management

Inspections of the drainage system, sand traps, settlement ponds and surface water channels should be performed regularly to identify areas necessary for maintenance, cleaning or repair.  Regular maintenance and replacement, if required, of the HDPE liner should be conducted to prevent degradation from affecting the performance of the capping system.  Monitoring of surface water quality should be conducted on a regular basis (see Section 11 for the monitoring requirements).

6.9.2                                Groundwater Management

The groundwater management facilities including the groundwater monitoring wells and the groundwater collection sumps will be inspected regularly during the routine groundwater monitoring programme.  Monitoring of groundwater quality will be conducted on a regular basis (see Section 11 for the monitoring requirements).

6.9.3                                Leachate Management

The leachate pump houses and related ancillary equipment should be inspected regularly and repaired, if necessary.  For equipment such as pumps that require routine scheduled maintenance, the maintenance should be performed following the manufacturer’s recommended frequency.  Monitoring of leachate levels above the basal liner and leachate quality should be conducted on a regular basis (see Section 11 for the monitoring requirements).

 

The design of the LTP has included two identical treatment trains (each with a treatment capacity of 750 m3 d-1).  Taking account of the predicted average combined leachate flow (about 355 m3 d-1) from the Extension and the restored SENT Landfill, there will be sufficient redundancy in the system to handle the anticipated leachate flow even if one treatment train is down for maintenance.  Preventive maintenance will be implemented so that the possibility for forced shutdown during the wet season will be kept to minimum.  However, emergency procedures or a contingency plan should be established should the LTP malfunction.  It may require that the leachate be stored temporarily within the landfill if the leachate buffer tanks are full and leachate cannot be transported to the LTP for treatment.  However, it is considered that the likelihood of this situation is very remote. 

6.10                                  Aftercare Phase Mitigation Measures

 

6.10.1                            Potential Leakage of Leachate

As discussed in Section 6.7.4, the assessment indicates that there will be no adverse impact on groundwater quality entering Junk Bay as a result of potential leachate leakage from the Extension.  Regular groundwater quality monitoring should be carried out to monitor the performance of the leachate containment system.  Maintenance and replacement of the capping system should be carried out, if necessary, to prevent leachate seepage in the event of a damaged cap.

In addition, long term measures to prevent any surface breakout of leachate include maintaining control of the leachate level through extraction; and/or maintaining the engineered capped system to control infiltration.

6.11                                  Cumulative Impact Assessment

 

The confirmed concurrent project in the vicinity of the Extension is the existing SENT Landfill and hence the following cumulative impact assessment will focus on the concurrent activities of the two landfills during different phases of the Extension. 

 

On the other hand, neither sufficient project details nor consolidated programme for the TKO Area 137 is available at the completion of this EIA Study.  It is uncertain whether any activities or what kind of activities, if any, will occur concurrently with the Extension activities and hence it will not be further discussed.

6.11.1                            Construction Phase

 

As discussed in Section 6.5, the construction runoff and the sewage generated by the construction activities and workforce will be well controlled with full implementation of mitigation measures.  No cumulative impacts are hence expected to occur during the construction phase.

 

6.11.2                            Operation Phase and Restoration Phase

 

During the first quarter of the operation at the Extension, restoration at the last filling area at the existing SENT Landfill will be undertaken.  In other words, the two landfills will not receive waste simultaneously.  It is hence not anticipated that cumulative impact will result from the concurrent operation of two landfills.

 

The clean surface runoff from the restored SENT Landfill will be intercepted and diverted away from the Extension.  As mentioned in Section 6.6.2, the surface runoffs generated at the Extension will be well managed by the proposed systems such as the perimeter drains to be provided surrounding the Extension and twin drainage channels to divert the collected surface water to the western side of the Extension.  The surface runoffs from the Extension will not flow into and influence the existing SENT Landfill.  Cumulative impacts on the surface water are hence not expected to be resulted by the operation of the Extension.

 

Cumulative impact on leachate treatment has also been assessed and discussed in Section 6.6.7.  The new LTP will be commissioned during the last year of operation at the existing SENT Landfill and will replace the existing Bioplant of the SENT Landfill.  A buffer storage tank with a capacity of 22,000 m3 will be provided and it will be able to cope with the anticipated peak leachate volume during the last year operation of the existing SENT Landfill when the existing Bioplant will be demolished, and subsequently during the Extension operation.  The LTP is capable of treating leachate generated from the existing SENT Landfill to comply with the discharge standards stipulated in the existing discharge license of the SENT Landfill.  Following closure and restoration of the existing SENT Landfill, leachate generation at the existing SENT Landfill will be reduced significantly.  It is estimated that the averaged combined leachate flow from the restored SENT Landfill and the operating Extension will be around 355 m3 d-1 while the peak flow will be less than 1,000 m3 d-1.  The LTP is capable of treating leachate to comply with discharge standards stipulated in EPD’s Technical Memorandum Standards for Effluents Discharged into Drainage and Sewage Systems, Inland and Coastal Waters.  It is therefore not expected that any cumulative impacts on leachate treatment will occur.

6.11.3                            Aftercare Phase

 

During the aftercare phase of the Extension, leachate will continue to be generated from both landfills but the leachate generated from the Extension is expected to be sufficiently reduced (see Section 6.6.7).  The established leachate control measures and treatment will continue to operate throughout the aftercare period of the Extension.  As discussed in Section 6.7, proper site maintenance will be undertaken during the aftercare period to ensure that the capping system, leachate collection and treatment systems will be performed to comply with the design requirements.  Surface water, groundwater and effluent quality monitoring will also be undertaken during the aftercare period in accordance to the monitoring plan.  With the provisions of all these control and monitoring systems, no cumulative impacts are expected to occur during the aftercare phase.

6.12                                  Summary of Environmental Outcomes and Conclusion

The potential impacts due to construction operation/restoration and aftercare of the Extension on surface water, groundwater and marine water quality have been assessed. 

The assessment indicates that with the implementation of the proposed design for surface water management system and recommended mitigation measures, there will be no unacceptable water quality impacts due to the construction activities.

With the proposed surface water, leachate and groundwater management systems and international good practice for landfill operation, the operation and restoration of the Extension will not result in adverse water quality impacts on the identified water sensitive receivers.  Discharge of treated effluent, which complies with the TM standards, to the foul sewer leading to the TKO STW will not cause adverse water quality impacts.

During the aftercare phase, the Extension Contractor will be responsible for operating the leachate collection system and LTP, and maintenance of the final capping system for 30 years.  Regular inspection of the capping system should be carried out to ensure that its integrity and performance meet the design requirements and that there is no leachate seepage from the cap.

The hydrogeological assessment concludes that, whilst the cap remains intact and leachate control is maintained, there are no significant impacts on groundwater quality.  Even in the very long term (on a timescale of several hundred years), assuming cap degradation and cessation of leachate control, the impacts on groundwater are predicted to be slight, and groundwater discharges to Junk Bay will remain within the limits.

6.13                                  Environmental Monitoring and Audit (EM&A) Requirements

Based on the impact assessment as detailed in Section 6.5, no adverse impacts are predicted provided that mitigation measures as recommended in Section 6.8 are fully implemented.  Monitoring of surface water and site inspections are recommended to be carried out during the construction phase in order to check the environmental performance of the construction works on a regular basis.

To monitor the performance of the operation/restoration and aftercare of the Extension, it is recommended to monitor the quality of surface water and groundwater at the monitoring wells and collection sumps at the Extension, and the effluent discharged from the LTP.   Detailed of the monitoring requirements are described in Section 11 of this Report and in the Environmental Monitoring and Audit Manual.


 



([1])        CH2M Hill (China) Ltd (2002).  Environmental Impact Assessment for the Proposed Fill Bank at Tseung Kwan O Area 137.  For Civil Engineering Department.  Final EIA Report.

([2])         The facilities for the existing SENT Landfill will be re-provided either as new separate facilities or a combined facilities with those of the Extension at the new infrastructure area.

([3])     Guidance Notes for Investigation and Remediation of Contaminated Sites of: Petrol Filling STations, Boatyards and Car Repair/Dismantling Worship, EPD, 1999.

([4])     A small cell (with separated containment and leachate collection system) will be constructed at the landfill.   The cell will be filled with the anticipated ratio of the stabilized IWMF residues and MSW to be disposed at the Extension.  The leachate collected from the lysimeter cell will be analysed (particularly for heavy metals) to see if there is any significant change in leachate quality due to the co-disposal arrangement.  The monitoring data will help to refine the treatment process and if necessary modify the treatment process of the LTP to ensure that the effluent will comply with the required discharge standards. 

 

([5])         Hydrogeological Risk Assessment for Landfills (Environment Agency, March 2003)

([6])         Incluidng a 30-year aftercare period after the closure of the Extension.  Further restoration contracts may be let for continuous management of the restored Extension until the landfill is stablised.