Environmental Impact Assessment: Proposed Reverse Osmosis ... of... · Reverse Osmosis (RO) seawater desalination plant be constructed to meet this water demand. In terms of existing

Environmental Impact Assessment: Proposed Reverse Osmosis Plant, Iron –ore Handling

Facility, Port of Saldanha

Specialist Marine Impact Assessment

Prepared for:

PD Naidoo and Associates Pty (Ltd) & SRK Consulting Engineers and Scientists Joint Venture (PDNA/SRK JV)

Jointly prepared by:

CSIR, Natural Resources and the Environment, PISCES Environmental Services (Pty) Ltd and

Nina Steffani, Marine Environmental Consultant

February 2008

Env i r onmental Ser v ices (Pt y ) Lt dPISC ES

This page intentionally left blank

CSIR/NRE/ECO/ER/2007/0149/C (Draft)

Prepared for:

PD Naidoo and Associates Pty (Ltd) and SRK Consulting Engineers and Scientists Joint Venture (PDNA/SRK JV)

On behalf of: Transnet Capital Projects This report was compiled by: Roy van Ballegooyen CSIR, Natural Resources and the Environment

PO Box 320, Stellenbosch 7599, South Africa Tel: + 27 21 888 2572 Fax: + 27 21 888 2693 Email: [email protected]

Dr Andrea Pulfrich PISCES Environmental Services (Pty) Ltd

PO Box 31228, Tokai 7966, South Africa Tel: + 27 21 782 9553 Fax: + 27 21 782 9552 Email: [email protected]

Nina Steffani Marine Environmental Consultant

21 Skippers End, Zeekoevlei 7941, South Africa Tel: + 27 21 705 9915 Fax: + 27 21 705 9915 Email: [email protected]

Published by:

CSIR P O Box 395 0001 PRETORIA Republic of South Africa

Issued and printed by, also obtainable from:

CSIR P O Box 320 STELLENBOSCH 7599 South Africa

Tel: + 27 21 888-2400 Fax: +27 21 888-2693 Email: [email protected]

The report to be cited as: Van Ballegooyen, N. Steffani and A. Pulfrich 2007. Environmental Impact Assessment:

Proposed Reverse Osmosis Plant, Iron –ore Handling Facility, Port of Saldanha - Marine

Impact Assessment Specialist Study, Joint CSIR/Pisces Report,

CSIR/NRE/ECO/ER/2007/0149/C, 184pp + 198pp App (Draft).


Marine Impact Assessment Specialist Study

EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page i

SCOPE OF WORK

The Iron Ore Handling Facility at the Port of Saldanha (including the proposed upgrades to the

facility), will require increased water supplies for dust suppression measures. It is proposed that a

Reverse Osmosis (RO) seawater desalination plant be constructed to meet this water demand. In

terms of existing legislation, this proposed activity (i.e. the construction and operation of an RO Plant)

requires that a Basic Assessment of potential environmental impacts be undertaken.

PD Naidoo & Associates (Pty) Ltd and SRK Consulting Scientists and Engineers Joint Venture

(PDNA/SRK Joint Venture) have been appointed by Transnet to undertake a Basic Assessment for a

proposed RO plant at the Port of Saldanha. The PDNA/SRK JV, in turn, have contracted the CSIR to

undertake the Specialist Marine Impact Assessment for the Basic Assessment. CSIR has sub-

contracted PISCES Environmental Services (Pty) Ltd and Nina Steffani (Independent Consultant) to

provide the ecological assessment component of this specialist study, while the Coastal Systems

Research Group of the CSIR Natural Resources and the Environment Unit, has undertaken the water

quality modelling to characterise the predicted changes in water quality associated with the proposed

discharges from the RO Plant.

Under this agreement the CSIR and its sub-consultants are to:

• Advise on the design of the RO Plant early in the process;

• Undertake an assessment of potential environmental impacts in the marine environment

associated with the construction and operation of an RO Plant;

• Identify the environmentally preferred site and intake and discharge positions from the

alternatives identified through the Basic Assessment process.

The detailed terms of reference for this study are contained in Section 1 of this report.

Roy van Ballegooyen

Coastal Systems Research Group Stellenbosch, South Africa

Natural Resources and the Environment CSIR February 2008


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page ii



EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page iii

EXECUTIVE SUMMARY

Introduction The Iron Ore Handling Facility at the Port of Saldanha (including the proposed upgrades to the

facility), will require increased water supplies for dust suppression measures. It is anticipated that

ultimately a total of 3600 kℓ of water per day will be required for this purpose. The municipality

currently supplies water to the Iron Ore Handling Facility, but Transnet propose to supplement or

possibly offset this supply with additional sources of freshwater. Of all potential sources of such

water, seemingly the most feasible is that of a Reverse Osmosis (RO) plant, for the desalination of

seawater.

In terms of existing legislation this proposed activity (i.e. the construction and operation of an RO

Plant) will require that a Basic Assessment of potential environmental impacts be undertaken.

Consequently, the PDNA/SRK Joint Venture approached the CSIR and Pisces Environmental

Services (Pty) Ltd to undertake a Marine Specialist Study to assess the impacts of the construction

and operation of the RO Plant on the marine environment. In summary, the CSIR was asked to:

• Advise on the design of the RO Plant early in the process and identify any fatal flaws

associated with any of the alternative sites, or the proposed project;

• Assess the significance of the potential impact of the proposed development on the marine

environment for each of the alternative sites (taking into consideration the differences in project

design in the case of each alternative site);

• Recommend mitigation measures to minimise impacts associated with the proposed RO Plant

and enhance benefits; and

• Identify an environmentally preferred site.

Project Description Three alternative locations within the Port of Saldanha are being considered for the location of the

RO plant (see figure below). At the three proposed sites, there are various opportunities and

constraints in terms of the specific intake and discharge alternatives that may be considered (see

Table 3.1 in the main body of this report). These are described in generic terms, followed by a

detailed description of the various alternatives assumed and assessed for the three proposed

locations for the RO Plant.

The development alternatives at these three sites include a number of combinations of intake and

discharge infrastructure relevant to each of these sites as summarised in the Table below. Intake

structures considered are beach wells, pipeline intakes or borehole intakes. Discharge structures

considered are various pipeline discharge configurations comprising primarily single point or multi-

point diffuser structures.


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page iv

Proposed locations for the RO Plant.

Development alternatives used in this assessment.

Site Alternative # Alternative Intake/Discharge Infrastructure Locations

Site 1 1 a) Beach well intake and pipeline discharge (Big Bay)

b) Pipeline intake and pipeline discharge (Big Bay)

Site 2 a) Beach well intake and pipeline discharge (Small Bay)

b) Pipeline intake and pipeline discharge (Small Bay)

Site 3

a) Pipeline intake (Small Bay) and pipeline discharge (Small Bay)

b) Pipeline intake (Small Bay) and pipeline discharge (Big Bay)

c) Borehole intake on quay (adjacent to existing iron-ore stockpiles) and pipeline discharge (Caisson 3, Big Bay)

d) Borehole intake on quay (adjacent to Multi-purpose Terminal) and pipeline discharge (Caisson 3, Big Bay)

Greater detail of these combinations of intake and discharge infrastructure are contained in Figures

3.3 to 3.6 in the main body of the report. 1 A beach well intake and beach well discharge option was also considered for Site 1, however the

groundwater specialist study indicated the beach sediment conditions were such that that a beach well discharge is considered not feasible

Small Bay

Big Bay


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page v

Key Issues and sources of Impact The key issues identified in the course of the public participation phase of the Basic Assessment (also referred to as EIA in this document) and the environmental screening process for the proposed RO Plant relate to the construction phase, operational phase and decommissioning phase. Key issues in the construction phase include the disturbance of the beach in the case of intake beach wells and the seabed (and possibly associated sediment dynamics) in the case of pipeline intakes and discharges. Most of the key issues and major potential impacts are associated with the operational phase. The key issues related to the presence of pipeline infrastructure and brine discharges into the marine environment are:

• altered flows at the intake and discharge resulting in ecological impacts (e.g. entrainment of biota at the intake, low distortion/changes at the discharge, and affects on natural sediment dynamics);

• the effect of elevated salinities in the brine water discharged to the bay;

• biocidal action of residual chlorine and/or other non-oxidising biocides such as dibromonitrilopropionamide (DBNPA)) in the effluent;

• the effects of co-discharged waste water constituents; including possible tainting effects affecting both mariculture activities and fish factory processing in the bay;

• the effect of the discharged effluent having a higher temperature than the receiving environment; and

• changes in dissolved oxygen that include:

• direct changes in dissolved oxygen content due to the difference between the ambient dissolved oxygen concentrations and those in the discharged effluent, and;

• indirect changes in dissolved oxygen content of the water column and sediments due to:

i) changes in phytoplankton production as a result of changes in nutrient dynamics (both in terms of changes in nutrient inflows and vertical mixing of nutrients), and;

ii) changes in remineralisation rates (with related changes in nutrient concentrations in near bottom waters) due to near bottom changes in seawater temperature associated with the brine discharge plume.

Additional engineering design considerations, not strictly constituting issues to be considered within the Basic Assessment of potential environmental impacts, include the following:

• structural integrity of the intake and discharge pipelines (e.g. related to shoreline movement);

• potential re-circulation of brine effluents if intakes and discharges are situated in close proximity to one another. The model results however indicate that this will not be a concern for all of the intake and discharge options considered in this study;


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page vi

• the permeability and particle size distributions of the sands (should beach intake and/or discharge wells be considered as a viable option); and

• water quality of feed-waters that may require specific mitigation measures or planned flexibility in the operations of the RO Plant.

The potential impacts during the de-commissioning phase are expected to be minimal in comparison to those occurring during the operational phase, and no key issues related to the marine environment are identified at this stage. The minimum anticipated life of the RO plant is 25 years. The individual RO modules will be replaced as and when required during this period. No decommissioning procedures or restoration plans have been compiled at this stage, as it is envisaged that the plant will be refurbished rather than decommissioned after the anticipated 25 years, as the beach wells and boreholes (if used) are expected to have a considerably longer lifespan than the RO modules. A full decommissioning will require a separate EIA at the time of decommissioning.

Changes in Environment due to Plant Discharges Model simulations were undertaken of the various options. The results for changes in salinity, seawater temperature and the footprint of biocide and potential co-discharges for all of the site alternatives are summarised in the figures below.

Available mariculture

concessions

Available/existing mariculture concessions

Existing mariculture concession

s


concessions



s


concessions



s


concessions



s


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page vii

Assessment of Environmental Impact The potential impacts associated with construction activities, the operation of the plant (brine and co-

discharges) and the long-term impacts associated with the intake and discharge structures are

assessed below.

Impacts During Construction

In the absence of detailed engineering specifications describing the construction of the intake and

discharge structures (both pipeline or beach wells), the assessment is based on generic assumptions

for such constructions. Should the final construction specifications be sufficiently different to those

assumed in this report, the impacts associated with the construction phase will need to be re-

evaluated. For the intake structure along the quayside for Site 3 and the proposed discharge

structure at Caisson 3, the impacts will be minimal as these construction activities will be utilising

existing infrastructure as their basis and construction activities will not be extensive.

Summary of construction impacts for the various Sites and associated intake and discharge combinations.

Site 1

Site 2

Site 3 (Small Bay)

Site 3 (Big Bay)

Site 3 (Caisson 3)

Construction Impacts

Construction of Beach Well intakes

no mitigation Medium Medium n/a n/a n/a

with mitigation No significant mitigation possible other than avoiding beach well construction

n/a n/a n/a

Construction of borehole intakes

no mitigation n/a n/a n/a n/a Very low

with mitigation n/a n/a n/a n/a No mitigation

required

Construction of Intake pipelines

no mitigation Medium Medium Very low Very low n/a

with mitigation

Limited mitigation is possible using “best practice” mitigation measures during

construction. It is not possible to propose specific mitigation measures

based on existing known detail of construction activities

None deemed necessary other than normal environmental ‘best practice’

n/a

Construction of discharge infrastructure: Comprises a pipeline with diffuser for Site 2 and for discharges into Small Bay and Big Bay from Site 3 (i.e. development options 2a, 2b, 3a and 3b), and a pipeline with a more complicated discharge infrastructure than simply a diffuser at Site 1 and for a discharge at Caissoon 3 from Site 3 (i.e. development options 1a, 1b, 3c and 3d) no mitigation Very low Medium Medium Medium Very low

with mitigation

None deemed necessary other

than the use of “best practice” mitigation measures during

construction.

Limited mitigation is possible using “best practice” mitigation measures during construction. It is not possible to propose specific mitigation measures based on existing known detail

of construction activities

None deemed necessary other

than the use of “best practice” mitigation measures during

construction.


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page viii

Impacts during Operation of the RO Plant

The impacts on the marine environment associated with the operation of the RO Plant include those

associated with entrainment of biota, flow distortion and changes in sediment dynamics and of

greatest concern potential impacts on water quality in the marine environment associated with

discharges from the RO Plant. These are tabulated below.

Assessment Criteria

Site 1

Site 2

Site 3 (Small Bay)

Site 3 (Big Bay)

Site 3 (Caisson 3)

Operational Impacts (other than Water Quality) of Intake and discharge operation

Entrainment of Biota (Pipeline Intakes) no mitigation Medium Medium Medium Medium n/a

with mitigation Low Low Low Low n/a

Entrainment of Biota (Beach Well Intakes) no mitigation insignificant insignificant n/a n/a n/a

with mitigation No mitigation deemed necessary n/a n/a n/a

Entrainment of Biota (Borehole intakes along Causeway) no mitigation n/a n/a Insignificant Insignificant Insignificant

with mitigation n/a n/a No mitigation deemed necessary

Flow Distortion

no mitigation Low Low Low Low Low

with mitigation Insufficient detail in project description to define specific mitigation measures, however appropriate engineering design should negate any potential impacts

Impacts on Sediment Dynamics

no mitigation insignificant Low Low Low insignificant

with mitigation No mitigation required

No detailed mitigation can be recommended as insufficient detail in project description, however appropriate

engineering design should negate any potential impacts

No mitigation required

Specific Impacts associated with the discharged brine (and associated co-discharged)

Salinity no mitigation Low Low Low Low Low

with mitigation No mitigation specified/considered, other than optimal discharge diffuser design

Temperature

no mitigation Low Low/Medium* Low Low Low


Oxygen (no O2 scavengers)


with mitigation Mitigation not deemed necessary unless oxygen scavenging compounds are added to the feed-water. This will only occur should chlorine be used as a biocide that results in a potential discharge of

residual chlorine into the marine environment.. Presently the project description specifies that chorine will not be used in this role

Oxygen (with O2 scavengers) no mitigation Medium Medium Medium Medium Medium

with mitigation Very Low Very Low Very Low Very Low Very Low

Oxidising Biocides (NaOCl): Beach well or borehole intakes

no mitigation Low Low n/a n/a Low

with mitigation If biocide dosage is as specified, no

mitigation is likely to be required, other than optimal discharge diffuser design

n/a n/a

No mitigation likely to be

required, other than optimal

discharge diffuser design


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page ix

Assessment Criteria

Site 1

Site 2

Site 3 (Small Bay)

Site 3 (Big Bay)

Site 3 (Caisson 3)

Oxidising Biocides (NaOCl): Pipeline intakes

no mitigation Medium Medium Low Low n/a

with mitigation** Very Low Very Low Very Low Very Low n/a

Non-oxidising Biocides (DBNPA): Beach well or borehole intakes


with mitigation*** If biocide dosage is as specified, no

mitigation is likely to be required, other than optimal discharge diffuser design

n/a n/a

No mitigation likely to be

required, other than optimal


Non-oxidising Biocides (DBNPA): Pipeline intakes

no mitigation Medium Medium Low Low Low

with mitigation*** Low Low Very Low Very Low Very Low

Co-discharges (beach well or borehole intakes) no mitigation Low Low n/a n/a Low

with mitigation****

Very Low Very Low n/a n/a Very Low

Co-discharges (pipeline intakes) no mitigation Low Low//Medium Low Low n/a

with mitigation****

Very Low Very Low Very Low Very Low n/a

* Note that the intensity of the thermal impacts at Site 2 may be reduced and assumed to be low, i) given appropriate engineering design mitigation (i.e. assuming that it would be possible to locate the pipeline intake in a manner that would ensure sufficiently lower seawater temperatures at the intake) or ii) if it is assumed that beach well intake water temperatures are lower that the near surface water temperatures assumed in the modelling.

** A proposed mitigation for oxidising biocides such an sodium hypochlorite (if required and to be applied to the extent necessary) is de-chlorination by the use of an oxygen scavenger such as Sodium metabisulphate, however as this does potentially exacerbate impacts associated with low concentrations of dissolved oxygen in the marine environment, the mitigation preferred is that of carefully monitored and managed dosing to ensure minimal residual chlorine concentrations in the discharge.

*** The proposed mitigation to reduce DBNPA impacts in the marine environment is to design the brine basin so as to ensure greater and sufficient dilution of the DBNPA residuals in the effluent stream before discharge or to revert to the use of an oxidising biocide (chlorine) in this role.

**** Mitigation in unlikely to be required except for perhaps Site 2, should monitoring studies indicate the accumulation of sediments to the extent that it becomes a concern. The assessment with mitigation assumes mitigation measures whereby particulates are removed from the flocculant/backwash sludge to the extent required and disposed elsewhere than the marine environment.

“No-development” Alternative

The “no-development” alternative implies that the RO Plant and associated infra-structure will not be

commissioned. From a marine perspective this is undeniably the preferred alternative, as all

identified non-negligible impacts associated with beach disturbance and effluent discharge will not

occur and therefore will no longer be of concern and/or require mitigation. This must, however, be

seen in context with the existing need for additional water to implement the required dust mitigation

measures, the proposed extensions to the Port, as well as the use of possible alternative water

sources for dust control. Furthermore, it needs to be weighed up against the potential positive socio-

economic impacts undoubtedly associated both with the RO Plant project itself, as well as the Port

extension.


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page x

Project Impacts and Environment Interaction Points

The figures providing the maximum footprint in terms of salinity, seawater temperature, biocide and

potential co-discharge impacts show that the plume footprints do not extend as far as any existing or

proposed mariculture activities, seawater intakes for fish processing factories, recreational and

commercial gill-netting areas, or National Parks and Marine Protected Areas. At Site 2, however,

the plumes extend close to the eastern boundary of the area demarcated for seaweed

harvesting.

Conclusions

Each Site and its associated intake and discharge combinations has its own unique set of potential

impacts. However if these sites were to be ranked qualitatively in terms of their potential impacts,

their ranking would be as tabulated below. A ranking of 1 indicates the most preferable option in

terms of minimising overall impacts on the marine environment.

Qualitative Ranking of Options

Overall Ranking Site 1

Site 2

Site 3

(Small Bay) Site 3

(Big Bay) Site 3

(Caisson 3) 4 5 3 2 1

All Sites (except perhaps for Site 2) remain viable options. Site 2, whilst not environmentally fatally

flawed in terms of the existing proposed project, is not recommended as it does have a significantly

larger impact area than the other sites. It also constitutes a discharge into a relatively quiescent and

poorly flushed region of the bay where there are existing water quality concerns and neither the

waves nor currents are sufficient to efficiently disperse the effluent plume and potential backwash

sediments. Accordingly the public perceptions around potential impacts a RO plant discharge in such

a location would be the most negative. Furthermore, Site 2 is expected to have significant limitations

mainly in terms of the limited assimilative capacity of Small Bay when dealing with possible future

increases in discharges to the marine environment e.g. possible discharges from other

industrial/processing activities or the expansion of RO plant operations (although, it should be noted

that no more than the 3 RO plant modules are being proposed by Transnet).

Recommendations

The recommendations from this study include mitigation measures (optional and required) and

monitoring requirements to be able to better assess and manage potential impacts.

Mitigation Measures

The recommended mitigation measures are listed below for both the construction and operational

phases of the RO Plant. These are listed in the following table with a clear indication of whether they

are required or merely recommended.


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page xi

Recommended and Essential Mitigation Measures

Mitigation Necessity

Construction Impacts: - Limiting and restricting vehicular traffic on the beach - Good house-keeping - Active rehabilitation above high water mark

Required Required Required

Use of sub-surface intakes (beach wells/boreholes) Required (highly recommended)

Pipeline intakes: - Adjustment of intake velocities and/or velocity caps - Use of intake screens - Operational cleaning (pigging) of intake pipes

Required Required Optional

Discharges: - Evaporation ponds / crystallisation - Beach well disposal - Carefully controlled dosing of biocides based on feedback from

monitoring systems - Dechlorination of purge water - Reduction of residual DBNPA concentration in effluent to be

discharged by designing the brine basin so as to ensure greater and sufficient dilution of the DBNPA residuals in the effluent stream before discharge, or to revert to the use of an oxidising biocide (chlorine) in this role.

- Aeration of purge water - Removal of sludge particles from backwashing of RO modules - Optimal diffuser design based on acceptable water quality

target values

Not feasible Not feasible

Required

Required only if chlorine is used in systems discharging to the marine

environment (presently not indicated) and then only in the case of pipeline intakes as chlorine dosing for beach

well and borehole intakes is assumed to be low

Required only in the case of pipeline intakes as DBNPA dosing for beach

well and borehole intakes is assumed to be low

Optional but highly recommended if chlorine is used in systems discharging to the marine environment (presently

not indicated) and sodium metabisulfite is used to neutralise chlorine residuals in the effluent stream discharged to the

marine environment

Optional, to be undertaken to the extent required (i.e. as informed by the monitoring results for the discharge.)

Required


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page xii

Monitoring Recommendations

The monitoring recommendations are as follows:

Monitoring Activity Nature and Duration Establishment of a baseline of intertidal and shallow subtidal invertebrate macrofaunal communities before any construction commences (Site 1 only)

Once-off survey Required

(Site 1 only)

Development of a monitoring programme to study the impact of the brine on potentially affected communities, particularly the subtidal benthic communities.

Surveys at recommended 6-monthly intervals (DWAF, 2007) or annually (whichever deemed more appropriate) over a period of approximately 4 years.*

Required

Monitoring of the RO plant effluent after commissioning for heavy metals.

Depending on plant design and associated perceived risks, monitoring over a range of operational conditions until a profile of the discharge in terms of heavy metal concentrations is determined.

Required

Toxicity testing of the RO Plant effluent at discharge point for a full range of operational scenarios (i.e. shock-dosing, etc).

Monitoring to continue until a representative profile of the discharge in terms of particularly biocide concentrations is determined.

Required

Monitoring (or audit of process chemical once RO Plant is commissioned) to ensure that tainting substances are absent from the RO Plant effluent.

Depending on plant design and associated perceived risks, monitoring over a range of operational conditions until it is confirmed that no tainting substances are present. However, it may be sufficient to simply audit the process chemicals used to confirm the absence of tainting substances.

Required

Monitoring to confirm performance of the discharge system and the numerical model predictions

Monitoring of salinity, temperature and total suspended solids in the near-field.

Required

Monitoring to confirm the numerical model predictions

As detailed above for confirmation of discharge system performance, but over a slightly more expansive area.

Optional

* Recommended intervals could be reduced to annual surveys if deemed appropriate and/or indicated by preliminary survey results from initial surveys.


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page xiii

TABLE OF CONTENTS

SCOPE OF WORK ............................................................................................................................................... I EXECUTIVE SUMMARY ................................................................................................................................ III TABLE OF CONTENTS................................................................................................................................ XIII LIST OF FIGURES........................................................................................................................................... XV LIST OF TABLES.......................................................................................................................................... XVII ABBREVIATIONS, UNITS AND GLOSSARY............................................................................................XIX

1 INTRODUCTION.......................................................................................................................................... 1

1.1 BACKGROUND ........................................................................................................................................ 1 1.2 SCOPE OF WORK..................................................................................................................................... 2 1.3 STRUCTURE OF THE REPORT ................................................................................................................... 2 1.4 LEGISLATIVE AND PERMITTING REQUIREMENTS .................................................................................... 4

2 PROJECT DESCRIPTION ........................................................................................................................ 11

2.1 DESCRIPTION OF THE RO PLANT FACILITIES ........................................................................................ 11 2.2 LIFE SPAN OF PROJECT ......................................................................................................................... 12 2.3 PROPOSED PLANT CAPACITY ................................................................................................................ 12 2.4 CONSTRUCTION AND COMMISSIONING OF INFRASTRUCTURE ................................................................ 13

3 IDENTIFICATION AND SELECTION OF ALTERNATIVES............................................................. 17

3.1 GEOGRAPHICAL SETTING...................................................................................................................... 17 3.2 SITING ALTERNATIVES ......................................................................................................................... 18 3.3 CONCEPTUAL DESIGNS OF POSSIBLE INTAKE AND DISCHARGE STRUCTURES....................................... 20

3.3.1 Intake Alternatives........................................................................................................................... 20 3.3.2 Discharge Alternatives .................................................................................................................... 24

3.4 “NO-DEVELOPMENT” ALTERNATIVE .................................................................................................... 27 3.5 SUMMARY OF DEVELOPMENT ALTERNATIVES...................................................................................... 27

4 APPROACH TO THE STUDY .................................................................................................................. 29

4.1 METHODOLOGY .................................................................................................................................... 30 4.1.1 Environmental Baseline................................................................................................................... 30 4.1.2 Modelling......................................................................................................................................... 30 4.1.3 Environmental Impact Assessment .................................................................................................. 32

4.2 LIMITATIONS AND ASSUMPTIONS ......................................................................................................... 35 5 DESCRIPTION OF THE AFFECTED ENVIRONMENT...................................................................... 39

5.1 PHYSICAL ENVIRONMENT..................................................................................................................... 39 5.1.1 General ............................................................................................................................................ 39 5.1.2 Climate and Winds........................................................................................................................... 39 5.1.3 Tides ................................................................................................................................................ 41 5.1.4 Waves............................................................................................................................................... 42 5.1.5 Currents........................................................................................................................................... 47 5.1.6 Water Column Stratification............................................................................................................ 47 5.1.7 Seawater Temperature..................................................................................................................... 50 5.1.8 Salinity............................................................................................................................................. 52 5.1.9 Water Quality .................................................................................................................................. 52


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page xiv

5.2 BIOLOGICAL ENVIRONMENT................................................................................................................. 54 5.2.1 Sandy Substrate Habitats and Biota ................................................................................................ 54 5.2.2 Rocky Habitats and Biota ................................................................................................................ 59 5.2.3 Pelagic Communities ....................................................................................................................... 60 5.2.4 Birds ................................................................................................................................................ 62 5.2.5 Beneficial Uses ................................................................................................................................ 63 5.2.6 Existing Environmental Impacts...................................................................................................... 68 5.2.7 Potentially Threatened Habitats and Beneficial Uses ..................................................................... 72

6 IDENTIFICATION OF KEY ISSUES AND SOURCES OF POTENTIAL ENVIRONMENTAL

IMPACT ....................................................................................................................................................... 73 6.1 CONSTRUCTION PHASE ......................................................................................................................... 73 6.2 OPERATIONAL PHASE ........................................................................................................................... 74 6.3 DECOMMISSIONING PHASE ................................................................................................................... 75

7 MODELLED CHANGES IN THE MARINE ENVIRONMENT DUE TO THE RO PLANT

EFFLUENT DISCHARGES TO THE MARINE ENVIRONMENT...................................................... 77 7.1 INTRODUCTION ..................................................................................................................................... 77 7.2 ASSUMED INTAKE AND DISCHARGE LOCATIONS.................................................................................... 77 7.3 INTAKE CHARACTERISTICS ................................................................................................................... 85 7.4 DISCHARGE CHARACTERISTICS ............................................................................................................ 86 7.5 SCENARIOS SIMULATED ........................................................................................................................ 91 7.6 MODEL RESULTS .................................................................................................................................. 93

7.6.1 Analysis of results............................................................................................................................ 93 7.6.2 Summary of results in terms of exceedance of selected water quality guidelines............................ 96

8 ASSESSMENT OF ENVIRONMENTAL IMPACTS ............................................................................ 107

8.1 IDENTIFICATION OF POTENTIAL ENVIRONMENTAL IMPACTS............................................................... 107 8.1.1 Construction of Intake and Discharge Structures.......................................................................... 107 8.1.2 Permanent Intake and Discharge Structures................................................................................. 109 8.1.3 RO Plant Effluents ......................................................................................................................... 112

8.2 ASSESSMENT OF POTENTIAL ENVIRONMENTAL IMPACTS ................................................................... 143 8.2.1 Assessment of Impacts During Construction ................................................................................. 143 8.2.2 Assessment of Impacts Associated with Brine Discharge.............................................................. 145 8.2.3 Assessment of Impacts Associated with Intake and Discharge Structures..................................... 154

8.3 “NO-DEVELOPMENT” ALTERNATIVE .................................................................................................. 155 8.4 PROJECT IMPACTS AND ENVIRONMENT INTERACTION POINTS............................................................ 155

9 CONCLUSIONS AND RECOMMENDATIONS ................................................................................... 157

9.1 ENVIRONMENTAL ACCEPTABILITY AND COMPARISON OF ALTERNATIVES ......................................... 157 9.2 PREFERRED ALTERNATIVE.................................................................................................................. 159 9.3 RECOMMENDATIONS........................................................................................................................... 165

9.3.1 Mitigation Measures...................................................................................................................... 165 9.3.2 Monitoring Recommendations....................................................................................................... 169

10 REFERENCES........................................................................................................................................... 171 Appendix A: Possible implications of policy, legislation and approval and licensing procedures for

the proposed RO Plant water discharge

Appendix B: Hydrodynamic and Water Quality Model Set-up and Calibration Appendix C: Model Results


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page xv

LIST OF FIGURES

Figure 2.2a: Proposed layouts for the Phase 2A development of Iron-ore export facilities in Saldanha Bay. 15

Figure 2.2b: Proposed layouts for the Phase 2B development of Iron-ore export facilities in Saldanha Bay. 15

Figure 3.1: The Saldanha Bay/Langebaan Lagoon system showing major port infrastructure. 17

Figure 3.2: Proposed locations for the RO Plant. 19

Figure 3.3: Location and extent of intake beach wells at Site 1 (PDNA/SRK Joint Venture, 2008). 21

Figure 3.4: Location and extent of intake beach wells at Site 2 (PDNA/SRK Joint Venture, 2008). 22

Figure 3.5: Location and number of intake boreholes alongside the existing iron-ore stockpiles (PDNA/SRK Joint Venture, 2008). 25

Figure 3.6: Location and number of intake boreholes alongside the existing MPT (PDNA/SRK Joint Venture, 2008). 26

Figure 5.1: Wind roses of the winds measured at Port Control in Saldanha Bay (see inset). 40

Figure 5.2: Wave height measured at the Slangkop Directional Waverider, located off Kommetjie on the Cape South-west coast (34°12'14.40"S, 18°17'12.01"E, 70 m depth) for the period 2001 to 2007. 43

Figure 5.3a: Simulated wave conditions in Saldanha Bay for large offshore swell conditions and low winds (Offshore Hmo = 3.6 m , Tp=13 s, Direction = SW’ly, Local wind = 3 m/s S’ly). 44

Figure 5.3b: Simulated wave conditions in Saldanha Bay for moderate offshore swell conditions and very strong winds (Offshore Hmo = 1.3 m , Tp=11 s, Direction = SW’ly, Local wind = 20 m/s S’ly). 44

Figure 5.4a: Flood tide surface and bottom currents in Saldanha Bay during spring tide and under relatively calm conditions. 45

Figure 5.4b: Ebb tide surface and bottom currents in Saldanha Bay during spring tide and under relatively calm conditions. 45

Figure 5.5a: Schematic of the wind driven and tidal currents in Saldanha Bay under S wind conditions. 46

Figure 5.5b: Schematic of the wind driven and tidal currents in Saldanha Bay under NW wind conditions. 46

Figure 5.6a: Surface and bottom water temperature and flow in Saldanha Bay under strong SE wind conditions (i.e. during the “active” upwelling phase). 48

Figure 5.6b: Surface and bottom water temperature and flow in Saldanha Bay after a strong SE wind event (i.e. relaxation phase of upwelling). 48


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page xvi

Figure 5.7: Vertical temperature structure of the water column along the “profile line” or cross-section indicated in Figure 5.6b. The upper panel shows the “active” upwelling phase while the lower panel shows the water column structure during the relaxation phase of the upwelling cycle. 49

Figure 5.8: Simulated surface and bottom water temperatures at North Buoy in Small Bay (see inset), showing the various temporal scales and magnitudes of seawater temperature variability in Saldanha Bay. 50

Figure 5.9: Conservation areas in Saldanha Bay (adapted from Taljaard and Monteiro, 2002). 64

Figure 5.10: Current mariculture lease holders in Saldanha Bay. 66

Figure 5.11: Designated beneficial use areas in Saldanha Bay (adapted from Taljaard and Monteiro, 2002). 68

Figure 7.1a: Proposed location of intake and discharge structures for Site 1. 78

Figure 7.1b: Proposed location of intake and discharge structures for Site 2. The yellow dotted line encloses the potential area within which beach wells may be located. 78

Figure 7.1c: Proposed location of intake and discharge structures for pipeline discharges into Small and Big Bay for Site 3. 79

Figure 7.1d: Proposed location of borehole intakes (adjacent to stockyard and MPT) discharge location for a pipeline discharge from Site 3 at Caisson 3. The intake boreholes will be located within the linear distance marked by the white lines in the figure opposite the existing iron-ore stockyard and the MPT (see Figures 3.5 and 3.6 in Section 3 for more detail). 79

Figure 7.2: Schematic of the near-field behaviour of a dense effluent. 91

Figure 7.3: Comparative maximum dimensions of the elevated salinity “footprint” (ΔS < 1 psu or S < 36 psu) for all discharge sites. 98

Figure 7.4: Comparative maximum dimensions of the elevated temperature “footprint” (ΔT < 1 ºC) for all discharge sites. 100

Figure 7.5: Comparative maximum dimensions of the biocide “footprint” (oxidising biocide concentration <3 µg.ℓ-1 or DBNPA residual concentrations less than the target values assumed to be appropriate) for all discharge sites. 102

Figure 7.6: Comparative maximum dimensions of the plume “footprint” (achievable dilution < 50) for all discharge sites. 105


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page xvii

LIST OF TABLES

Table 2.1: Proposed RO plant capacity for the various development phases of the Saldanha Bay Iron-ore Export Expansion project. 12

Table 3.1: Development alternatives used in this assessment (confirmed by Transnet on 22 November 2007). 28

Table 5.1: Tidal characteristics for Saldanha Bay. 41

Table 5.2: Wave height exceedances measured at a wave buoy near the entrance to Small Bay.42

Table 5.3: Dominant species/taxa in Saldanha Bay as reported from benthic macrofauna studies conducted in 1975, 1999 and 2004 (adapted from Atkinson et al., 2006). 56

Table 5.4: Current Mariculture lease holders in Saldanha Bay. 65

Table 7.1: Location of the intake and discharges for the various proposed sites. 81

Table 7.2: Main characteristics and conceptual design of intake structures for the RO plant. 82

Table 7.3: Main characteristics and conceptual design of discharge structures for the RO plant. 87

Table 7.4: Discharge rates and salinity of the brine discharge for the various development phases and plant capacities. 89

Table 7.5: The five intake and discharge combinations simulated in the hydrodynamic and water quality modelling. 92

Table 7.6: Summary of effluent plume dimensions around the discharge point (based on exceedances of the salinity water quality guidelines for cumulative periods of 6 hours and approximately 5 days), and the magnitude of the salinity elevation at the intake. 97

Table 7.7: Summary of effluent plume dimensions around the discharge point (based on exceedances of the temperature water quality guidelines for cumulative periods of 6 hours and approximately 5 days), and the magnitude of the temperature elevation at the intake. 99

Table 7.8: Summary of effluent plume dimensions around the discharge point (based on exceedances of the biocide water quality guidelines) for cumulative periods of 6 hours and approximately 5 days, respectively. 101

Table 7.9: Summary of effluent plume dimensions around the discharge point (based on non-exceedances of a required dilution of 50 times) for cumulative periods of less than 6 hours and less than approximately 5 days, respectively. 103

Table 7.10: Summary of effluent plume dimensions around the discharge point (based on non-exceedances of a required dilution of 100 times) for cumulative periods of less than 6 hours and less than approximately 5 days, respectively. 104

Table 8.2: Likely profile of residual concentrations of DBNPA in discharges to the marine environment from the RO Plant. (after Klaine et al., 1996) 128

Table 8.3: Potential chemicals used for the Reverse Osmosis process - information as supplied by Transnet on 5 December 2007. Quantities based on intakes flows of 8000 m3.day-

1. 135


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page xviii

Table 9.1: Summary of potential impacts for the various Sites and associated intake and discharge combinations. 161

Table 9.2: Optional and required mitigation measures 168


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page xix

ABBREVIATIONS, UNITS AND GLOSSARY

Abbreviations

ANZECC Australian and New Zealand Environment and Conservation Council

CD Chart Datum

CIP Clean in Place

CSIR Council for Scientific and Industrial Research

CTD Conductivity-Temperature-Depth probe

DBNPA dibromonitrilopropionamide, a non-oxidising biocide

DMF Dual Media Filters

DSP Diarrehetic Shellfish Poisoning

DWAF Department of Water Affairs and Forestry

E East

EC50 median effective concentration

EDTA Ethylenediaminetetraacetic acid

EIA Environmental Impact Assessment

ESE East-southeast

HAB Harmful Algal Blooms

LC50 median lethal concentration

MCM Marine and Coastal Management

MPA Marine Protected Area

MPT Multi-purpose terminal

NaOCl Sodium Hypochlorite

NOEC no observed effect concentration

NNW North-northwest

TNPA Transnet National Ports Authority

NWA National Water Act (1998)

NSF National Sanitation Foundation

NW Northwest

PDNA PD Naidoo & Associates (Pty) Ltd


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page xx

PNEC predicted no effect concentrations

PSP Paralytic Shellfish Poisoning

RO Reverse Osmosis

RSA Republic of South Africa

RSA DWAF Republic of South Africa, Department of Water Affairs and Forestry

SBWQFT Saldanha Bay Water Quality Forum Trust

SE Southeast

SLS Sodium lauryl sulphate

SSE South-southeast

SSW South-southwest

STPP Sodium tripolyphosphate

SWRO Seawater Reverse Osmosis

TOC Total Organic Carbon

TRC Total Residual Chlorine

TSP Trisodium phosphate

WSW West-southwest

Units used in the report

cm centimetres

g.kg-1 grams per kilogram

g C.m-2 .day-1 grams Carbon per square metre per day

gfd gallons per square foot per day

h hours

ha hectares

kg kilogram

km kilometres

km2 square kilometres

µg.ℓ-1 micrograms per litre

Mℓ Megalitres

m metres


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page xxi

mm millimetres

m2 square metres

m3.day-1 cubic metres per day

m3.hr-1 cubic metres per hour

m3.s-1 cubic metres per second

m3 yr-1 cubic metres per year

m.s-1 metres per second

mg.ℓ-1 milligrams per litre

ng.ℓ-1 nanograms per litre

mg Chl a.m-3 milligrams Chlorophyll a per cubic metre

ppt parts per thousand

psu practical salinity units which in the normal oceanic salinity ranges are the

same as parts per thousand (ppt)

% percentage

~ approximately

< less than

≤ less than or equal to

> greater than

≥ greater than or equal to

°C degrees centigrade


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page xxii

Glossary

Acute toxicity Rapid adverse effect (e.g. death) caused by a substance in a

living organism. Can be used to define either the exposure or

the response to an exposure (effect).

Anoxia: The absence or near absence of oxygen, i.e. < 0.1 ml O2/ℓ.

Baroclinic A condition in the ocean in which isobaric (pressure) and

constant-density surfaces are not parallel, typically resulting in

vertically sheared flows driven by these density differences in

the water column.

Benthic Referring to organisms living in or on the sediments of aquatic

habitats (lakes, rivers, ponds, etc.).

Benthos The sum total of organisms living in, or on, the sediments of

aquatic habitats.

Benthic organisms Organisms living in or on sediments of aquatic habitats

Biodiversity The variety of life forms, including the plants, animals and

micro-organisms, the genes they contain and the ecosystems

and ecological processes of which they are a part.

Biocide A substance, such as a chlorine, that is capable of destroying

living organisms if applied in sufficient doses.

Biomass The living weight of a plant or animal population, usually

expressed on a unit area basis.

Biota The sum total of the living organisms of any designated area.

Bioturbation The displacement and mixing of sediment particles by benthic

fauna (animals) or flora (plants).

Bivalve A mollusk with a hinged double shell.

Community structure All the types of taxa present in a community and their relative

abundance.

Community An assemblage of organisms characterized by a distinctive

combination of species occupying a common environment and

interacting with one another.

Contaminant Biological (e.g. bacterial and viral pathogens) and chemical

introductions capable of producing an adverse response

(effect) in a biological system, seriously injuring structure or

function or producing death.


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page xxiii

Detritus Unconsolidated sediments composed of both inorganic and

dead and decaying organic material.

Dilution The reduction in concentration of a substance due to mixing

with water

Dissolved oxygen (DO) Oxygen dissolved in a liquid, the solubility depending upon

temperature, partial pressure and salinity, expressed in

milligrams/litre or millilitres/litre

Diurnal Having a 24 hour cycle or recurring on a 24 hour basis.

Drfit-line A line of debris left by waves at the high-tide line.

Ecosystem A community of plants, animals and organisms interacting with

each other and with the non-living (physical and chemical)

components of their environment

Effluent A complex waste material (e.g. liquid industrial discharge or

sewage) that may be discharged into the environment.

Environmental impact A positive or negative environmental change (biophysical,

social and/or economic) caused by human action

Environmental quality A statement of the quality requirement for a body of water to

objective be suitable for a particular use (also referred to as

Resource Quality Objective)

Epifauna Organisms, which live at or on the sediment surface being

either attached (sessile) or capable of movement.

Eutrophication Strictly speaking, means an increase in chemical nutrients

(typically compounds containing nitrogen or phosphorus) in an

ecosystem. However this term is often used to mean the

resultant increase in the ecosystem's primary productivity (i.e.

excessive plant growth and decay) and even further impacts,

including lack of oxygen and severe reductions in water quality.

Far-field The region beyond the near-field where secondary dilution

effects such as dispersion by environmental flows, etc

predominate rather than initial dilution effects due to buoyancy

forcing and entrainment.

Guideline trigger values These are the concentrations (or loads) of the key performance

indicators measured for the ecosystem, below which there

exists a low risk that adverse biological (ecological) effects will

occur. They indicate a risk of impact if exceeded and should

‘trigger’ some action, either further ecosystem specific


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page xxiv

investigations or implementation of management/remedial

actions.

Habitat The place where a population (e.g. animal, plant, micro-

organism) lives and its surroundings, both living and non-living.

Headloss The head or pressure lost by water flowing in a pipe or

channel, caused by the roughness of the pipe or channel walls.

Infauna Animals of any size living within the sediment. They move

freely through interstitial spaces between sedimentary particles

or they build burrows or tubes.

Initial dilution Dilution that occurs in the near-field whilst the effluent plume is

still under the influence of strong buoyancy forcing and

entrainment effects.

LC50 (Median Lethal Concentration) A statistically derived concentration of a substance that can be

expected to cause death in 50% of test animals.

Level of Concern (LOC) A level (concentration) of a pollutant or toxicant at above which

the toxicant will pose a threat to an ecosystem or component

thereof.

Lowest Observed Effect The lowest tested concentration which produces a statistically

Concentration (LOEC) significant effect on the test organism in relation to the control

within a certain exposure time

Macrofauna Animals >1 mm.

Macrophyte A member of the macroscopic plant life of an area, especially

of a body of water; large aquatic plant.

Mariculture Cultivation of marine plants and animals in natural and artificial

environments

Marine discharge Discharging wastewater to the marine environment either to an

estuary or the surf zone or through a marine outfall (i.e. to the

offshore marine environment)

Marine environment Marine environment includes estuaries, coastal marine and

near-shore zones, and open-ocean-deep-sea regions.

Meiofauna Animals <1 mm.

Near-field A region in close proximity to the discharge where buoyancy

forcing and entrainment effects primarily determine plume

dynamics


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page xxv

Oligotrophic Refers to a body of water with very low nutrient levels. Usually

these waters are poor in dissolved nutrients, have low

photosynthetic productivity, and are rich in dissolved oxygen.

Osmotic pressure The hydrostatic pressure produced by a solution in a space

divided by a semi-permeable membrane due to a differential in

the concentrations of solute.

Pollution The introduction of unwanted components into waters, air or

soil, usually as a result of human activity; e.g. hot water in

rivers, sewage in the sea, oil on land.

Population Population is defined as the total number of individuals of the

species or taxon.

Pycnocline A transition layer of water in the ocean, with a steeper vertical

density gradient than that found in the layers of ocean above

and below, i.e. a narrow range of depths at which density

changes abruptly between warm surface waters and deeper,

colder waters.

Recruitment The replenishment or addition of individuals of an animal or

plant population through reproduction, dispersion and

migration.

Reverse Osmosis A filtration process that removes dissolved salts and metallic

ions from water by forcing it through a semi-permeable

membrane that removes molecules larger than the pores of the

membrane.

Sediment Unconsolidated mineral and organic particulate material that

settles to the bottom of aquatic environment.

Sludge Residual sludge, whether treated or untreated, from urban

wastewater treatment plants

Species A group of organisms that resemble each other to a greater

degree than members of other groups and that form a

reproductively isolated group that will not produce viable

offspring if bred with members of another group.

Subtidal The zone below the low-tide level, i.e. it is never exposed at

low tide

Supralittoral zone Also known as the spray zone, is the area above the spring

high tide line that is regularly splashed, but not submerged by

ocean water.


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page xxvi

Surf zone Also referred to as the ‘breaker zone’ where water depths are

less than half the wavelength of the incoming waves with the

result that the orbital pattern of the waves collapses and

breakers are formed.

Suspended material Total mass of material suspended in a given volume of water,

measured in mg.ℓ-1l.

Suspended matter Suspended material in the water column.

Suspended sediment Unconsolidated mineral and organic particulate material that is

suspended in a given volume of water, measured in mg.ℓ-1.

Swash zone part of the foreshore washed by waves

Tainting This refers to the change in taste of seafood products as a

result of the presence of objectionable chemical constituents

which may greatly influence the quality and market price of

cultured products.

Taxon (Taxa) Any group of organisms considered to be sufficiently distinct

from other such groups to be treated as a separate unit (e.g.

species, genera, families).

Thermocline A transition layer of water in the ocean, with a steeper vertical

temperature gradient than that found in the layers of ocean

above and below, i.e. a narrow range of depths at which

temperature changes abruptly between warm surface waters

and deeper, colder waters.

Toxicity The inherent potential or capacity of a material to cause

adverse effects in a living organism.

Turbidity Measure of the light-scattering properties of a volume of water,

usually measured in nephelometric turbidity units.

Turgor pressure The pressure of the cell contents against the cell wall, in plant

cells, determined by the water content of the vacuole, resulting

from osmotic pressure. i.e. the hydrostatic pressure produced

by a solution in a space divided by a semipermeable

membrane due to a differential in the concentration of solute.

Vulnerable A taxon is vulnerable when it is not Critically Endangered or

Endangered but is facing a high risk of extinction in the wild in

the medium-term future.


EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay page 1

1 INTRODUCTION

The Iron Ore Handling Facility at the Port of Saldanha (including the proposed upgrades to the

facility), will require increased water supplies for dust suppression measures. It is anticipated that

ultimately a total of 3600 kℓ of water per day will be required for this purpose. The municipality

currently supplies water to the Iron Ore Handling Facility, but Transnet propose to supplement this

supply with additional sources of freshwater. Of all potential sources of such water, seemingly the

most feasible is that of a Reverse Osmosis (RO) seawater desalination plant.

In terms of existing legislation this proposed activity (i.e. the construction and operation of an RO

Plant) will require that a Basic Assessment of potential environmental impacts be undertaken.

1.1 Background

SRK Consulting and PD Naidoo & Associates (Pty) Ltd have been jointly appointed by Transnet to

undertake a Basic Assessment for a proposed RO plant at the Port of Saldanha. The principal

objectives of the Basic Assessment are to:

• Assess the potential impacts associated with the construction and operation of the proposed

RO plant;

• Provide an assessment of all the alternatives, including the three location alternatives

(including various combinations of intake and discharge infrastructure at the various sites) and

the no development alternative;

• Indicate whether the sites are environmentally acceptable or unacceptable for an RO plant in

terms of the respective impacts assessed by the relevant specialists;

• Recommend appropriate and practical mitigation measures to minimise the negative impacts

and maximise potential benefits associated with the three sites; and;

• Indicate the environmentally preferred site.

During the screening phase of the project a number of specialist studies were identified as being

necessary. These included a specialist study on the potential environmental impacts in the marine

environment, with a focus on predicting i) the behaviour and footprint of the brine discharge plume for

a number of proposed discharge locations and under a representative range of environmental

conditions, and ii) the associated potential impacts on the marine ecology.

Consequently, the PDNA/SRK Joint Venture approached the CSIR and Pisces Environmental

Services (Pty) Ltd to undertake a Marine Specialist Study to assess the impacts of the construction

and operation of the RO Plant on the marine environment.



1.2 Scope of Work

The detailed Terms of Reference specific for this Specialist Marine Impact Assessment are to:

• Advise on the design of the RO Plant early in the process;

• Identify any fatal flaws associated with any of the alternative sites, or the proposed project;

• Recommend and implement additional terms of reference, based on professional expertise and

experience;

• Based on previous specialist studies (e.g. for the Phase 1B and Phase 2 iron-ore Terminal

Upgrade EIAs), existing knowledge of the area, and additional studies where required, provide

a brief baseline description of the marine environment (including aspects relating to water

quality, currents, marine ecology, etc.) that may be affected by the proposed RO Plant;

• Assess the significance of the potential impact of the proposed development on the marine

environment for each of the alternative sites (taking into consideration the differences in project

design in the case of each alternative site), according to the standard impact assessment

methodology supplied by PDNA/SRK Joint Venture and outlined in Section 4.1.3 of this report;

• Briefly describe any relevant legislative and permitting requirements that apply to the location

of the RO Plant (e.g. authorisation required for the construction of structures below or near the

high water mark, disposal of brine to the marine environment, etc);

• Recommend mitigation measures to minimise impacts associated with the proposed RO Plant

and enhance benefits;

• Identify an environmentally preferred site;

• Prepare a report comprising both baseline information and an impact assessment;

• Attend an integration workshop with the project team to discuss all the specialist studies;

• Address marine issues raised by registered Interested and Affected Parties; and

• If required, to attend a public meeting to discuss the issues highlighted during the public

participation phase.

1.3 Structure of the Report

This report describes the effects on the marine environment (i.e. the coastal zone below the high

water mark) of the construction and operation of the proposed RO Desalination Plant and

significance within the context of the receiving environment in Saldanha Bay. The report outlines the

approach to the study, assesses impacts identified by marine specialist consultants, and makes

recommendations for mitigation, monitoring and management of these impacts. The report is

structured as follows:



• Chapter 1: Introduction - provides an introduction and background to the proposed project,

outlines the Terms of Reference and report structure and includes a brief overview of the

legislation relevant to the proposed activity;

• Chapter 2: Project Description - provides an overview of the proposed RO Plant, giving

some technical detail on the project designs being considered and the volume, nature and

water quality of the proposed discharges from the RO Plant;

• Chapter 3: Identification and Selection of Alternatives - describes the study area, and

identifies and summarises various alternative project scenarios that will be considered as part

of the proposed development;

• Chapter 4: Approach to the Study - provides an overview of the modelling approach applied

in the assessment and the scenarios simulated in the modelling study. Information sources,

assumptions and limitations to the study are provided, and the assessment methodology is

outlined;

• Chapter 5: Description of the Affected Environment - briefly describes the receiving

biophysical environment that could be impacted by the RO Plant. Existing impacts on the

environment are discussed and sensitive and/or potentially threatened habitats or species are

identified;

• Chapter 6: Identification of Key Issues and Sources of Potential Environmental Impact –

here key issues identified during the public consultation and environmental screening process

for the proposed RO Plant are identified and summarised in terms of the construction phase,

operational phase and decommissioning phase;

• Chapter 7: Anticipated Changes in the Marine Environment due to the Effluent - provides

a detailed description of the discharge plume modelling studies undertaken and a summary of

the modelling results;

• Chapter 8: Assessment of Environmental Impact - identifies and assesses the significance

of potential direct, indirect and cumulative environmental impacts on the marine environment

associated with the construction and operation of the RO Plant, based on information provided

by the client and the results of the modelling studies. The impacts identified for each of the

three alternative sites are assessed and summarised separately;

• Chapter 9: Conclusions and Recommendations - here the environmental acceptability of

the development alternatives are discussed, and the environmentally preferred alternative is

identified. A comparison between the “no development” alternative and the proposed

development alternatives is also included. Mitigation measures and monitoring

recommendations are presented;

• Chapter 10: References - provides a full listing of all information sources and literature cited

in this document.



1.4 Legislative and Permitting Requirements

The primary policy and legislation that needs to be considered is that related to the brine (and co-

discharges) from the proposed RO Plant. Under South African legislation the effluent water

discharge (and potential associated co-discharges) from the proposed RO plant in Saldanha Bay is

classified as “industrial wastewater” and thus requires a license, under Section 21 of the National Water Act 1998 (NWA) for disposal to the marine environment. In terms of policy, legislation and

practice the following documents are of relevance:

• South Africa’s “Operational policy for the disposal of land-derived wastewater to the marine environment” (DWAF, 2004a-c) that captures the legislative framework of relevance to

this study;

• National Water Quality Management Framework (DWAF, 2002); and;

• Documentation of relevance to licensing a wastewater discharge to the marine environment

(e.g. DWAF, 2000, 2003a, 2003b).

The key management institutions and role-players in the decision on the acceptability of the proposed

water discharges into Saldanha Bay are the Saldanha Bay Water Quality Forum Trust (SBWQFT) in

an advisory role, the DWAF2 and the DEAT, all of whom will ensure transparent and adequate

stakeholder and public participation.

Specifically, environmental quality objectives need to be set for the marine environment, based on

the requirements of the site-specific marine ecosystems, as well as other designated beneficial uses

(both existing and future) of the receiving environment. The identification and mapping of marine

ecosystems and the beneficial uses of the receiving marine environment provide a sound basis from

which to derive site-specific environmental quality objectives (Taljaard et al. 2006). To ensure that

environmental quality objectives are practical and effective management tools, they need to be set in

terms of measurable target values, or ranges for specific water column and sediment parameters, or

in terms of the abundance and diversity of biotic components.

The South African Water Quality Guidelines for Coastal Marine Waters (DWAF 1995) provide

recommended target values (as opposed to standards) for a range of substances, but these are not

exhaustive. Therefore, in setting site-specific environmental quality objectives, the information

contained in the DWAF guideline document is supplemented by additional information obtained from

2 The DWAF previously has indicated that the Water Quality Management Plan developed by the SBWQFT

and associated studies (e.g. Taljaard and Monteiro, 2002, Monteiro and Kemp, 2004), together with representations from the DEAT, will play a significant role in its decision-making, i.e. the present Saldanha Bay water quality management plan and associated scientific studies, together with the relevant policy legislation and licensing procedures, will form the basis upon which a decision is made on the viability of the proposed RO Plant discharge into Saldanha Bay and the subsequent granting of a licence for such a discharge. The reader is referred to Section 8.1.3 of this report for more detailed discussion of the requirements of the SBQWQFT Water Quality Management Plan and associated requirements.



published literature, best available international guidelines (e.g. ANZECC 2000; World Bank 1998),

as well as site-specific data and information (e.g. obtained through numerical modelling outputs).

The relevant and expanded set of target values for the Benguela Current Large Marine Ecosystem

countries (Angola, Namibia and South Africa) are summarised in Taljaard (2006).

Saldanha Bay supports an important mussel farming industry. In South Africa, standards controlling

the quality of fish and shellfish flesh for human consumption (i.e. concentration limits of constituents

as required by law) are set out in the following legislation:

• Foodstuffs, Cosmetics and Disinfectants Act (Act 54 of 1972), Regulation - Marine food, 2

November 1973.

• Foodstuffs, Cosmetics and Disinfectants Act (Act 54 of 1972), Regulations related to metals

and foodstuffs, 9 September 1994.

In principle, these food quality standards should be met if the quality of the water from which these

organisms are harvested or cultured complies with the recommended target values for mariculture,

as specified in the South Africa Water Quality Guidelines for Coastal Marine Waters (DWAF, 1995).

The specific water quality guidelines adopted for this study are summarized in Chapter 8.

The possible implications of the policy, legislation and approval and licensing procedures referred to

above, for the proposed RO Plant water discharge into Saldanha Bay, are summarized in Appendix

A.

In addition to the above, there will need to be due consideration of the requirements of the following

policy and legislation related primarily to the siting and construction of the RO Plant and its

associated intake and discharge infrastructure. A more detailed description of these policy and

legislation requirements is given in DWAF (2007). It should be noted however that in this specialist

report, only marine impacts (i.e. impacts below the high water mark) are considered, with a specific

focus on the intake of seawater and discharge of brines from the proposed RO Plant and associated

infrastructure (including the impacts of possible intake and discharge through beach wells where

deemed feasible). Construction impacts and any ongoing impacts (e.g. disruption of sediment

transports in the marine environment) therefore are limited to those occurring below the high water

mark. Thus only specific aspects (text in bold) of the policy and legislation described below is of

relevance to this study.



In addition to the requirements of the NWA policy and legislation, the following also are of relevance

to the siting, construction and operation of an RO Plant:

• National Environmental Management Act (Act No 107 of 1998): The NEMA serves as the

framework legislation, promoting sound environmental management, co-operative governance

and sustainable development. The NEMA also contains the principle that environmental

management must place people and their needs at the forefront of its concern, and serve their

physical, psychological, developmental, cultural and social interests equitably. Specifically, one

of these principals, according to Section 2(4)(r), is that: “sensitive, vulnerable, highly

dynamic or stressed ecosystems, such as coastal shores, estuaries, wetlands or similar systems require specific attention in management and planning procedures, especially

where they are subjected to significant human resource usage and development pressure”. As summarised in DWAF (2007), of specific interest to the marine impacts of

desalination are the following activities, which are listed in Regulation 386 (Appendix C) and

which therefore need to follow a “basic assessment” procedure:

2: Construction or earth moving activities in the sea or within 100 metres inland of the high-water mark of the sea, in respect of –

2(d) embankments

2(e) stabilising walls;

2(f) infrastructure, or

2(g) buildings

3: The prevention of the free movement of sand, including erosion and accretion, by

means of planting vegetation, placing synthetic material on dunes and exposed sand

surfaces within a distance of 100 metres inland of the high-water mark of the sea.

5: The removal or damaging of indigenous vegetation of more than 10 square metres

within a distance of 100 metres inland of the high-water mark of the sea.

6: The excavation, moving, removal, depositing or compacting of soil, sand, rock

or rubble covering an area exceeding 10 square metres in the sea or within a distance of 100 metres inland of the high-water mark of the sea.

12: The transformation or removal of indigenous vegetation of 3 hectares or more or of

any size where the transformation or removal would occur within a critically endangered

or an endangered ecosystem listed in terms of section 52 of the National Environmental

Management: Biodiversity Act, 2004 (Act No. 10 of 2004).

13: The abstraction of groundwater at a volume where any general authorisation issued

in terms of the National Water Act, 1998 (Act No. 36 of 1998) will be exceeded.

25: The expansion of or changes to existing facilities for any process or activity, which requires an amendment of an existing permit or license or a new permit or license in terms of legislation governing the release of emissions, pollution, effluent.



NEMA also governs issues such as the Control of Vehicles in the coastal zone. The

Minister has proclaimed regulations under section 44 of the NEMA to control the use of

vehicles in the coastal zone in order to protect and conserve the sensitive ecosystem of the

coastal zone. The regulatory authorities have indicated that Transnet will not need to obtain a

permit for vehicle access to the beach for the pre-feasibility studies and construction activities

proposed.

• Integrated Coastal Management Bill (Draft): The main purpose of this Bill is to establish a

system of integrated coastal and estuarine management in South Africa, including norms,

standards and policies, in order to promote the conservation of the coastal environment, and

the ecologically sustainable development of the coastal zone. Chapter 7 of the Bill deals with

the protection of coastal resources.

Section 57 defines the principal that adverse effects need to be avoided, and also provides a

clear link with NEMA, which will be discussed. Section 57(b) specifically draws the attention to

the operator of a pipeline that ends in the coastal zone (sub-section (iv)) and any person who

produce a substance which causes an adverse effect (sub-section (v)). Section 58 stipulates

that authorisation needs to be applied for at the relevant authority if an activity which might have a possible adverse effect on the marine environment is planned.

Section 63 of the Bill deals with specific activities which are prohibited in the “coastal buffer

zone” (see DWAF, 2007 for more detail). Activities listed in Part A of Schedule 3 are prohibited

in the coastal buffer zone, while activities listed in Part B of Schedule 3 require a permit. All

activities within coastal public property and exclusive economic zones listed in Part C of

Schedule 3 also require a permit.

Activities of importance to desalination in Part A include:

• The disposal of solid waste, rubble, unprocessed sewage or any other effluent likely to

cause an adverse effect on the coastal environment.

Activities of importance to desalination in Part B include:

• The erection, construction, placing, or any significant alteration or extension of a building

or structure;

• The construction or any significant alteration or extension of a road;

• The clearing of indigenous vegetation other than cultivated indigenous vegetation;

• The stabilization or destabilization of dunes.

Activities of importance to desalination in Part C include:



• The erection, construction, placing, alteration or extension of a building or structure on or

in any coastal public property, including an artificial reef, or any structure designed to

prevent coastal erosion or to promote accretion of the seashore;

• The disturbance of any coastal public property in a manner that has or is likely to

have an adverse effect on the coastal environment, including any excavations,

dredging, draining, drilling or tunnelling;

• The destruction, damage or disturbance of any coastal public property in a manner

that has or is likely to have an adverse effect on biodiversity or habitat;

• The abstraction of water from coastal waters for agricultural, commercial or

industrial purposes, including for aquaculture and desalination, or in a manner that is likely to have an adverse effect;

• The stabilization or destabilization of dunes.

Section 74 of the Bill makes provision for permits to be issued, in co-operation with the NWA,

to discharge effluent into coastal waters.3 No reference, however, is made to the need to apply

for the abstraction of seawater (DEAT, 2007). The requirement for such a permit may however

be implicit in that all activities listed under Part C of Schedule 3 require a permit and one of

these activities is ”the abstraction of water from coastal waters for agricultural,

commercial or industrial purposes, including for aquaculture and desalination, or in a manner that is likely to have an adverse effect.”

• Sea Shore Act (Act No 21 of 1935): The Sea Shore Act provides the legal decision-making

legislation for all sea-related construction work. This administrative function and authority is

delegated to the Provincial Department responsible for environmental management. It should

be noted that the Act also does not address other issues such as the management of the

Admiralty Reserve, found along the high water mark in certain sections of the coast

(Department of Environmental Affairs and Tourism, 2000).

• National Ports Act 12 of 2005: The National Ports Act 12 of 2005 and its delegated

authorities may be of relevance in terms of some of the developments proposed. Specifically,

where a land-based activity, which requires a licence under section 21 of the NWA, falls within

a commercial harbour area, the Transnet National Ports Authority, as the landowner, is

responsible to ensure that such an activity meets the requirements of the relevant laws (DEAT,

2007).

• National Heritage Resource Act (Act No 25 of 1999): The National Heritage Resource Act

(NHRA) states in section 38 that it is required that any person who intends to undertake a development, such as the construction of a pipeline in excess of 300 m in length, or any

3 This bill presently is in draft form and consequently as yet there are no such permit requirements.



development or activity which will change the character of the site, must, at the very early stages, notify the responsible authority and supply details regarding the intended development. If the NEMA requires the assessment of a heritage resource, then the

comments and recommendations of the heritage resource authority must be obtained, and

unless this authority stipulates that section 38 of the NHRA does not apply, the Act clearly

stipulates what such an assessment must include. Should there be construction activities in the marine environment, it is likely that the opinion of a marine archaeologist may be required prior to environmental approval of the project. Note that this specialist study

explicitly excludes issues to be assessed in terms of the NHRA).

Also of relevance here are:

• Marine Living Resources Act 18 of 1998

• National Environmental Management: Biodiversity Act [No. 10 of 2004]

• National Environmental Management Protected Areas ACT 57 of 2003 as amended by the

National Environmental Management: Protected Areas Amendment Act 31 of 2004

If due consideration is given to NEMA, the NWA, the principles in the Integrated Coastal

Management Bill and associated environmental legislation, compliance with the above legislation is

likely to be ensured.

It is however recommended that there is adequate and timeous consultation with both the DWAF and

DEAT regarding the intake and discharges, as well of the infrastructure, associated with the proposed

RO Plant. Transnet Projects has indicated that such consultation has occurred and is ongoing.






2 PROJECT DESCRIPTION

2.1 Description of the RO Plant Facilities

The proposed RO Plant comprises the following:

• An RO plant containment building (approximately 1500 m² surface footprint) with room for three

RO modules, an electric sub station, a motor control room, a pump house, a store room, office

and ablution facilities and space for parking;

• Sea water supply and brine discharge via pipelines, beach wells or boreholes;

• Submersible pumps and piping for the extraction and discharge of sea water;

• A sea water intake storage tank alongside the RO building;

• A potable water storage tank alongside the RO building;

• Storage reservoir(s) totalling 5 Mℓ, 48 hour potable water storage capacity next to the existing

potable water reservoir or at the alternative site adjacent to the borrow pit area;

• A small service road connecting the RO building to the nearest road infrastructure; and

• All requisite electrical and communication facilities between the RO installations and the Iron

Ore Handling Facility.

The basic process for the treatment of water in the proposed RO plant is as follows (PDNA/SRK Joint

Venture, 2007). Sea water is supplied through a seawater intake and appropriately treated or

supplied and pre-filtered through beach wells or boreholes and then pumped to a seawater buffer

storage tank (Figure 2.1). The salt water is then pumped at high pressure through to the RO

membranes. The desalinated water is piped to the potable water tank and the concentrated sea

water (brine) is released back in to the ocean through discharge pipes with appropriate diffuser

structures or beach wells. The concentrated sea-water may or may not require treatment before

being disposed, but this will be addressed as part of this specialist study.

Figure 2.1: Indicative components of an RO Plant (PDNA/SRK Joint Venture, 2007).



2.2 Life Span of Project

The anticipated life span of the RO plant is 25 years. After this period it is likely that the RO units will

be refurbished or replaced and that the RO Plant will continue to operate. It is, however, not clear

whether or when “re-construction” or major maintenance will be required should beach

wells/boreholes be used at the point of intake and/or discharge as considered for all 3 of the

proposed sites.

2.3 Proposed Plant Capacity

There is an immediate requirement for potable water/freshwater for dust suppression measures

associated with the Phase 1A and 1B expansion of the Saldanha Bay iron-ore export facilities. It is

intended that the initial phase of the RO Plant (Phase 1A/1B) comprising output from a single RO unit

(1200 kℓ/day capacity) be commissioned in approximately March 2009, depending on authorisation of

the project by DEAT. The freshwater requirement will increase with the proposed further Phase 2A

development of the iron-ore export facilities when the requirement will be such that a further 2 RO

units will need to be commissioned. It is anticipated that the additional 2 RO units will need to be

operational some 6 months to one year after the initial commissioning of the first RO unit.

The development phases and the associated RO plant capacities required are indicated in Table 2.1

below. These are based on the potable water requirements for the various phases as tabulated in

Figure 11 of the Saldanha Iron Ore Terminal Water Supply and Demand Report (Transnet Report by

Alistaire Dick, 28 May 2007) and as amended in feedback at a meeting with Transnet (SRK offices on

13 August 2007) where a potable water recovery rate of 45% was specified.

Table 2.1: Proposed RO plant capacity for the various development phases of the Saldanha Bay Iron-ore Export Expansion project.

Development Phase

Number of 1200 kl

RO Units Required

Potable Water Produced (m3/day)*

Brine Discharge Volumes (m3/day)

Intake Volumes (m3/day)

Phase 1A/1B 1 1200 1467 2 667 Phase 2A/2B 3 3600 4 400 8 000 * 1000 m3/day = 1000 kℓ/day = 1 Mℓ/day

It should be noted that, although the intention would be to initially install only one of the three RO

units, the full capacity of the proposed plant with three units has been assessed for the purposes of

the Basic Assessment and this marine specialist study.



2.4 Construction and commissioning of infrastructure

It is planned that the RO support infrastructure (pipelines, beach wells, boreholes, buildings, etc)

however will be sized for the full three modules (3600 kℓ/day) and installed during the initial phase of

the project (approximately Sept 2008 to March 2009). This may include potential beach wells for

both intake of seawater and/or discharge of the brine from the RO Plant. Should beach wells be

constructed and/or intake discharge pipelines be installed, the full infrastructure will be installed

during the initial construction phase. Consequently no further disruption of the beach is expected

once the RO support infrastructure is installed during the initial construction phase. Thus it will only

be the additional RO modules that will be installed in a phased approach (i.e. initial installation of one

RO unit followed by installation of additional 2 RO units some 6 to 12 months later). It is not

anticipated that the installation of the additional 2 RO units will have any environmental impacts as

the installation activities will be confined to buildings within the existing infrastructure.

Construction activities of relevance to assessing marine impacts include:

• The construction of seawater intakes that may include the impacts of beach well construction

for intake waters, beach sumps and other appropriate intake infrastructure;

• The construction of discharge structures (including possible beach well discharges).

Beach wells for intake are being proposed only for Site 1 and Site 2 (see Figure 3.2 in Section 3). The option of discharge beach wells, initially considered for only Site 1, is no longer deemed to be feasible (SRK, 2007) and is consequently excluded from this marine impact assessment. Discharge beach wells thus are not being considered for any of the proposed sites. The construction activities associated with intake and discharge structures are as follows. The installation of pipeline intakes and/or discharges will require the standard techniques for the construction of such infrastructure that include the provisions of an on-land construction site (albeit limited) for pipeline laying activities and possible other activities such as trenching through the surf-zone and/or associated activities below the high-water mark. Presently the siting and detailed design of such intake pipelines (and associated intake structures) as well as the exact nature of the discharge pipelines and discharges structures (diffusers, etc) remains non-specific. It is intended that this study of potential environmental impacts, together with other engineering design considerations, will inform the exact nature of such infrastructure. Consequently, presently only a generic description of construction activities is possible. Should there be a requirement for the drilling of beach wells, access will be required for a drilling rig and support vehicles to possibly provide drilling media, fuel, etc. The exact nature of the construction activities will depend on the beach well design which will come from the successful RO Plant tenderer. The drilling activities involved are similar to those that will be employed to sink the test wells as described for the groundwater assessment studies (Visser et al., 2007).



The physical layout of the port is likely to change over time. In particular the proposed Phase 2A and 2B developments4 will result in significant infrastructural changes in the marine environment, however, only Site 1 and some of the discharge options from Site 3 (see Section 3 for details of the proposed RO Plant locations) is likely to be affected by these changes. Under the assumption that the RO discharge plume and its associated impacts will be of limited

spatial extent (and therefore not substantively affected by the proposed changes to the marine

environment i.e. dredging and reclamation), the initial assessment considers only the proposed RO

discharges under the Phase 1A and 1B development scenario, i.e. under a development scenario

where there are no changes in the marine environment compared to the presently existing situation.

Had these initial results indicated that the RO discharge plume extents would be significant, it would

have been necessary to assess further development scenarios (i.e. Phase 2A and 2B) in the

assessment. The proposed layouts associated with the Phase 2A and 2B developments are given in

Figures 2.2a and 2.2b.

In all cases the plume dynamics may be affected to some degree, however these differences in plume behaviour are not deemed sufficient to warrant detailed modelling studies of these discharge options under the various Phase 2 development scenarios. As noted above, changes in the discharge plume behaviour under the Phase 2 development scenarios are only likely to be substantive for Site 1 and possibly two of the discharge options for Site 3, namely brine discharges alongside the causeway into Big Bay and at caisson 3. Should the large reclamation area be constructed during Phase 2 (see Figures 2.2a and 2.2b) there is likely to be some changes in the plume dynamics for a discharge at Site 1 (development options 1a & 1b). These changes are likely to result in stronger flows at or near the discharge location, particularly under southerly wind conditions. This is likely to result in greater dispersion of the plume. However a relatively quiescent area is likely to develop in the lee of the reclamation area (between the reclamation area and the causeway), particularly under southerly wind conditions. There is potential for the accumulation of discharged brine should it reach this area before being significantly dispersed. The modelling results indicate that the plume dimensions are likely to be sufficiently limited that this will not occur. The changes in the plume dynamics for a discharge at Caisson 3 from Site 3 are expected to be limited, the most likely change being enhanced dispersion of the RO Plant discharge plume due to increased tidal (and upwelling) flows past Caisson 3 associated with extensions in the shipping channel proposed for Phase 2.

4 Phase 2A will expand the iron-ore handling capacity of the Port of Saldanha to approximately 67 MTPA

and will require the construction of an additional single berth. It is anticipated that the dredging activities associated with the Phase 2A development will commencce in January 2009, however there remains a degree of uncertainty around this date due to a number of factors (e.g. dredge equipment availability, environmental approvals, etc). Phase 2B (following immediately after Phase 2A or delayed for a period of approximately 5 to 10 years later) will expand the iron-ore handling capacity of the Port of Saldanha to approximately 93 MTPA and will require extension of the dredge channel and the construction of a further additional berth (i.e. a total of two additional berths compared to presently existing berthing facilities).



Figure 2.2a: Proposed layouts for the Phase 2A development of Iron-ore export facilities in Saldanha Bay.

Figure 2.2b: Proposed layouts for the Phase 2B development of Iron-ore export facilities in Saldanha Bay.

Proposed Reclaim Area dimensions

Proposed Phase 2A shipping channel

Proposed Reclaim Area dimensions

Proposed Phase 2B shipping channel



A discharge into Big Bay from Site 3 (development option 3b) may result in the accumulation of warm salty waters in the dredge channel when developed to its maximum proposed extent Phase 2 of the proposed iron-ore export expansion project, should the discharge plume not be sufficiently dissipated before reaching the deeper waters of the proposed shipping channel. The modelling results indicate that under worst case conditions there is a possibility of such an occurrence approximately for up to a total duration of 24 hours within a season, however propeller wash is likely to dissipate any such plume reaching the shipping channel. Such occurrences, if and to the extent that they occur, will not change the conclusions of this study. Similar behaviour could occur for a discharge from Site at Caisson 3, however the potential consequences at this site are considered to be significantly less than indicated above for a discharge from Site 3 into Big Bay. The reason for this is that the plume dimensions for a discharge at Caisson 3 are significantly smaller than for a discharge into Big Bay from Site 3. Furthermore, the tidal flushing at Caisson 3 and consequently dispersion of the brine plume is likely to increase under the Phase 2 development scenarios (although the brine may spread somewhat more extensively into the proposed extended dredge channel into Big Bay).

The observations indicate that a detailed modelling study of the potential RO Plant discharges under the Phase 2 port layouts is not indicated



3 IDENTIFICATION AND SELECTION OF ALTERNATIVES

3.1 Geographical Setting

Saldanha Bay is situated on the west coast of South Africa, approximately 120 km north of Cape

Town and is directly linked to the shallow, tidal Langebaan Lagoon (Figure 3.1). The bay contains

five offshore islands, namely Malgas, Jutten, Marcus, Meeuw and Schaapen Island.

Figure 3.1: The Saldanha Bay/Langebaan Lagoon system showing major port

infrastructure.

Saldanha Bay is the only natural harbour of significant size on the west coast of South Africa, and

hosts a substantial fishing industry and fish processing factories. In 1971 the harbour was upgraded

into an international port, which was followed by two major developments:

• a causeway that linked Marcus Island to the mainland, providing shelter for ore-carriers, and

• the construction of the iron ore causeway, which was later extended to provide for the import

of oil.

Reclaim dam



A multipurpose terminal was later added to the iron-ore causeway and a small-craft harbour was built

to cater for the increase in recreational and tourism activities in the bay. The construction of the iron

ore causeway essentially divided Saldanha Bay into two sections: a smaller area bounded by the

causeway, the northern shore and the ore causeway (called Small Bay); and an adjacent larger,

more exposed area called Big Bay, which leads into Langebaan Lagoon. Langebaan Lagoon forms

part of the West Coast National Park and is internationally recognised as a Ramsar site in terms of

the Convention on Wetlands of International Importance, especially as waterfowl habitat. Small Bay,

Big Bay and Langebaan Lagoon can be considered to comprise one large ecosystem with strong

interdependencies between the various regions in this semi-enclosed coastal embayment.

Reference to a semi-enclosed coastal embayment does not imply that there is not significant

exchanges and flushing of the various sub-components (i.e. Small Bay, Big Bay and Langebaan

Lagoon) by water from the adjacent continental shelf. Estimated exchanges between the various

regions of the bay are as tabulated in the footnote below5.

3.2 Siting Alternatives

Three alternative locations within the Port of Saldanha are being considered for the location of the

RO plant (Figure 3.2). At the three proposed sites, there are various opportunities and constraints in

terms of the specific intake and discharge alternatives that may be considered (see Table 3.1).

These are described in generic terms, followed by a detailed description of the various alternatives

assumed and assessed for the three proposed locations for the RO Plant.

A detailed description of the three proposed sites follows.

• Alternative 1: Located to the east of the Iron Ore Handling Facility. The RO plant would be

situated in the vicinity of the coastal dunes, with both intake and discharge occurring in Big

Bay. It is proposed that the intake waters are provided via seawater supply beach wells or a

pipeline intake from the surf-zone. The discharge infrastructure comprises a 20 to 30 m

“diffuser” pipeline laid along the existing revetment. Brine discharge would be into the larger

5 Estimated exchanges between the various regions of the bay are as tabulated below.

Exchange fluxes* Location of cross-section Neap tides

(m3.s-1) Spring tides

(m3.s-1)

Flux across the mouth of Small Bay 210 1 680

Flux across the mouth of Big Bay (i.e between Saldanha Bay and the adjacent continental shelf)

1 100 7 950

Flux across the mouth of Small Bay(both channels) 500 3 260

* It should be noted that the salt input into Saldanha Bay from the RO Plant discharge are only 0.012%, 0.06% and 0.026% of the salt fluxes through the mouth of Big Bay, the mouth of Small Bay and through the two channels at the mouth of Langebaan Lagoon under Neap tides, respectively . Under Spring tides the percentages are almost an order of magnitude less than those for Neap tides.



Saldanha Bay (Big Bay). The two intake and discharge combinations considered under

alternative 1 are summarised in Table 3.1.

• Alternative 2: Located north and northwest of the Iron Ore Handling Facility. The RO plant

would be established in an area currently containing stockpiles of gravel and construction

rubble. Intake options include beach well intakes or surf-zone (sump) intakes. The discharge

option considered is a pipeline discharge into the North-eastern corner of Small Bay. The two

intake and discharge combinations considered under alternative 2 are summarised in Table

3.1.

• Alternative 3: Located on the southern section of the Quay of the Iron Ore Handling Facility.

The RO plant would be positioned on a gravel area adjacent to the Multi-purpose Terminal

(MPT). There is no beach of sufficient size in this area for intake beach wells. Consequently,

seawater supply and brine discharge would need to be via a pipeline intake or borehole intakes

along the causeway. A number of alternative discharge locations are possible, namely a

discharge into Small Bay, a discharge at depth into Big Bay or a discharge at Caisson 3 at the

end of the causeway. The four intake and discharge combinations considered under

alternative 3 are summarised in Table 3.1.

Figure 3.2: Proposed locations for the RO Plant.

Small Bay

Big Bay



3.3 Conceptual Designs of Possible Intake and Discharge Structures

The conceptual designs of possible intake and discharge structures follows.

3.3.1 Intake Alternatives The three intake alternatives for the supply of seawater to the RO Plant, namely beach wells,

boreholes and feed-water pipelines, are described in more detail below:

Beach Wells Beach intake wells with submersible pumps have been proposed as a means of supplying seawater

to the RO Plant. The proposed intake wells will consist of perforated PVC and stainless steel

cylinders of 300 - 500 mm diameter equipped with continuous-slot stainless steel screens. The wells

will be sunk above the high water mark vertically to a maximum of 20 m into the beach sediments.

The tops of the wells will protrude out of the surface of the beach by approximately one metre. As an

alternative, horizontal collector wells with horizontal collector pipes could be used. Seawater filtered

through the sand will percolate into the wells from where it will be pumped by underground pipeline

(not exceeding 360 mm internal diameter) to the seawater buffer tank. However, the amount of water

that can be taken in by subsurface intakes is a function of the type of substrate, its permeability and

other geotechnical characteristics (California Coastal Commission, 2004). Beach wells will not work

well in areas where the substrate contains high proportions of silt or clay, or where the sandy

substrate is unstable or insufficiently deep to ensure that the intakes are not exposed during seasonal

sand movement or storms.

For the Saldanha Bay RO Project, this has been determined for Site 1 by a specific geohydrological

groundwater study (Visser et al., 2007) and shoreline stability studies associated with proposed

future port developments in Saldanha Bay (Smith et al., 2007). Numerical modelling conducted as

part of the groundwater study indicated that ~10 vertical intake wells drilled 50 m apart parallel to the

shoreline above the high water mark at Site 1 would be required to abstract the anticipated raw water

demand of 8000 m3/day for the RO plant (Figure 3.3). Alternatively, if horizontal collector wells are

decided on, at least two with ~100 m of horizontal collector pipes each will be required (Visser et al.,

2007). Although no similar detailed studies have taken place to date for the proposed beach well

intakes at Site 2, a worst case scenario of up to 10 intake wells, spaced ~50 m apart have been

roposed for this site (Figure 3.4).



Figure 3.3: Location and extent of intake beach wells at Site 1 (PDNA/SRK Joint Venture, 2008).



Figure 3.4: Location and extent of intake beach wells at Site 2 (PDNA/SRK Joint Venture, 2008).



Designed with appropriate intake velocities and installed at the proper depth within the substrate,

beach wells can operate with negligible effects on local marine life. A further operational advantage

is that the natural filtering effects of the beach sediments ensure that the feed-water is comparatively

clean, thereby reducing the necessity for the use of pre-treatment substances, and thus reducing

operating costs as well as likely discharges of concern.

Single-point Intakes

As an alternative, feed-water may be supplied by a single-point intake. The intake will be located

either in the surf-zone or just beyond (Site 1), in the quay wall (Site 3), or up to 75 m offshore (Site 2).

Surf-zone intakes typically consist of one or more excavated sumps sunk into the sediments a

distance of ~15-20 m below the high water mark in the surf-zone. The sumps will be covered in sand

transported through the area by waves and currents. The feed-water is thus drawn in through the

overlying sediments, which act as a natural filter thereby removing fine particulate material and

improving water quality delivered to the seawater buffer tank. The intakes and associated pipelines

usually deliver seawater to a pump house located above the high water mark, from where the

seawater is pumped to the buffer tank at the RO Plant. In essence therefore, a surf-zone intake is

similar to a beach well intake.

However, depending on specific circumstances an alternative deeper water structure may be

considered that lies beyond the active surf-zone. Pipeline intakes for desalination plants are

commonly 1-m in diameter and composed of HDPE (High Density Polyethylene), however for the

small volumes being considered here the pipelines diameters are likely to be only 0.2 to 0.4 m

diameter HDPE pipelines for intakes and 0.25 to 0.5 m diameter HDPE pipelines for discharges. The

pipeline diameter must be sufficiently large so that the total headloss is limited, but not too large as

this may result in settlement of solids within the pipeline. HDPE is a flexible material and can

relatively easily accommodate seabed level fluctuations without leading to unacceptable stresses in

the material. The seaward end of the pipeline is usually stabilised within a concrete intake structure

placed on the seabed.

To obtain the warmest water possible to ensure maximum efficiency of the RO process, the intake

needs to be near the sea surface, but at the same time at sufficient depth to avoid the ingress of

floating debris and possible oils (i.e. approximately -1 m Chart Datum or deeper). If located too near

the seabed, intakes have a greater chance of entraining sediments re-suspended during rough sea

conditions, or near-bottom low oxygen waters and/or associated poor quality waters (if present). The

intakes will be fitted with appropriate screens to further minimise entrainment of foreign matter.

To avoid scour around the structure through wave and current action, some form of rock protection is

usually placed around the intake structure to ensure its stability. As fine sediments and organic

matter in the feed-water not captured by the screens will block the filters and/or membranes of a RO

Plant, a receiving sump large enough to accommodate the anticipated sediment loads is usually



installed between the intake and seawater buffer tank. The entrained sediment settles in the sump,

which needs to be regularly removed and disposed of in an environmentally acceptable manner.

Onshore, and within the nearshore area off Sites 1 and 2, the feed-water pipeline will most likely

need to be trenched. In the offshore area it will be lying on, or partially within, the seabed sediments,

and will most likely be kept in place by weight collars. For Site 3 the single point intake will be

attached to the quayside.

Borehole Intakes A further intake alternative that has been proposed for Site 3, is the intake of feed-water via a series

of boreholes drilled into the causeway. Two approximate locations have been proposed for these

intake boreholes. The first is a series of boreholes with a 50 m spacing drilled along the rail

embankment toe alongside the existing iron-ore stockpiles (Figure 3.5). The second option

considered for intake boreholes is a linear series of boreholes with a 50 m spacing located adjacent

to the MPT (Figure 3.6).

The actual number of boreholes to be drilled at each of these locations will depend on the seawater

yields from each borehole. Currently up to 10 boreholes are proposed for the rail embankment toe

alongside the existing iron-ore stockpiles (Figure 3.5), or alternatively, up to 10 boreholes adjacent to

the MPT (Figure 3.6). These borehole intakes are expected to have the same advantages as beach

wells in terms of reducing the necessity for the use of pre-treatment substances and thus reducing

operating costs as well as likely discharges of concern.

3.3.2 Discharge Alternatives

Brine-discharge Beach Wells Brine discharge through beach wells was initially considered for Site 1. However, geotechnical

studies have indicated that the transmissivities of the sediments are too low to accommodate

discharge beach wells (Visser et al., 2007).

Surf-zone Discharge

A surf-zone discharge is likely to comprise a concrete discharge channel with or without a screen. It

is likely that the discharge will occur just above the high water mark with the brine flowing into the

active surf-zone.



Figure 3.5: Location and number of intake boreholes alongside the existing iron-ore stockpiles (PDNA/SRK Joint Venture, 2008).



Figure 3.6: Location and number of intake boreholes alongside the existing MPT (PDNA/SRK Joint Venture, 2008).



Single-port Discharge at Depth Where discharge occurs through a pipeline at depth, it is anticipated that the end of the pipeline (if

comprising a single port diffuser) will be directed upwards at approximately 40 to 60 degrees from the

horizontal, to ensure adequate mixing of the dense brine through the water column. The

performance of such a single port diffuser will need to be such that it is consistent with the near-field

behaviour of the brine effluent as assumed in this study. A predicted6 near-field or initial dilution

exceeding 507 is recommended. It may be necessary to design and construct a simple multi-port

diffuser to meet appropriate near-field dilution requirements. In this study, we have assumed a

relatively conservative or “worst case” near-field behaviour (i.e. effluent remains in the bottom one

third to half of the water column depending on water depth), providing some leeway in engineering

design possibilities should they be required to improve the near-field dilution of effluents.

3.4 “No-development” Alternative

The “no-development” alternative for the RO Plant implies the establishment of alternative water

sources for dust control as part of the port expansion. Possible alternatives investigated included

additional potable water from municipal supplies, reclaimed sewage (treated effluent), groundwater,

and seawater. However, due to the potential lack of available yields, environmental costs, and to

ensure suitable water quality, these alternative water sources were not considered as feasible

options during the planning stages of the port development, and will therefore not be dealt with

further here.

3.5 Summary of Development Alternatives

The development alternatives include a range of sites at which it is proposed to locate the RO plant

as well as a number of combinations of intake and discharge infrastructure relevant to each of these

sites. Table 3.1 below summarises all of the development alternatives considered in this study.

More precise detail of the proposed location and combination of the various intake structures together

with associated discharge infrastructure are described in Chapter 7.

6 The predicted near field dilution are typically obtained using near-field models such as CORMIX (Jirka et al.,

1996) or UOUTPLM (Baumgartner et al., 1971; US-EPA, 1985). These models are typically used for engineering design despite their limitations (i.e. their inability to accurately predict near- and/or far-field impacts accurately should there be an accumulation or build-up of effluent in the immediate vicinity of the discharge). These models assume that the effluent, at all times, is being diluted with “clean” ambient waters, which is not always necessarily the case. These models if used in isolation and not in conjunction with far-field models, depending on the flow rates that are considered, do not always provide a conservative assessment. In this study we have assumed conservative near-field behaviours (i.e. have not undertaken near-field modelling) and based the assessment on a far-field model (Delft3D-FLOW that takes into account that the ambient water entrained are not necessarily “clean” but may already be somewhat contaminated.

7 A dilution of 50 implies that each unit volume of effluent is mixed (diluted) with 50 unit volumes of the ambient waters into which the effluent is being discharged.



Table 3.1: Development alternatives used in this assessment (confirmed by Transnet on 22 November 2007).

Site Alternative # Alternative Intake/Discharge Infrastructure Locations

Site 1 8 a) Beach well intake and pipeline discharge (Big Bay)

b) Pipeline intake and pipeline discharge (Big Bay)

Site 2 a) Beach well intake and pipeline discharge (Small Bay)

b) Pipeline intake and pipeline discharge (Small Bay)

Site 3

a) Pipeline intake (Small Bay) and pipeline discharge (Small Bay)

b) Pipeline intake (Small Bay) and pipeline discharge (Big Bay)

c) Borehole intake on quay (adjacent to existing iron-ore stockpiles) and pipeline discharge (Caisson 3, Big Bay)

d) Borehole intake on quay (adjacent to Multi-purpose Terminal) and pipeline discharge (Caisson 3, Big Bay)

Note: For Site 3 the RO Plant building will be located on the southern extremity of the MPT. Other

elements of the infrastructure (water storage reservoirs, etc) may be located closer to the

iron-ore stockpiles.

8 A beach well intake and beach well discharge option was also considered for Site 1, however the

groundwater specialist study (Visser et al., 2007) indicated the beach sediment conditions were such that a beach well discharge is not feasible.



4 APPROACH TO THE STUDY

This marine specialist study comprises two components, namely:

• the physico-chemical characterisation and (where appropriate) assessment of potential

impacts, primarily the brine discharges, on the marine environment due to the construction and

operation of the proposed RO Plant, and;

• an assessment of the associated potential ecological impacts in the marine environment.

The physico-chemical characterisation of potential impacts, mainly comprises a three dimensional

numerical modelling of transport and fate of discharges from the RO desalination plant.

Based on the above physico-chemical characterisation of impacts, an assessment of the potential

ecological impacts in the marine environment has been undertaken. As determined by the terms of

reference for the marine ecological component of this specialist study, this ecological assessment

has adopted primarily a ‘desktop’ approach utilising existing information, for the purpose of the EIA.

Based on the review of available information from international peer-reviewed scientific literature, and

internal reports and EIAs available on the internet, the ecological assessment component of the

report identifies the nature and magnitude of impacts on marine communities likely to result from the

RO Plant, in the context of natural and other anthropogenic impacts in the Saldanha Bay area.

In summary, the impact assessment is structured as follows:

• A description of the marine and coastal ecosystems and biological communities of the region,

including:

− Climate and weather of the region;

− Physical oceanography of the region;

− Water quality within the bay;

− Beach and nearshore ecology;

− Rocky intertidal ecology;

− Pelagic communities and shore birds; and

− Beneficial uses and existing impacts.

• A description of the project background and design, including the identification of all

development options and alternative scenarios to the proposed project, including the ‘no-

development’ option. These have been described as a number of scenarios to be assessed for

potential impacts in the marine environment. As the ‘no-development’ option is unlikely to

influence the potential impacts identified as being associated with the proposed project and

being assessed here, this alternative is not discussed further in this specialist report;

• Identification and assessment of potential environmental impacts (both physical and ecological

impacts, as well as those on beneficial uses in the region);



• Recommendation of mitigation measures and management actions for marine and coastal

issues of concern; and

• Identification of monitoring requirements.

4.1 Methodology

The methods and the motivation for using these methods are described below.

4.1.1 Environmental Baseline The baseline description of the receiving environment in the vicinity of the proposed RO Plant is

limited to a ‘desktop’ approach utilizing existing information and data only. This was sourced and

reviewed from existing studies conducted in the area, as well as information available on the internet.

These included the ‘State of the Bay’ report (Atkinson et al., 2006), scientific literature, previous

environmental impact assessments for various developments in Saldanha Bay, e.g. Phase 1 and 2 of

the expansion of iron ore loading facilities, and from personal knowledge of one of the authors (N.

Steffani), especially of the rocky shore environment.

While a literature research and review are usually adequate for a study of this nature, a lack of

baseline information on sandy beach macrofauna was identified. Consequently, in the event of

beach wells and/or surf-zone discharges being considered in the potential engineering design,

existing information on this habitat is inadequate to confidently base an assessment of the impacts.

More appropriate in this case would be to design and implement a quantitative baseline survey of

beach macrofauna in the area on which to base the assessment. In the absence of such data we

have invoked the pre-cautionary principle in determining the significance of potential impacts, i.e. we

have assumed a significance rating that ensures that the potential impacts are not underestimated.

4.1.2 Modelling The purpose of the modelling undertaken as part of this assessment is to predict the transport and

fate of the brine discharge (including potential co-discharges) from the RO plant for the various RO

plant locations and combinations of intake and discharge structures.

The modelling studies have been undertaken for the maximum planned plant capacity (i.e. 3 RO

Plant modules generating 3.6 Mℓ of potable water per day with an intake volume of 8 Mℓ per day and

a discharge volume of approximately 4.4 Mℓ per day. Based on the project description available at

the commencement of the model simulations, the brine discharge specified in the model is assumed

to comprise an effluent with a salinity of approximately 63 practical salinity units (psu), a zero

temperature elevation above the ambient water temperature at the intake and containing a residual of

oxidising biocide (potentially NaOCl at a discharge concentration of 0.1 mg.ℓ-1) or non-oxidising

biocide (assumed to be DBNPA at a discharge concentrations ranging between 1.15 mg.ℓ-1 and 2.475

mg.ℓ-1). Potential co-discharges and their likely concentrations have been specified, however, the



exact residual concentrations at the discharge point will need to be confirmed. The modelling study

has been designed to accommodate this uncertainty by including the discharge of a conservative

tracer that is a proxy for all co-discharges providing that the behaviour of the co-discharge is

conservative after discharge into the marine environment (i.e. does not decay or transform into other

substances). Should the co-discharge constituent decay or be transformed, the approach taken will

provide a conservative result, i.e. the potential impacts will be assessed as being greater than would

be the case in reality. In the (unlikely) event of the transformation resulting in more toxic constituents,

the modelling approach taken then would not be considered conservative. Based on the co-

discharges specified, no clear potential synergistic effects could be identified.

The discharges are significantly less than those that would be assessed for discharge from, for

example, a power plant (typically 5 to 10 times the magnitude of the discharge being considered

here), of a similar magnitude to those recently assessed for a proposed RO plant discharging to an

open coastline along the East Coast of South Africa, but significantly less (5 and 13.5 times smaller)

than those discharged from an existing RO plant into the Bushman’s River estuary (Bornman and

Klages, 2004).

Potential assessment approaches include:

• Simple near-field studies based on conceptual designs and near-field models. However the

build-up of effluent cannot be simulated by these near-field models as in these models it is

assumed that the effluent is continually diluted by ‘clean’ water. Consequently such near-field

models provide results that are not necessarily conservative, particularly when simulations are

undertaken in relatively poorly flushed embayments such as Small Bay, Saldanha Bay.

• More comprehensive three dimensional (3D) hydrodynamic and water quality modelling where

the potential build-up of effluent is explicitly taken into consideration but where near-field

behaviours need to be schematised (in this case as a bottom discharge or an optimally

designed diffuser where the effluent is mixed throughout the water column or at least a

significant proportion thereof). This type of study does not depend on a specific discharge

design but can consider a broad range of possible near-field behaviours.

• More sophisticated water quality modelling where bay productivity, oxygen concentrations and

nutrient dynamics are explicitly simulated. This is the nature of previous studies undertaken for

potentially larger (5 to 10 times) brine discharges.

Whilst the “footprint” of the brine (and any associated pollutants) is relatively modest and could

possibly be assessed using a near-field model only, here it has been deemed prudent to undertake a

fully three-dimensional hydrodynamic and water quality study to ensure a quantitative assessment

should there be accumulation of effluent around the discharge location. However, given that the

“footprint” or impacted area is relatively modest, it is deemed that a detailed study, where bay

productivity, oxygen concentrations and nutrient dynamics are explicitly simulated, is not warranted

here.



The scenarios simulated will be for the conceptual designs (shoreline discharge, discharge at depth

through single or multi-port diffusers) that are appropriate for the particular discharge location.

A total of 5 discharge locations have been assessed (see Chapter 7), namely:

• a discharge into Big Bay alongside the existing reclamation dam for Site 1;

• a discharge into the NE corner of Small Bay from Site 2;

• a discharge alongside the causeway into Small Bay from Site 3;

• an alternative discharge from Site 3 alongside the causeway into Big Bay; and

• a discharge located at Caisson 3 in the vicinity of the southern extremity of the causeway

separating Small Bay and Big Bay.

Each of the above discharge scenarios will be simulated for a range of environmental conditions

comprising:

• a two month late summer period when the bay is most stratified resulting in relatively quiescent

bottom waters and limited vertical mixing;

• a two month winter period when the water column is well-mixed resulting in more active vertical

mixing and typically stronger bottom flows; and

• a “worst case” one month simulation comprising a both calm and stratified period (typically April

to June in Saldanha Bay) that is representative of the calmest period in 15 years or more in

Saldanha Bay.

4.1.3 Environmental Impact Assessment The Environmental Impact Assessment is based on the results of the three-dimensional

hydrodynamic and water quality numerical modelling study (Chapter 7).

The information for the assessment of impacts related to the brine discharge (and other co-

discharges) was drawn from various scientific publications as well as information sourced from the

Internet and provided by Transnet Projects. The sources consulted are listed in the reference list

(Chapter 10). The method used to assess the potential environmental impacts is described below.

The significance of all potential impacts that would result from the proposed project is determined in

order to assist decision-makers. The significance rating of impacts is considered by decision-makers,

as shown below.

• INSIGNIFICANT: the potential impact is negligible and will not have an influence on the decision regarding the proposed activity.

• VERY LOW: the potential impact is very small and should not have any meaningful influence on the decision regarding the proposed activity.



• LOW: the potential impact may not have any meaningful influence on the decision regarding the proposed activity.

• MEDIUM: the potential impact should influence the decision regarding the proposed activity.

• HIGH: the potential impact will affect a decision regarding the proposed activity.

• VERY HIGH: The proposed activity should only be approved under special circumstances.

The significance of an impact is defined here as a combination of the consequence of the impact

occurring and the probability that the impact will occur. The significance of each identified impact

was thus rated according to the method described below.

1. The Consequence Rating for the impact is determined by adding the score for each of the

three criteria (A-C) listed below:

Rating Definition of Rating Score

A. Extent– the area over which the impact will be experienced 0

Local Confined to the project or study area or part thereof (i.e. site specific) 1

Regional Confined to the region, which may be defined in various ways, e.g. cadastral, catchment, topographic

2

(Inter) national Nationally or beyond 3

B. Intensity– the magnitude of the impact in relation to the sensitivity of the receiving environment None 0 Low Natural and/or social functions and processes are negligibly altered 1 Medium Natural and/or social functions and processes continue albeit in a

modified way 2

High Natural and/or social functions or processes are severely altered 3 C. Duration– the time frame for which the impact will be experienced None 0 Short-term Up to 2 years 1 Medium-term 2 to 15 years 2 Long-term More than 15 years 3

A Consequence Rating is subsequently determined by combining the scores of these three criteria

as follows:

Combined Score (A+B+C) 0 – 2 3 – 4 5 6 7 8 – 9 Consequence Rating Not significant Very low Low Medium High Very high

The probability of the impact occurring was assessed according to the following definitions:

Probability – the likelihood of the impact occurringImprobable < 40% chance of occurring Possible 40% - 70% chance of occurring Probable > 70% - 90% chance of occurring Definite > 90% chance of occurring



2. The overall significance of the impact is determined as a combination of the consequence

and probability ratings as indicated in the table below:

Significance Rating Consequence Probability Insignificant Very Low & Improbable Very Low & Possible Very Low Very Low & Probable Very Low & Definite Low & Improbable Low & Possible Low Low & Probable Low & Definite Medium & Improbable Medium & Possible Medium Medium & Probable Medium & Definite High & Improbable High & Possible High High & Probable High & Definite Very High & Improbable Very High & Possible Very High Very High & Probable Very High & Definite

3. The status of the impact (i.e. whether the effect of the impact will be negative or positive) is

noted.

4. The level of confidence in the assessment of the impact is stated as either high, medium or

low.

Practical mitigation measures that can be implemented effectively to reduce the significance of the

impact are identified and described. The impact is then re-assessed following mitigation, by

repeating Steps 1-5 to demonstrate how the extent, intensity, duration and/or probability changed

after implementation of the proposed mitigation measures.

Mitigation measures are described as either:

• Essential: must be implemented and are non negotiable; or

• Optional: must be shown to have been considered and sound reasons provided if not

implemented.



4.2 Limitations and Assumptions

The following are the assumptions and limitations of the study:

• The study is based on the project description made available to the specialists at the time of the commencement of the studies (plant capacities, discharge locations, constituents, volumes, etc.) and as updated on 22 November 2007, 5 December 2007 and mid-January 2008. The assessment is restricted to only those constituents specified by Transnet as being contained within the effluents from the RO plant. Based on feedback from Transnet at the time of the commencement of the model simulations, the modelling study has explicitly assumed that there is no change of the seawater temperature between the intake and the discharge structures. Thus the only thermal impacts being assessed in the modelling study are the ambient temperature differences that exist between the intake location and discharge location. Subsequently Transnet has indicated that the assumption of no change of the seawater temperature between the intake and the discharge structures is not necessarily correct and that the temperature elevation between the intake and discharge point will ultimately depend on the detail of the system design (i.e. retention time in the brine basin, hydraulic design, energy recovery device specifics, etc.). Should there be a significant temperature rise in the seawater on transit from the intake to the discharge location, then there is a potential for the thermal impacts associated with the discharge to be underestimated. Transnet (through their contractor) has indicated that a typical value for such a temperature increase is a maximum of one degree Celsius. Should the temperature elevation of the seawater/brine between intake and discharge be restricted to one degree Celsius as reported above, the magnitude of the associated changes in plume dimensions9 are expected to be sufficiently limited not to change the significance of the assessed thermal impacts of the discharge and consequently the conclusions of this study.

Based on information available at the time of undertaking the modelling studies, if utilised, the residual oxdising biocide (NaOCl) concentration in the effluent stream is assumed to be 0.1 mg.ℓ-1 . Furthermore a continuous discharge has been assumed, however subsequent communications has indicated that if a non-oxidising biocide is utilised it will be applied in shock doses. As the modelling study did not explicitly simulate discrete shock treatment processes, to ensure the validity of the modelling study as well as a conservative approach to assessment, it is assumed that the residual oxidising biocide concentrations will be managed such that a residual biocide concentration of 0.1 mg.ℓ-1 is not exceeded at the point of discharge at any stage of the treatment process (i.e. at any stage during the shock treatment process). Furthermore, it has been stated that non-oxidising biocides will be utilised in a “shock treatment” rather than continuous dosing mode. Similarly, to ensure the validity of the modelling study as well as a conservative approach to assessment, it is assumed that the

9 A temperature elevation between the intake and discharge point due to the system design will result in

increased mixing of the effluent with ambient waters in the near-field due to the increase in the buoyancy of the effluent, resulting in greater dispersion of the effluent plume. However the “thermal load” of the effluent will be greater, resulting in potentially more extensive thermal plumes and an associated increase in potential thermal impacts. It is thus not clear which of these influences would be greatest and whether thermal impacts would be more extensive or not. For the limited increase in temperature indicated, there is reasonable confidence that any changes occurring would not invalidate the conclusions of this study.



residual concentrations of the preferred biocide10, i.e. the non-oxidising biocide (DBNPA), will be managed such that residual non-oxidising biocide concentrations will not exceed a range of between 1.15 mg.ℓ-1 and 2.475 mg.ℓ-1 (see Section 8.1.3.4) at the point of discharge at any stage of the treatment process (i.e. at any stage during the shock treatment process.

• The assessment is limited to the project description as supplied to the specialists by Transnet at the time that this assessment was undertaken and the issues that could be identified using the detail provided in the project description. Specifically, a number of aspects of the engineering design had not yet been finalised at the time of the study (e.g. exact number and design of beach wells, detailed description of construction activities, etc), making confident and comprehensive assessments of potential marine impacts related to these activities difficult. Where such detail is missing from the project description a generic, but conservative, approach to the assessment (i.e. precautionary approach) has been undertaken.

• The three dimensional modelling study comprises a far-field model and does not resolve detailed near-field features of the discharge plumes. For this reason we have assumed that the resolution in the modelling is inadequate to provide detail within an approximate 50 m radius of the discharge point. Consequently we have assumed that the water quality guidelines may regularly be exceeded within a 50 m radius of the discharge point, despite the fact that this will not necessarily be the case. This constitutes a conservative approach.

• Unlike near-field models such as CORMIX, the far-field modelling undertaken here incorporates the effects of the accumulation / re-circulation of effluent in the vicinity of the discharge and the consequent reduction in the dilution/dispersion of the effluents in the discharge plume. The far-field modelling undertaken here also provides a much more realistic simulation of environmental conditions in the marine environment than is typically possible for near-field models. The implementation of near-field models requires a detailed rather than a conceptual description of the discharge structures (i.e. alternatives such as an open channel of single/multi-port diffuser outfall) that is not yet available for this study. As it is intended that this study inform which of the above discharge locations and structures is environmentally acceptable or preferable, the modelling study has been based on broad conceptual discharge designs that parameterise the near-field behaviour of the plume. This combination of conceptual designs and assumed associated near-field effluent behaviour is deemed adequate for the purposes of this study that is intended to assess primarily the potential far-field impacts in the receiving environment. The performance of the discharge infrastructure when designed and constructed will need to be such that it is consistent with (or better than) the near-field behaviour of the brine effluent assumed in this study. In this study we have assumed a relatively conservative near-field behaviour (i.e. effluent remains in the bottom one third to half of the water column), providing some leeway in engineering design possibilities should they be required to improve the near-field dilution of effluents.

10 The RO membranes proposed for use comprise the most recent and efficient membrane technology,

however this implies the use of membrane material that is affected by the sodium hypochlorite (the oxidising biocide) and this results in deterioration of the membranes. It is for this reason that the use of DBNPA is proposed as this does not affect the membranes in the same way.



• The three dimensional modelling study is based on existing measurements (i.e. design and associated environmental measurement for Saldanha Bay) and no new measurement were obtained specifically for this study. The existing observations are deemed to be adequate for the purpose of using the model to predict discharge plume behaviours.

• Other than the detailed modelling component of the study, the ecological assessment is limited to a “desktop” approach and thus relies on existing information only. No new data or measurements (physical or biological) have been obtained as part of this study.

• While a literature research and review are usually adequate for a study of this nature, a lack of baseline information on sandy beach macrofauna was identified. Consequently, in the event of beach wells and/or surf-zone discharges being considered in the potential engineering design, existing information on this habitat is inadequate to confidently base an assessment of the impacts. More appropriate in this case would be to design and implement a quantitative baseline survey of beach macrofauna in the area on which to base the assessment.

• The assessment is based on the current Port configuration, and therefore does not consider future port expansions not clearly covered under the current specifications of the Saldanha Bay Iron-ore Expansion Project (i.e. Phase 2). If deemed necessary, the scenarios comprising discharges into Big Bay could be simulated with the “maximum” proposed changes in port layout. However, the “footprint” of the effluent discharged from the RO Plant under the various discharge scenarios is sufficiently limited in extent to not be influenced sufficiently by the changes in port layout to invalidate the conclusions of this study (see Section 2.4 for more detail).

• Some important conclusions and associated assessments and recommendations made in the EIA section of this study are based on results from the detailed three dimensional modelling study. The predictions of these models, whilst considered to be robust in terms of the major discharge constituent (salinity)11, need to be validated by field observations and subsequent monitoring. If field observations and monitoring, however, fail to mirror predicted results, the forecasted impacts will need to be re-assessed.

• Potential changes in the marine environment such as sea level rise and/or increases in the severity and frequency of storms related to climate change are not explicitly considered here. Such scenarios are difficult to assess due to the uncertainties surrounding climate change. Should evidence or more certain predictions of such changes become available, Transnet should re-assess their development and management plans to include the impacts of these anticipated macroscale changes. However, it is not expected that these climate changes will affect the effluent plume behaviour to the extent that the conclusions of this study will be altered. (These changes may however significantly affect the existing beneficial or designated uses in the bay.) The most significant change that may occur due to long-term climate change is a change in the beach configuration at Site 1 (and most probably to an insignificant extent at

11 Should the field observations and monitoring fail to mirror the model predictions for the various

constituents of the effluent stream, the concentration of constituents of concern in the effluent stream can be reduced by the various mitigation measures proposed. The only exception is the salinity of the brine discharge that is inherent to the RO Plant operation and that can only be manipulated by changing RO Plant efficiencies and then not to any significantly degree. It is for this reason that it is important that the reported model predictions of salinity “footprints” in the marine environment, are considered to be robust.



Site 2) that may need to be considered when designing intake beach wells. It is expected that the beach in the vicinity of Site 1 will continue to accrete as it has done since the construction of the causeway. Should Phase 2 of the Saldanha Bay Iron-ore Expansion Project proceed, it is predicted that the present rate of accretion of the beach at Site 1 will be slower than would be the case if not further development take place (Smith et al. 2007).



5 DESCRIPTION OF THE AFFECTED ENVIRONMENT

Here a detailed description is given of the physical, chemical and biological environment of Saldanha Bay.

5.1 Physical Environment

5.1.1 General The Saldanha Bay-Langebaan system can be divided into Outer Bay, Saldanha Bay (comprising Big Bay and Small Bay) and Langebaan Lagoon (Figure 5.2a and b). The boundary between Big Bay and Small Bay is the iron-ore causeway built in 1974/75 which impacts significantly on the water circulation in Saldanha Bay. Most of the commercial activities in Saldanha Bay are concentrated in Small Bay or just outside of Small Bay while Langebaan Lagoon remains largely pristine. There are no significant river inputs into either Saldanha Bay or Langebaan Lagoon, consequently the Saldanha Bay-Langebaan Lagoon system is marine with its waters originating in the shelf waters of the adjacent Benguela upwelling system (Shannon and Stander, 1977). In winter, however, there is a small seepage of fresh water due to rain (Day, 1981).

The overall surface area of the Saldanha Bay-Langebaan system is estimated to be 9.61×107 m

2. Of

this surface area, Small Bay comprises 1.41×107 m

2, Big Bay 4.31×10

7 m

2 and Langebaan Lagoon

3.89×107 m

2. The mid-tide volume of the whole system is 7,34×10

8 m

3. Of this total volume, Small

Bay contributes 1.28 ×108 m

3, Big Bay 5.17×10

8 m

3 and Langebaan Lagoon 8.91×10

7 m

3 (Weeks et

al., 1990).

5.1.2 Climate and Winds Saldanha Bay and Langebaan Lagoon are situated on the Cape West Coast, approximately 100 km

north of Cape Town. The climate of this area is mild to cool and is strongly influenced by the cold

Benguela Current that moves up the west coast of Southern Africa. Temperatures are mostly less

than 20°C and rarely exceed 30°C (CSIR, 1996). The area has a semi-arid Mediterranean climate

with an average annual rainfall of about 300 mm. Most of the rainfall occurs in winter with summers

generally being dry. Coastal fogs caused by the interaction between cold marine air (the result of the

Benguela Current) and the warmer land mass are common, particularly in autumn. There is a strong seasonality in the winds over Saldanha Bay, reflecting the changes in the synoptic weather patterns

prevailing at different times during the year. Southerly winds pre-dominate in this region for most of

the year, modulated by short periods of calm conditions or north-westerly winds which are associated

with the propagation of coastal lows southwards along the west coast of southern Africa. Only in the

mid-winter months do north to north-westerly winds predominate.



Figure 5.1: Wind roses of the winds measured at Port Control in Saldanha Bay (see inset).

Port Control



Wind data from Port Control in Saldanha Bay (Figure 5.1) indicate that in summer the winds are

predominantly southerly with significant southwesterly and to a lesser extent southeasterly wind

components. In autumn the winds are predominantly southerly with the development of a

northwesterly wind component as the season progresses. The regular passage of cold fronts in

winter results in predominantly northwesterly winds with the occurrence of significant southwesterly

and southeasterly wind components. The spring wind regime is similar to the summer wind regime

but with increased southeasterly wind components.

The winds along the West Coast have a significant diurnal component (Jury and Guastella, 1987)

that is clearly observable in the wind records for Saldanha Bay. The wind speed typically reaches a

maximum in the late afternoon when the sea breeze cycle is at its maximum. These diurnal changes

in the winds are expected to impact significantly on the heat fluxes at the sea surface over a 24 hour

period.

The interannual variability in the winds along the West Coast has been well documented

(Taunton-Clarke, 1990; Taunton-Clarke and Shannon, 1988; Shannon et al., 1992).

5.1.3 Tides The tides along the west coast of southern Africa, including Saldanha Bay, are semi-diurnal (two high and two low tides per tidal day). The tidal characteristics for Saldanha Bay (Table 5.1) are typical of a micro tidal regime and indicate an approximate 2 m tidal range during spring tides.

Table 5.1: Tidal characteristics for Saldanha Bay.

Tidal Characteristic Tidal level relative to Chart Datum

(m) Highest Astronomical Tide 2.03 Mean High Water Springs 1.75 Mean High Water Neaps 1.27 Mean Level 0.99 Mean Low Water Neaps 0.70 Mean Low Water Springs 0.24 Lowest Astronomical Tide 0



5.1.4 Waves The wave conditions inside the bay are sheltered compared to those outside the bay, since all energy

reaching the bay has to pass through the relatively narrow channel between Marcus Island and

Elandspunt. The median significant wave height measured in the entrance to the bay is 1.1 m, while

the greatest occurrence of peak periods lies in the 10 to 12 s range. The seasonal wave height

exceedances measured at a buoy near the entrance to Small Bay are summarised for the various

seasons in Table 5.2 below.

The wave height outside the bay can be deduced from the wave data recorded at Slangkop, west of

the Cape Peninsula, where the median significant wave height is 2.3 m. Figure 5.2 shows the wave

rose measured at Slangkop. The most frequent direction of wave approach is from the south-west.

In addition to waves originating from offshore and refracted into the bay, small wind-waves (up to

some 1 m in height) are generated by strong winds within Saldanha Bay.

Simulated wave conditions for large offshore swell conditions and strong wind conditions are plotted

in Figures 5.3a and 5.3b, respectively.

Table 5.2: Wave height exceedances measured at a wave buoy near the entrance to Small Bay.

Significant Wave Height (Hmo) Exceeded (m)

1% 5% 10% 25% 50%

Summer 2.38 1.85 1.64 1.35 1.07

Autumn 2.77 2.15 1.84 1.41 1.09

Winter 3.57 2.63 2.27 1.79 1.35

Spring 2.82 2.18 1.89 1.50 1.14

All Seasons 2.96 2.25 1.94 1.51 1.15

Measurements of long wave energy in Saldanha Bay indicate significant energy in the period range

of 30 s to 200 s.



Figure 5.2: Wave height measured at the Slangkop Directional Waverider, located off Kommetjie on the Cape South-west coast (34°12'14.40"S, 18°17'12.01"E, 70 m depth) for the period 2001 to 2007.



Figure 5.3a: Simulated wave conditions in Saldanha Bay for large offshore swell conditions and low winds (Offshore Hmo = 3.6 m , Tp=13 s, Direction = SW’ly, Local wind = 3 m/s S’ly).

Figure 5.3b: Simulated wave conditions in Saldanha Bay for moderate offshore swell conditions and very strong winds (Offshore Hmo = 1.3 m , Tp=11 s, Direction = SW’ly, Local wind = 20 m/s S’ly).



Figure 5.4a: Flood tide surface and bottom currents in Saldanha Bay during spring tide and under relatively calm conditions.

Figure 5.4b: Ebb tide surface and bottom currents in Saldanha Bay during spring tide and under relatively calm conditions.



Figure 5.5a: Schematic of the wind driven and tidal currents in Saldanha Bay under S wind conditions.

Figure 5.5b: Schematic of the wind driven and tidal currents in Saldanha Bay under NW wind conditions.



5.1.5 Currents The currents in the bay are predominantly forced by the wind and the tide, the relative importance of

the two processes changing with depth and location in the bay. In general, wind is the dominant

physical forcing mechanism determining the surface layer current speed and direction in both Small

and Big Bay (van Ballegooyen et al., 2002). Tidal forcing (Figure 5.4a and b) is stronger at depth, in

the vicinity of the mouth of Saldanha Bay (Shannon and Stander, 1977) and with increasing proximity

to Langebaan Lagoon (Weeks et al., 1991a). In the surf-zone, wave-driven currents are expected to

dominate.

Although residual flows associated with the tides occur in the bay, the greatest exchange between

Saldanha Bay and the shelf is a consequence of synoptic weather events occurring on time scales of

3 to 10 days. South-southeasterly wind events are reported to result in a general surface outflow and

a subsurface inflow of cold bottom water (Spolander, 1996; Monteiro and Largier, 1999), while

indications are that northwesterly wind events lead to the inflow of surface waters in the northern

region of the mouth of Saldanha Bay (Figure 5.5a and b). Thus the surface, mid-water and bottom

currents often are observed to be flowing in different, and at times, opposite directions. While this is

expected in a highly stratified water column, observations of three-dimensional flow structure are not

restricted to strongly stratified conditions. Weeks et al. (1991b) recorded an event in late winter

(August 1990) where the flow was strongly three-dimensional under well-mixed conditions.

During periods of slack winds, tidal currents dominate and are the sole mechanism for flushing the

bay. The tidal currents are generally weak, however strong tidal flows are observed at the entrance

to the lagoon, particularly during spring tides. During tidal exchange, it is estimated that

approximately half of the lagoon water passes through the Lagoon entrance channels into Saldanha

Bay (Shannon and Stander, 1977) and velocities of up to 1.0 m.s-1 are observed in the two channels

connecting Big Bay and Langebaan Lagoon (Krug, 1999).

5.1.6 Water Column Stratification The water column structure in Saldanha Bay is seasonal, varying from a strongly thermally stratified

water column for most of the year (August to May) to well-mixed conditions during the mid-winter

months (June to July). For most of the year, strong stratification is maintained by atmospheric heat

fluxes into the surface waters and the inflow of cold bottom waters from upwelling on the adjacent

open shelf, with the local winds acting to vertically mix the water column and break down the

thermocline (Monteiro and Largier, 1999). During the mid-winter months the water column within the

bay is largely well-mixed both due to reduced heat fluxes into surface waters and the reduced

occurrence of south to south-easterly winds that drive the upwelling over the adjacent shelf which in

turn drives cold bottom waters into the bay.



Figure 5.6a: Surface and bottom water temperature and flow in Saldanha Bay under strong SE wind conditions (i.e. during the “active” upwelling phase).

Figure 5.6b: Surface and bottom water temperature and flow in Saldanha Bay after a strong SE wind event (i.e. relaxation phase of upwelling).



The above processes control the thermocline dynamics and vertical mixing of the water column

which, together with wind- and tidally-driven currents, ultimately determine the behaviour of

biogeochemical parameters and pollutants within the bay.

Figure 5.7: Vertical temperature structure of the water column along the “profile line” or cross-section indicated in Figure 5.6b. The upper panel shows the “active” upwelling phase while the lower panel shows the water column structure during the relaxation phase of the upwelling cycle.

The variability in the water column stratification is predominantly synoptic and responds strongly to

wind-forcing which has a periodicity of 6 to 10 days in this region (Nelson and Hutchings, 1983). A

typical sequence of events in the bay is as follows.

The dominant south to southeasterly winds over the bay and adjacent shelf result in upwelling over

the adjacent shelf which sets up a baroclinic pressure gradient which drives cold bottom water into

the bay on a time scale of one or two days (Figure 5.6a and 5.7 – upper panel). This constitutes a

strong buoyancy input which acts to strengthen the thermocline within the bay. However, the same

strong south to southeasterly winds results in vigorous vertical mixing of the water column within the

bay as well as a heat loss from the surface waters. These processes act to reduce the water column

stratification. Initially the mixing processes dominate and the water column becomes well-mixed,

particularly in the shallower regions. However, after a period of approximately 24 hours or more the

effects of the input of cold bottom waters predominate and the thermocline starts to strengthen. With

the passing of the coastal low, the winds typically moderate resulting in either calm conditions or



weaker northwesterly winds. The resulting reduction in the vertical mixing of the waters in the bay

results in the rapid appearance of a strongly stratified water column within the bay. This is followed

by the retreat of the cold bottom waters from the bay under NW or calm wind conditions (Figure 5.5b

and 5.7 – lower panel). This sequence of events is important as not only does it explain the high

degree of natural temperature variability observed in the bay, but it also describes the physical

processes that serve to reduce the thermal impacts associated with a thermal discharge into the bay.

Typically any additional anthropogenic heat input into the bay would be removed by mixing of these

heated waters with the cold bottom waters or by heat loss to the atmosphere.

5.1.7 Seawater Temperature The natural seawater temperature fluctuations in Saldanha Bay are substantial and typically occur on

four time scales, namely diurnal, synoptic, seasonal and interannual (Figure 5.8).

Figure 5.8: Simulated surface and bottom water temperatures at North Buoy in Small Bay (see inset), showing the various temporal scales and magnitudes of seawater temperature variability in Saldanha Bay.

Thermistor chain data in Small Bay (Monteiro and Largier, 1999; CSIR, 1995) indicate that the

diurnal temperature changes are greatest in summer when the surface waters experience diurnal

surface bottom

surface bottom



temperature changes of typically 0,5°C to 1°C but up to 2°C on occasion. Much larger diurnal

temperature fluctuations may occur at a specific depth in the deeper waters due to increased or

decreased vertical mixing mainly associated with diurnal changes in local wind velocity. In winter,

when heat fluxes into the surface waters are smaller and the water column is largely unstratified, the

diurnal temperature fluctuations are substantially reduced at all depths in the water column.

Changes in the synoptic weather events lead to substantial variability in water column temperature

both within the bay and over the adjacent shelf region (CSIR, 1976; CSIR, 1995; Monteiro and

Largier, 1999; Shannon, 1985). These temperature changes occur as a result of advection of warm

surface or cold bottom waters into Saldanha Bay from the adjacent shelf region as well as vertical

mixing of the water column by local winds in the bay. Investigations by Spolander (1996) and

Monteiro and Largier (1999) indicate that changes in atmospheric heat fluxes at the surface are

expected only to play a secondary role in the synoptic seawater temperature variability.

The thermistor chain data in Small Bay (CSIR, 1995; Monteiro and Largier, 1999) show that the

temperature at any particular depth may change during a synoptic cycle by as much as 6°C to 8°C in

summer and 1°C to 2°C in winter. This is mostly due to changes in the vertical mixing of the water

column due to local winds, however, advection of cold bottom waters into Small Bay also play a

significant role in these temperature fluctuations during the upwelling season (Monteiro and Largier,

1999). There is an indication in both thermistor chain data (Sea Fisheries Research Institute) and

CTD profiles (CSIR, 1976) that there is flow of warmer surface waters into the Saldanha Bay under

northwesterly wind conditions.

Long-term (daily) sea surface temperature observations within Small Bay (Greenwood and

Taunton-Clark, 1992) indicate that the mean seasonal change in sea surface temperature is about

6°C. The magnitude of seasonal changes in sea surface temperature is expected to be highest in

Langebaan Lagoon and much smaller near the more exposed mouth of Saldanha Bay, however, we

are not aware of any temperature time-series of sufficient length to substantiate the above statement.

The seasonal changes in water temperature of the deeper waters in the bay (approximately 2.5 °C)

are substantially less than those observed in the surface waters of the bay.

Daily sea surface temperature observations in Small Bay indicate that the interannual sea surface

temperature variability typically has a magnitude of between 1°C and 2°C (Greenwood and

Taunton-Clark, 1994). These longer period changes in temperature are most likely due to persistent

changes in the local synoptic weather conditions. Monteiro and Brundrit (1990) have documented

four major episodic inflows of higher temperature and higher salinity oceanic waters into Big Bay that

occurred between 1974 and 1979, however, the temperature signal associated with these episodic

events is largely masked in the surface waters by seasonal temperature variations due to changes in

atmospheric heat input and vertical mixing of the water column.



In summary, the temperature of the surface waters in Saldanha Bay are determined by atmospheric

heat fluxes at the sea surface, entrainment of cooler subsurface waters into the surface layers and

horizontal exchange of the surface waters with the adjacent lagoon and shelf waters. The

temperature of the bottom water is predominantly determined by upwelling inflows and vertical mixing

of the warmer surface waters into the deeper waters of the bay under strong local wind conditions.

5.1.8 Salinity The salinity of the water in Saldanha Bay has been monitored at frequent intervals in conjunction with

temperature and shows little variation over time. Salinities of the inshore waters along the west coast

typically vary between 34.6-34.9 psu and the salinity values recorded for Saldanha Bay usually fall

within this range. Atkinson et al. (2006) report salinities from various sources that range between

34.7 and 35.2 psu and average approximately 34.9 psu. During summer months wind-driven coastal

upwelling bring cooler less saline water into Saldanha Bay. Consequently the salinity within the bay

usually is slightly lower in summer than in winter when the upwelling front breaks down and warmer,

more saline surface waters enter the bay.

5.1.9 Water Quality Oxygen

The primary sources of oxygen in the marine environment are atmospheric oxygen, which enters the

system via gaseous exchange across the air-sea surface interface and in situ production via

photosynthesis of algae and other aquatic plants. Dissolved Oxygen (DO) is measured as a

concentration (mg. ℓ-1) or as a percent saturation (%). Of critical importance to marine organisms are

the fate and behaviour of dissolved oxygen and the factors affecting fluctuations in DO levels. The

principal anthropogenic activity resulting in changes in DO concentrations in the marine environment

is the addition of organic matter. In the Saldanha Bay system, Small Bay experiences a fairly regular

oxygen deficit during the late summer and winter months, whilst Big Bay experiences less frequent

and lower magnitude oxygen deficits (Atkinson et al., 2006). Monteiro et al. (1990) attributed the

oxygen deficit in Small Bay largely to anthropogenic causes, namely reduced flushing rates (due to

the causeway and ore jetty construction) and discharges of organic rich effluents from fish processing

factories. There is evidence of anoxia in localized areas of Small Bay (e.g. under the mussel rafts,

within the yacht basin) that is caused by excessive organic inputs (Stenton-Dozey et al. 2001).



Turbidity The water of Saldanha Bay is fairly turbid, the turbidity comprising both organic and inorganic

particulates. During active upwelling it is expected that the turbidity of particular the bottom waters

will decrease, however, under strong wind conditions both wind and wave action result in significant

water column turbidity, the light coloured sediments resulting in significant discolouration of the

waters within particularly Big Bay. The waters of Langebaan Lagoon, in contrast, are typically very

clear and of low turbidity.

Dissolved trace metals The Mussel Watch Programme regularly records concentrations of Cadmium, Copper, Lead, Zinc,

Iron and Manganese present in the flesh of mussels. The data collected in Small Bay as part this

programme indicate that there do not appear to be significant increases in concentrations of heavy

metals in the flesh of mussels in Saldanha Bay. Consequently, the quality of the water can be

considered suitable for mariculture purposes and heavy metal accumulation is currently not a

concern (Atkinson et al., 2006).

Microbial Contamination

Pathogenic microorganisms, which are primarily introduced into coastal waters by faecal pollution,

pose a risk to both water users and mariculture ventures. Regular monitoring of microbiological

indicators within Saldanha Bay, initiated by the Saldanha Bay Water Quality Forum Trust, began in

1999 (Monteiro et al., 2000), and the data set covering the six year period from 1999 to 2005 was

summarized by Atkinson et al. (2006). In most areas of Small Bay the target limits for mariculture

and/or recreational uses set by the South African Water Quality Guidelines (DWAF, 1995) were

exceeded. Within Big Bay the faecal coliform counts were much lower, falling within the

recommended limits for recreation at all sites. At sites close to Langebaan town, however, bacterial

counts were in excess of target values for mariculture purposes (Atkinson et al., 2006).



5.2 Biological Environment

Saldanha Bay and Langebaan Lagoon fall within the Namaqua biogeographic province that extends

from Cape Point to Lüderitz within the southern Benguela upwelling region (Emanuel et al., 1992).

The bay and the lagoon together form one of the few sheltered habitats along the South African West

Coast, with graded changes in wave action and substratum between Saldanha Bay and Langebaan

Lagoon. The lagoon, which is shallow (2-6 m deep), 16 km long and 2-3 km wide, is fully marine with

a strong tidal exchange (Shannon and Stander, 1977). It has extensive intertidal sandflats and salt

marshes (Day, 1959). Marine ecosystems within the bay comprise a range of habitats each

supporting a characteristic biological community. Habitats in the Saldanha Bay system include:

• Sandy intertidal and subtidal substrates,

• Macrophyte beds,

• Intertidal rocky shores and subtidal reefs,

• Salt marshes,

• Unvegetated sand flats,

• Sea grass beds, and

• The water body

The biological communities in each of these habitats are described briefly below, with the main focus

on potentially sensitive communities, which may be affected by the proposed project.

5.2.1 Sandy Substrate Habitats and Biota The benthic biota of soft bottom substrates constitutes invertebrates that live on, or burrow within, the

sediments, and are generally divided into macrofauna (animals >1 mm) and meiofauna (<1 mm).

Intertidal Sandy Beaches Sandy beaches are one of the most dynamic coastal environments. The composition of their faunal

communities is largely dependent on the interaction of wave energy, beach slope and sand particle

size, which is called beach morphodynamics. Three morphodynamic beach types are described:

dissipative, reflective and intermediate beaches (McLachlan et al., 1993). Generally, dissipative

beaches are relatively wide and flat with fine sands and high wave energy. Waves start to break far

from the shore in a series of spilling breakers that ‘dissipate’ their energy along a broad surf zone.

This generates slow swashes with long periods, resulting in less turbulent conditions on the gently

sloping beach face. These beaches usually harbour the richest intertidal faunal communities.

Reflective beaches have low wave energy, and are coarse grained (>500 µm sand) with narrow and

steep intertidal beach faces. The relative absence of a surf-zone causes the waves to break directly

on the shore causing a high turnover of sand. The result is depauperate faunal communities.

Intermediate beach conditions exist between these extremes and have a very variable species

composition (McLachlan et al., 1993; Jaramillo et al., 1995; Soares, 2003). This variability is mainly



attributable to the amount and quality of food available. Beaches with a high input of e.g. kelp wrack

have a rich and diverse drift-line fauna, which is sparse or absent on beaches lacking a drift-line

(Branch and Griffiths, 1988; Field and Griffiths, 1991).

There is a noticeable scarcity of published information on the intertidal beach biota of Saldanha Bay

as work on the West Coast of South Africa has primarily focussed on ‘open coast’ beaches (e.g.

Soares 2003). Day (1959) gives an account of four sandy shores from the Saldanha Bay system,

comparing the more wave-exposed beaches of Saldanha Bay with the sheltered beaches in

Langebaan Lagoon. He describes an increase in species richness and a significant change in

species composition with increasing shelter. His work, however, was carried out prior to the

construction of the causeway and ore jetty, and thus prior to the changes in wave and current pattern

in the bay. Over the past few years, Zoology students from the University of Cape Town have

sampled beaches in the bay as part of their course work. These data are, however, unpublished and

the quality of the data is questionable, but a short summary of the results is none the less provided

here (unpublished UCT student data from 1995 and 1996, provided by Prof. C. Griffith). Most of the

beaches sampled are in Langebaan Lagoon or at the head of the lagoon; the only beach sampled in

Saldanha Bay is at Lynch Point. An ‘open coast’ beach at 16 Mile Beach was also surveyed. The

results confirm the dramatic change in species richness and composition between the exposed

Saldanha Bay beach and the sheltered lagoon beaches. At Lynch Point, the fauna is sparse, and

includes the semi-terrestrial isopod Tylos granulatus and the talitrid amphipod Talorchestia spp. in

the supralittoral zone (above the high water spring mark (HWS)), and the amphipods Pontogeloides

latipes, Eurydice longicornis, the polychaetes Glycera tridactyla and Scololepis squamata in the

midlittoral zone. The mysid Gastrosaccus psammodytes was found at, and below, the low tide level.

A similar faunal composition was recorded from 16 Mile Beach (unpublished UCT student data from

1995 and 1996). The macrofaunal species encountered are generally ubiquitous to the West Coast

(Day 1959; Soares 2003). In contrast, the extremely sheltered intertidal flats in Langebaan Lagoon

harboured >30 species (Day (1959) recorded 55 species), many of which are either South Coast

species known to occur on the West Coast only in Langebaan Lagoon, typical estuarine species, or

species normally found in pools and crevices on exposed rocky shores (Day 1959). Noteworthy is

that many of the typical West Coast beach species (e.g. Tylos, Talorchestia, Eurydice) are not found

in the lagoon.

No data were found on sandy beach biota north of Lynch Point or from Small Bay. Due to the

general lack of knowledge of sandy shore fauna in Saldanha Bay it is strongly recommended that a

baseline survey of sandy intertidal habitats in the vicinity of the proposed RO Plant be undertaken

before any construction work commences, i.e. at Site 1 and at Site 2 should it be anticipated that

construction activities with disrupt the sandy shore fauna. The purpose of such surveys are both to

provide a baseline to assess construction impacts as well as the data and necessary understanding

to inform future impact assessments.



Subtidal Sandy Habitats The structure and composition of benthic soft bottom communities is primarily a function of water

depth and sediment grain size, but other factors such as current velocity, organic content, and food

abundance also play a role (Snelgrove and Butman, 1994; Flach and Thomsen, 1998; Ellingsen,

2002). Benthic infauna is the preferred group for environmental monitoring studies, and first

accounts of benthic assemblages in Saldanha Bay date back to the 1940s. Major developments

have occurred in the bay since then and changes in benthic community structure as a result of

anthropogenic impacts have been reported by numerous authors (Christie and Moldan, 1977;

Moldan, 1978; Jackson and McGibbon, 1991, amongst others). For example, in Small Bay there has

been a shift from communities dominated by suspension-feeders to communities characterized by

deposit-feeders. More specifically, the sea pen Virgularia schultzei, a suspension feeder, was

historically widespread in the bay, but was not been recorded at all after 1989. Although it re-

appeared in Big Bay in 2004, it is still absent in Small Bay. On the other hand, the deposit-feeding

polychaete Polydora sp. has undergone a dramatic increase over the last decades, especially in

Small Bay (Jackson and McGibbon, 1991). This shift in community composition has been attributed

to changes in water circulation patterns in the Bay, as well as organic pollution from fish factories and

mussel farming in Small Bay.

Table 5.3: Dominant species/taxa in Saldanha Bay as reported from benthic macrofauna studies conducted in 1975, 1999 and 2004 (adapted from Atkinson et al., 2006).

Taxa Common name Scientific name

Echiuroidea Tongue worm Ochaetostoma capense Pennatulacea Sea pen Virgularia schultzei Echinodermata Sea cucumber Holothuroidea Brittle star Ophiuroidea Pelecypoda Tellinid mussel Tellina gilchristi Tellinid mussel Macoma crawfordii Black mussel Choromytilus meridionalis Gastropoda Plough snail Bullia digitalis Crustacea Mud prawn Upogebia capensis Sand prawn Callianassa kraussi Three-legged crab Thaumastoplax spiralis Crown crab Hymenosoma orbiculare Amphipods Various species Isopods Various species Polychaeta Segmented worms Various species

The most recent study on benthic macrofauna was commissioned by the Saldanha Bay Water

Quality Forum in 2004 (Anchor Environmental Consultants, 2004), and is summarized by Atkinson et



al. (2006) and compared to previous studies. Table 5.3 lists the dominant benthic macrofauna

species in Saldanha Bay.

The mud prawn Upogebia capensis is one of the most dominant species in the bay, particularly so in

Small Bay (Anchor Environmental Consultants, 2004). Other important species in Small Bay include

the polychaete Polydora sp., the amphipod Ampelisca spinimana, the tongue worm Ochaetostoma

capense and the crab Thaumastoplax spiralis, which lives commensally in the tube of the tongue

worm (Day, 1974). Aside from the mud prawn and the tongue worm, Big Bay was dominated by two

amphipod species (Ampelisca spinimana and Urothoe grimaldi), and the polychaete Orbinia

angrapequensis. In 2004, the sea pen Virgularia schultzei made a first reappearance in the Bay after

being absent for >10 years, but its occurrence was restricted to Big Bay while it was still absent from

Small Bay. Upogebia capensis, Ochaetostoma capense and Callianassa kraussi contributed the most

to the overall biomass in Small Bay and Big Bay, respectively. In the turbulent surf zone, particularly

between 2 – 5 m depth, the faunal diversity is usually lower and primarily includes amphipods and

polychaetes (Christie, 1976).

Dominant species in unvegetated sandflats in Langebaan Lagoon are the sand prawn Callianassa

kraussi, U. capensis, Ampelisca palmata (amphipod), Notomastus latericeus (polychaete) and

Cirolana hirtipes (isopod). In the mid-1990s, the alien invasive Mediterranean mussel Mytilus

galloprovincialis began establishing dense intertidal beds on the centre sandbanks of Langebaan

Lagoon (Hanekom and Nel, 2002). This invader had previously been restricted to rocky shores along

the South African coastline (Griffiths et al., 1992). The alien mussel beds significantly altered natural

community structure by the creation of a new habitat, which promoted the establishment of rocky-

shore hard-substrate species. Furthermore, the mussel beds excluded many sediment-dwellers

through smothering, and by denying burrowing species access to surface waters (Robinson et al.,

2002). Interestingly, after supporting an estimated biomass of nearly 8 t in 1998, the population had

died-off completely by mid-2001, with only empty shells and anoxic sand remaining (Robinson et al.,

2007a). The reason for the die-off is, however, unknown. In an effort to prevent the re-settlement of

M. galloprovincialis in this area, South African National Parks began removing all dead mussel shells

from the centre banks in late 2001, as these shells offer a suitable settlement substrate for mussel

larvae. The removal aided in the recovery of the previously invaded areas. However, 5 months after

clearance more than 50% of the species recorded in non-invaded areas were still absent from

cleared areas, including the important bioturbator Callianassa kraussi (Robinson et al., 2007a).

Subtidal macrophyte beds are dominated by the agarophyte alga species Gracilaria gracilis, which

occurs in Small Bay and adjacent to Schaapen Island in the southern portion of Big Bay. The alga is

also characteristic of the subtidal sandy sediments in the Langebaan Lagoon (Schils et al., 2001).

The alga occurs on sandy substrates at 2-10 m depths, and may either be anchored or drifting

(Anderson et al., 1993). Gracilaria forms the basis of a small industry that collects cast material from



the beaches for export to agar processing plants. Another important macroalga in Langebaan

Lagoon is Sargassum incisifolium (Schils et al., 2001).

The southern end of Langebaan Lagoon is dominated by beds of the eelgrass Zostera capensis.

Faunal species that are associated with these sea-grass beds are the snails Assiminea globulus and

Hydrobia sp., the limpet Siphonaria compressa, the polychaetes Ceratonereis erythraeensis and

Perinereis nuntia vallata, an amphipod Paramoera capensis, a crab Cleistostoma edwardsii and the

mudprawn Upogebia africana (Siebert and Branch, 2007). The invertebrate fauna of the mudflats are

an important food source for waders; for example, A. globulus is the major prey item of the curlew

sandpiper (Puttick 1980) and U. africana is important to the kelp gull, grey plover, and common tern

(http://www.environment.gov.za/soer/nsoer/resource/wetland/ langebaan_ris.htm). The inter-tidal salt

marsh vegetation in the lagoon is classified as ‘dwarf succulent shrubland’ and includes the grass

Spartina maritime and the succulent plants, Chenolea diffusa, Sarcocornia perennis, S. pillansii, and

Salicornia meyerana. The algal species growing in the salt marshes (belonging to the Bostrychietum

association) have a large geographical distribution and occur commonly in tropical mangroves and

warm temperate salt marshes. On the West Coast, however, they are restricted almost entirely to

Langebaan Lagoon (Schils et al., 2001). The salt marshes of Langebaan Lagoon are the largest

along South Africa’s shoreline, and being in a relatively pristine condition, they are thus a biologically

valuable area (Schils et al., 2001).

In general, there has been a decline in species diversity but an overall increase in biomass in the bay

from 1999 to 2004, and a corresponding decrease in biomass in Langebaan Lagoon (Atkinson et al.,

2006). It has been suggested that this is related to the increase in particulate organic carbon in the

bay, which serves as an important food source for particulate feeders such as the crustaceans (the

group that accounted for most of the observed increase in biomass).

Similarly, the epifaunal (animals living on the sediment surface) community composition appears to

have undergone dramatic changes after the harbour development (Kruger et al., 2005).

Comparisons of data from the 1960s (prior to the jetty and causeway construction) with those from a

dredge survey in 2001, demonstrated a decline in species number and a shift in species composition.

Polychaetes, in particular, showed a substantial decline in species number. The species that

contributed most to the dissimilarity between the epibenthic communities of the 1960s and the 2001

were the whelk Nassarius speciosus and the crab Hymenosoma orbiculare. Both species had

increased significantly in abundance in 2001. It was suggested that altered wave energy, a shift

towards finer sediment and increased organic matter within Saldanha Bay as a result of harbour

construction, and fish factory and mussel-farm outputs, were responsible for these changes (Kruger

et al., 2005).

Saldanha Bay is a sheltered bay and the benthic faunal and floral species occurring in the bay (e.g.

Upogebia capensis, Callianassa kraussi, Ochaetostoma capense, Nassarius speciosus, Gracilaria



gracilis) are typical inhabitants of shallow sheltered mud or sand banks and are common to the

southern African West Coast (Day, 1974; Branch et al., 1994). Langebaan Lagoon, however, from a

biological perspective is unique in that numerous species more typical of warmer waters occur here

and certain species normally restricted to estuarine conditions are also present, despite the system

being fully marine (Day 1959; Schils et al., 2001).

5.2.2 Rocky Habitats and Biota

Intertidal Rocky Shores Despite the known changes that have taken place within the Saldanha Bay system over the last fifty

years, almost no historical data exists on the state of rocky shores in the area. A survey covering a

range of different rocky habitats has only recently been undertaken (Atkinson et al., 2006). This

study showed that, similar to other South African rocky shores (e.g. McQuaid and Branch, 1984),

wave exposure and the type of rock substratum were important determinants of community structure.

The rocky intertidal can be divided into different zones according to height on the shore. Each zone

is distinguishable by its different biological communities, which is largely a result of the different

exposure times to air. The level of wave action is particularly important on the low shore. Generally,

biomass is greater on exposed shores, which are dominated by filter-feeders. Sheltered shores

support lower biomass, and algae form a large portion of this biomass (McQuaid and Branch, 1984;

McQuaid et al., 1985).

The rock topography is also important, and for example boulder beaches in Small Bay were found to

be impoverished in terms of species abundance and biomass compared to more wave exposed sites.

Construction of the iron ore causeway and the Marcus Island causeway altered the wave exposure

zones in the Bay. The causeway increased the extent of sheltered and semi-sheltered zones in

Small Bay, with semi-exposed shores being absent in this area (Luger et al., 1999). Although wave

exposure in Big Bay was altered less dramatically, the extent of sheltered and semi-sheltered wave

exposure areas increased after harbour development (Luger et al., 1999). Although no historical data

prior to the construction of the causeway exist, it has been suggested that the sheltering effect of the

causeway has negatively affected the intertidal communities along the Small Bay shoreline and

changed their compositions (Atkinson et al., 2006).

In terms of zonation, important species in the high shore are the grazers Littorina africana

knysnaensis, Oxystele variegata, the filter feeding barnacle Chthalamus dentatus, and the alga

Porphyra capensis. Mid-shore levels are dominated by C. dentatus, the limpets Siphonaria capensis

and Scutellastra granularis, the carnivorous whelk Burnupena sp., the algae Ulva spp. and

Caulacanthus ustulatus, and the alien mussel Mytilus galloprovincialis. At sheltered sites, the low

shore is characterized by algae such as Ulva spp., Gigartina radula, and crustose algae, and the

faunal component includes Burnupena sp., M. galloprovincialis and the indigenous mussel

Choromytilus meridionalis. At more exposed sites, the low shore is covered primarily by M.



galloprovincialis. All of these species are widespread on western and southern Cape shores

(Emanuel et al., 1992). However, M. galloprovincialis is an alien invasive species that is displacing

the indigenous species Choromytilus meridionalis and Aulacomya ater (Robinson et al., 2007b).

Further, because of greater structural complexity within beds of M. galloprovincialis compared to

those of indigenous mussels, there have been changes to overall community composition in areas

colonised by this species (Robinson et al., 2007b).

A study of the intertidal macroalgal assemblages in Saldanha Bay and Langebaan Lagoon identified

two distinct floral entities on rocky shores: (i) Saldanha Bay (including Small Bay and Big Bay) and (ii)

Langebaan Lagoon (Schils et al., 2001). The transition between the floral entities is located at the

mouth of the Lagoon. The species richness of the bay area is greater than in the lagoon. The

change in algal composition was explained by environmental variables of which wave exposure is the

most significant. In terms of biogeographical affinities of the different algal entities it was shown that

the bay area supports a typical West Coast flora. The algal flora of the lagoon is also dominated by

West Coast species, but is typified by species characteristic of sheltered habitats, and with a number

of species which otherwise only occur on the geographically distant South Coast (east of Cape

Agulhas) (Schils et al., 2001).

Rocky Subtidal Habitats Rocky subtidal reefs are not extensive in Saldanha Bay, but artificial habitats such as harbour

structures and their reinforcements serve as additional settlement substrates. The dominant

organisms on these structures are mussels (M. galloprovincialis and to a lesser extent C. meridionalis

and A. ater), Pyura, and whelks and barnacles with associated macroalgae. Typical kelp species

along the West Coast of South Africa are the kelps Ecklonia maxima and Laminaria pallida

(Stegenga et al., 1997). In Saldanha Bay, however, E. maxima appears to be replaced by L. pallida.

Simons (1977) considers this to be a response to reduced wave exposure within the bay. Individuals

of E. maxima occur as far as the entrance of Langebaan Lagoon but do not penetrate further into the

lagoon, whereas isolated specimens of L. pallida can be found further into the lagoon (Schils et al.,

2001).

5.2.3 Pelagic Communities The pelagic communities are typically divided into plankton (phytoplankton and zooplankton including

ichthyoplankton) and fish, and their main predators, marine mammals (seals, dolphins and whales).

Plankton Saldanha Bay is protected from the high-energy coastline, but remains a highly productive system

owing to its link on its western side to the Benguela upwelling system (Pitcher and Calder, 1998).

Due to the nutrient supply from this upwelling system, phytoplankton concentrations in Saldanha Bay

can attain concentrations of 18 mg Chl a .m-3 with a mean value of 8.62 mg Chl a .m-3 (Pitcher and

Calder, 1998). Highest values typically occur during the upwelling season. Phytoplankton exhibits



short term variability in distribution as it responds to variations in light levels, induced by natural

turbidity, nutrient supply to the surface layers resulting from wind mixing of the water column, and the

presence and location of thermoclines dividing oligotrophic surface layers from cooler, nutrient rich

subsurface water. Through the above processes, high phytoplankton biomasses can occur in

surface waters, be limited to subsurface maxima associated with thermoclines, or be reduced to low

levels as characteristically occurs in winter (Pitcher and Calder, 1998). Phytoplankton production is

estimated at 3.40 g C. m-2 .day-1 (Pitcher and Calder, 1998). This is comparable to estimates for the

adjacent southern Benguela upwelling system (Shannon and Pillar, 1986). The phytoplankton

species assemblage within Saldanha Bay appears to be largely similar to that of the adjacent

continental shelf.

Harmful algal blooms (HABs) are a regular late summer feature in the southern Benguela region

(Pitcher and Calder 2000). The occurrence of on Harmful algal blooms and their dynamics has been

recently summarised by Carter (2008).

Paralytic shellfish poisoning (PSP) due to Alexandrium catenella, and diarrehetic shellfish poising

(DSP) caused primarily by Dinophysis acuminate and D. fortii pose a threat to shellfish mariculture

operations, and mussel harvesting in Saldanha Bay was compromised for the first time in 1994 by

PSP (Pitcher et al., 1994). In subsequent years, both PSP and DSP have become regular problems

for the mariculture operations in the bay (Probyn et al., 2001). The geographical scales of Saldanha

Bay are considered unsuitable for in situ development of HABs (Pitcher et al., 1994). Blooms,

however, can be advected into Saldanha Bay from the adjacent continental shelf waters, but their

development and duration in the bay is restricted by the system of exchange that operates between

the bay and the coastal upwelling system, in that there is a net export of surface waters from the bay

(Probyn et al., 2001). In contrast, blooms of the brown tide organism, Aureococcus anophagefferens,

have been recorded in Saldanha Bay but not on the adjacent continental shelf (Pitcher and Calder,

2000; Probyn et al., 2001). The blooms were mainly limited to the reclamation (oyster) dam in 1997,

but spread throughout the entire system, including Langebaan Lagoon, in 1998 (Probyn et al., 2001),

and led to retarded growth rates in mussels and oysters.

Zooplankton species in Saldanha Bay are composed predominately of species similar to those of the

adjacent continental shelf (Grindley, 1977). The zooplankton species of Langebaan Lagoon,

however, were found to be distinctly different from that of Saldanha Bay, although elements of the

Saldanha Bay communities did penetrate the lagoon to various extents. Surprisingly, the

zooplankton communities at the head of the lagoon were found to be estuarine in character, even

though the system is not an estuary. This was attributed to the wide salinity tolerance range of these

estuarine species, which enables them to withstand the hyper-saline conditions often present at the

head of the lagoon (Grindley, 1977).



Fish Atkinson et al. (2006) report on surveys of fish distributions in Saldanha Bay and Langebaan Lagoon,

which were conducted using a variety of sampling gear. The waters of the Saldanha Bay system

support an abundant and diverse fish fauna with a total of 48 species being recorded. Overall

species richness and abundance was highest in Langebaan Lagoon. There was a trend of

increasing fish diversity and abundance with decreasing wave exposure. This was also reported

previously by Clark (1997). For example, wave exposed beaches yielded <1 fish. per square metre

less, exposed beaches around 2 fish. m-2, and >4 fish. m-2 at the top of the lagoon where waves are

all but absent.

Dominant species in Saldanha Bay and Langebaan Lagoon are harders (Liza richardsonii),

silversides (Antherina breviceps) and gobies (Caffrogobius sp.). Other important fish species in the

bay are the white stumpnose Rhadosargus globiceps, West Coast steenbras Lithognathus aureti,

steentjie Spondyliosoma emarginatum, gurnard Cheilidonichtyes capensis, Cape sole Heteromyctus

capensis, super klipvis Clinus superciliosus, and sand shark Rhinobatos annulatus

(Atkinson et al. 2006; http://www.environment.gov.za/soer/nsoer/resource/wetland/langebaan_ris.htm).

The Saldanha Bay/Langebaan Lagoon complex is an important nursery area for a number of

ecologically important fish species as its sheltered, nutrient rich and sun-warmed waters provide a

refuge from the cold and highly energetic adjacent continental shelf (Atkinson et al., 2006).

Marine Mammals The Cape fur seal Arctocephalus pusillus pusillus no longer breeds on islands in Saldanha Bay, but is

a regular visitor in both the inner and outer bays during all months of the year (Cooper, 1995). Five

whale species have been recorded within Saldanha Bay: Killer whale (Orcinus orca), Humpback

whale (Megaptera novaeangliae) and southern Right whales (Balaena glacialis), along with Minke

(Balaenoptera acutorostrata) and Bryde's (B. edeni) whales in the outer bay between Malgas, Jutten

and Marcus Islands (Cooper, 1995). Dusky dolphins (Lagenorhynchus obscurus) and Heaviside's

dolphin (Cephalorhynchus heavisidii) have been observed along the seaward side of the Marcus

Island causeway (Cooper, 1995).

5.2.4 Birds Saldanha Bay and the associated islands provide important shelter, feeding and breeding habitat for

at least 53 species of seabirds, 11 of which are known to breed on the islands (Atkinson et al., 2006).

The islands of Malgas, Marcus, Jutten, Schaapen and Vondeling support breeding populations of

African Penguin, Cape Gannet, four species of marine cormorants, Kelp and Hartlaub’s Gulls, and

Swift Terns. The islands also support important populations of the rare and endemic African Black

Oystercatcher. Langebaan Lagoon provides an important habitat for 67 species of waterbirds, of

which half are waders. The lagoon has been identified as the most important wetland for waders on

the west coast of southern Africa, with 17 of the wader species being regular migrants from the

Palearctic region of Eurasia. Waterbird abundance is thus highest in summer, and decreases in



winter. Since 1980, there has been a decline in the numbers of waders, which has been attributed to

the siltation of the lagoon reducing the amount of suitable feeding grounds, and increasing levels of

human disturbance (Atkinson et al., 2006).

5.2.5 Beneficial Uses Besides the natural environment, beneficial uses of the coastal marine waters of South Africa are

subdivided into three categories (DWAF, 1995):

• Recreational use;

• Mariculture use (including collection of seafood for human consumption); and

• Industrial uses (e.g. intake of cooling water and water for fish processing and/or mariculture).

The identification and mapping of designated uses of the marine environment in Saldanha Bay is

drawn from the study by Taljaard and Monteiro (2002), and summarized below.

Conservation Areas Langebaan Lagoon was designated as a Ramsar site in April 1988 under the Convention on

Wetlands of International Importance especially as Waterfowl Habitat. The Ramsar site includes the

islands Schaapen (29 ha), Marcus (17 ha), Malgas (18 ha) and Jutten (43 ha), the Langebaan

Lagoon (15 km long and 12.5 km wide), and a section of Atlantic coastline. The Langebaan Lagoon is

also included within the boundaries of the West Coast National Park, which was established in 1985.

The lagoon is divided into three different utilization zones namely: wilderness, limited recreational and

multi-purpose recreational areas (Figure 5.9). The wilderness zone has restricted access and

includes the southern end of the lagoon and the inshore islands, which are the key refuge sites of the

waders and breeding seabird populations respectively. The limited recreation zone includes the

middle reaches of the lagoon, where activities such as sailing and canoeing are permitted. The

mouth region is a multi-purpose recreation zone for power boats, yachts, water-skiers and fishermen.

However, no collecting or removal of perlemoen and crayfish is allowed in the lagoon.

There are also a number of marine protected areas (MPAs) declared under the Marine Living

Resources Act 18 of 1998 (Figure 5.9):

• Langebaan Lagoon MPA

• Sixteen Mile Beach MPA

• Malgas Island MPA

• Jutten Island MPA

• Marcus Island MPA



Figure 5.9: Conservation areas in Saldanha Bay (adapted from Taljaard and Monteiro,

2002). Conservation areas of the Cape Nature Conservation Board include:

• An area within the military base, SAS Saldanha

• Vondeling Island

Mariculture Areas The Transnet National Ports Authority (TNPA) has set aside a total of 395 ha of sea area within

Saldanha Bay for mariculture activities, of which 200 ha are situated in Big Bay, 130 ha are located in

Small Bay and a further 65 ha lie adjacent to the breakwater and Small Craft Harbour. Marine

aquaculture operations currently undertaken within these areas are:

• Mussel farming;

• Oyster farming; and

• Commercial harvesting of seaweed in Saldanha Bay.

Sixteen Mile Beach MPA



Table 5.4 and Figure 5.10 provide a summary of the lease holders and the areas and location of their

leases.

Table 5.4: Current Mariculture lease holders in Saldanha Bay.

Lease Holder Lease Area (ha) Location

Blue Bay Aquafarm (Pty) Ltd 50.9 Small Bay Saldanha Bay Sea Farms (Pty) Ltd 10.6 Small Bay Blue Sapphire Pearls cc 5 Small Bay Christopher Heineken 5 Small Bay JDL Fisheries (Pty) Ltd 10 Small Bay Masiza Mussel Farm (Pty) Ltd 30 Small Bay Ocean Deli Fishing 5 Small Bay OWK (Pty) Ltd 5 Small Bay TELVEX 16 cc 5 Small Bay West Coast Seaweed(Pty) Ltd 5 Small Bay Keep the Dream 81 (Saldanha Mariculture Association) 15 Small Bay Striker Fishing 10 Small Bay 25 Big Bay West Coast Oyster Growers 10 Big Bay

i) Mussel Farming

The alien Mediterranean mussel Mytilus galloprovincialis and the indigenous black mussels

Choromytilus meridionalis are cultured on ropes, clustered 60 cm apart and suspended from rafts to

a depth of 6 m. Settlement of larvae onto the ropes occurs naturally from the water column. During

years of poor recruitment mussel spat may, however, be harvested off the caissons of the iron-ore

jetty to a depth of ~5 m, and subsequently transferred to the culture rafts. Mussels are harvested,

washed and graded on board a boat and juvenile mussels are hung back onto the ropes and held in

place by mesh ‘socks’ until attachment.

After the discovery of populations of the alien mussel Mytilus galloprovincialis in Langebaan Lagoon

and at the head of the lagoon near Schaapen Island, the South African National Park initiated a

project, in collaboration with the local community, to remove the alien mussel from Langebaan

Lagoon, selling these to local markets. However, after some time it was discovered that the alien

mussel populations had reduced drastically, and that > 80% of the mussels collected were in fact the

indigenous species Choromytilus meridionalis. It was thus decided to discontinue with this harvesting

initiative, and there are no further plans for harvesting of wild mussel populations (P. Nel, West Coast

National Park, pers. comm.).



Figure 5.10: Current mariculture lease holders in Saldanha Bay.

ii) Oyster Farming

Until recently, Saldanha Bay Oyster Company farmed the Pacific oyster Crassostrea gigas in a

completely enclosed tidal dam (reclamation dam) situated in the Port of Saldanha (Figure 5.10). This

activity has now stopped. Partial reclamation of this dam is proposed as part of the Phase 1B for the

expansion of the Sishen-Saldanha Iron Ore Export Corridor development. It is proposed that the

remainder of the dam be reclaimed as part of the proposed Phase 2 expansion of the Sishen-

Saldanha Iron Ore Export Corridor development. There are, however, currently a number of new

oyster farming ventures proposed for TNPA lease areas in both Small Bay and Big Bay (see Table

5.3 and Figure 5.10).



iii) Seaweed Harvesting

The agarophyte Gracilaria gracilis is being harvested commercially in Saldanha Bay by Taurus

Saldanha Seaweed (Pty) Ltd. The beach-cast seaweed is collected and dried, before being exported

primarily for agar processing. Annual yields, however, vary enormously and the resource has

collapsed and recovered twice over the past few decades (Anderson et al., 1996).

The first collapse occurred in 1974, after the construction of the iron ore jetty and breakwater. After

initial recovery, the resources declined again in 1989. This was attributed to over-grazing by fish in

shallow water and by keyhole limpets and urchins in deep water (Anderson et al., 1993).

A further disruption to the industry occurred in 1993, when a bloom of Ulva lactuca appeared for the

first time in Saldanha Bay and persisted throughout the summer. Ulva wash-ups contaminated the

beach and part of the commercial Gracilaria beach-cast had to be discarded (Anderson et al., 1996).

In summer, the water in Small Bay becomes highly stratified with cold nutrient-rich bottom water and

warm oligotrophic surface water. The discharge of nitrogen-rich effluents from fish factories into the

surface layer in a sector of the bay provided localised conditions for Ulva to out-compete Gracilaria at

depths of 2-5 m, demonstrating the powerful disruptive effect of eutrophication in this strongly

stratified system (Anderson et al., 1996).

Commercial and Recreational Fisheries The commercial fishery in Saldanha Bay consists mainly of line fishing from small boats and gill

netting (Figure 5.11). Gill-netting is conducted from small ski boats close to or within the surf-zone,

primarily at night (S. Lamberth, MCM, pers. comm.). Currently, there are 15 gill-net permit holders, of

which ten operate in Langebaan Lagoon and five in Saldanha Bay (MCM, 2007). Those from

Saldanha Bay operate both in Small Bay and Big Bay, but the permit conditions allow some of the

Langebaan Lagoon permit holders to also operate up to the Iron Ore Jetty in Big Bay (MCM, 2006).

Gill-net permit holders target harders and in 1998-1999 landed an estimated 590 tons annually,

valued at approximately R 1.8 million (Hutchings and Lamberth, 2002). There is one beach-seine

netting right available for Saldanha Bay but at present this right has not been taken up (MCM, 2007).

Species such as white stumpnose, white steenbras, kob, elf, steentjie, yellowtail and smoothhound

shark support the commercial line fisheries, and also a large shore angling and recreational boat

fishery, which contributes significantly to the tourism appeal and regional economy of Saldanha Bay

and Langebaan (Atkinson et al., 2006).



Figure 5.11: Designated beneficial use areas in Saldanha Bay (adapted from Taljaard and

Monteiro, 2002).

5.2.6 Existing Environmental Impacts Existing activities that potentially have a negative impact on the quality of the marine environment in

the Saldanha Bay system have been described in detail by Taljaard and Monteiro (2002). An

overview of these activities and sources are provided below (see Figures 5.11 and 5.12).

Discharges from seafood processing industries There are four seafood processing industries situated in Saldanha Bay, namely:

• Sea Harvest Corporation Ltd;

• Southern Seas Fishing;

• SA Lobster Exporters (Marine Products); and

• Live Fish Tanks (West Coast) (Lusithania).



Each of these seafood industries discharges of their effluents into the sea at Small Bay. The main

pollutants in these effluents are:

• Inorganic nitrogen;

• Organic nitrogen and carbon;

• Suspended solids; and

• Microbiological contaminants.

This nitrogen-rich discharge, particularly that from the larger fish processing factories, has been

found to have measurable effects on benthic macrofauna (Christie and Moldan, 1977), and has

caused an outbreak of the opportunistic green alga Ulva lactuca, which reduced the benthic

Gracilaria stocks in 1993/94 (Anderson et al,. 1996; Monteiro et al., 1997). The fish waste has also

been found to provide a significant source of nitrogen for seaweed cultivated throughout the northern

area of Small Bay, particularly when the water is highly stratified in summer (Anderson et al., 1999).

Sewage In the Saldanha Bay and Langebaan area, sewage can enter the marine environment via the

following routes:

• Sewage effluent from a sewage treatment works (effluent from the Saldanha Bay sewage

treatment works is into the Bok River from where it drains into Saldanha Bay opposite the

Blouwaterbaai Resort);

• Overflow from sewage pump stations (usually the result of pump malfunction or power failures);

and

• Seepage or overflow from septic or conservancy tanks, respectively.

Storm water runoff Although it is very difficult to characterize storm water runoff due to the widely varying contaminant

concentrations, it is one of the major non-point sources of pollution. In the case of the Saldanha Bay

and Langebaan, storm water runoff that could potentially have a marked effect on marine water

quality primarily originates from industrial areas (industrial zone where seafood processing industries

are situated and the Port of Saldanha and surrounding industrial sites), and the residential areas of

Saldanha Bay and Langebaan, including the area up to Club Mykonos.

Port activities and associated ship traffic Activities, associated with shipping traffic and the Port of Saldanha that can potentially impact on

marine water and sediment quality in the area include:

• Ore dust fallout during ship loading operations;



• Oil spillages;

• Dredging operations; and

• Ballast water discharge.

Activities associated with smaller harbours There are numerous smaller harbour areas in and around Saldanha Bay and Langebaan Lagoon:

• Small craft harbours (also under the jurisdiction of the TNPA);

• Fishing harbours;

• Military harbour (SAS Saldanha);

• Yacht clubs of Saldanha Bay and Langebaan, and at Club Mykonos.

Activities and operations in harbours that could contribute to the deterioration of marine water quality

include:

• Cleaning of vessels in harbour areas, as well as emptying of water closets and toilets into

harbour areas;

• Dumping of blood water from fishing vessels into sheltered harbour areas;

• Off-cuts and offal from fish cleaning operations being washed down into storm water drains and

eventually ending up in the harbour;

• Poor waste disposal practices in the scraping and cleaning of vessels (maintenance). Anti-

fouling paints are of particular concern as these often contain significant levels of tributyl-tin, a

toxin that can result in the shell deformation in shellfish.

Mussel farming Mussel farming in Saldanha Bay uses the Spanish raft system, where mussel are cultured on ropes

that are suspended in the water column from rafts. This mussel farming technique can affect marine

sediment and water quality through reducing the turbulence in the benthic boundary layer, and

through high sedimentation rates from faeces, pseudofaeces, fallen mussels and foulers under the

rafts. Research has shown that mussel debris under rafts can accumulate to a depth of 20 cm,

creating organic enrichment and anoxia in sediments. Benthic macrofaunal communities under the

rafts were found to be disturbed, displaying a reduction in biomass and an alteration of trophic groups

and taxa (Stenton-Dozey et al., 1999, 2001).



Harmful algal blooms Harmful (toxic) blooms have become a regular seasonal occurrence in the bay since 1994 (see

section 5.2.3). These blooms pose a risk to both the sensitive ecosystems in the area as well as to

beneficial uses, such as mariculture operations, and recreation and tourism.

Groundwater abstraction Langebaan Lagoon supports extensive marsh plant communities. Some of the marsh vegetation,

e.g. bulrush, Typha capensis relies on freshwater input from groundwater. Over-exploitation of

groundwater feeding into the lagoon may therefore affect the diversity of marsh vegetation.

Littering Littering, particularly plastics, has become a major problem associated with urban development, not

only in terms of unpleasant aesthetics, but also in terms of the physical harm caused to marine life.

Towards improving the quality of South African beaches the Department of Environmental Affairs and

Tourism initiated their Coastcare programme, involving local communities. The Saldanha Municipality

acts as implementing agent for the Coastcare programme in their region.

Figure 5.12: Existing activities potentially impacting negatively on the marine environment in

Saldanha Bay (adapted from Taljaard and Monteiro 2002).



5.2.7 Potentially Threatened Habitats and Beneficial Uses Taking into account the wastewater characteristics of the proposed discharge from the RO plant, potential impacts are most likely to target important marine ecosystems and beneficial uses that rely on the health of marine organisms and plants, such as the marine aquaculture activities and the fisheries. Certain areas of special interest likely to be impacted by discharges from the RO Plant into the marine environment were identified. These specific areas include:

• The natural intertidal and shallow subtidal environments adjacent to the harbour;

• Commercial and recreational fisheries;

• Seaweed harvesting;

• National Parks;

• Marine Protected Areas;

• Nearby existing and proposed mariculture activities; and

• Seawater intakes to the fish factories



6 IDENTIFICATION OF KEY ISSUES AND SOURCES OF POTENTIAL ENVIRONMENTAL IMPACT

In the course of the public participation/consultation phase of the Basic Assessment and environmental screening process for the proposed RO Plant, key issues were identified relating to potential environmental impacts. These, and additional issues identified by the specialists, are summarised briefly below in terms of the construction phase, operational phase and decommissioning phase. They are dealt with in more detail in Section 8 of this report.

6.1 Construction Phase

The potential impacts associated with the construction of feed-water intake and brine discharge structures into the marine environment are likely to be far less significant than impacts associated with current and proposed port expansion activities. Nonetheless, they need to be addressed as part of the proposed RO Plant development, and are related to:

• Onshore construction (human activity, air, noise and vibration pollution, dust, blasting and piling driving, disturbance of coastal flora and fauna). Assessment of these impacts is not considered to be the scope of work for this specialist study;

• Construction and installation of a surf-zone or deeper water discharge and intake via pipeline (construction site, pipe lay-down areas, trenching of pipeline(s) in the marine environment, vehicular traffic on the beach and consequent disturbance of intertidal and subtidal biota); and

• Construction and installation of intake wells on the beach above the high water mark, and associated pipelines leading to and from the plant (vehicular traffic in the dunes and on the beach, well excavations and consequent disturbance of dune and beach biota).

At all three of the alternative sites the RO Plant will be constructed a distance from the existing shoreline. Consequently, issues associated with the location of the plant, drilling of the beach wells, boreholes, and the associated pipelines leading from the beach wells to the plant are not deemed to be of relevance to the marine environment, and have been dealt with by other specialist studies. The project-specific geotechnical groundwater study (Visser et al., 2007) estimated that installation of the 10 beach wells required at Site 1 would disturb an area of 10 x 10 m each, a total of 1000 m2. In contrast, construction of horizontal collector wells will disturb a much larger estimated area of at least 20 x 250 m (5000 m2) along the beach. In the case of surf-zone intakes, infrastructure extending into the sea will require temporary removal of seabed material (blasting if required, piling driving and potential increased turbidity) and seabed preparation, and potential impacts to beach biota. It is assumed here that the design and construction of the necessary infrastructure on the beach or extending into the sea, will take into consideration:

• the natural dynamics of the shoreline in the region;

• potential changes in shoreline stability associated with expansion of the ore terminal facilities; and

• possible climate change issues such as sea level rise and increased storminess.



6.2 Operational Phase

The key issues and major potential impacts are mostly associated with the operational phase. The key issues related to the presence of pipeline infrastructure and brine discharges into the marine environment are:

• altered flows at the intake and discharge resulting in ecological impacts (e.g. entrainment of biota at the intake, low distortion/changes at the discharge, and affects on natural sediment dynamics);

• the effect of elevated salinities in the brine water discharged to the bay;

• biocidal action of residual chlorine (or other non-oxidisng biocides) in the effluent;

• the effects of co-discharged waste water constituents, including possible tainting effects affecting both mariculture activities and fish factory processing in the bay;

• the effect of the discharged effluent having a higher temperature than the receiving environment; and

• direct changes in dissolved oxygen content due to the difference between the ambient dissolved oxygen concentrations and those in the discharged effluent, and indirect changes in dissolved oxygen content of the water column and sediments due to changes in phytoplankton production as a result of changes in nutrient dynamics (both in terms of changes in nutrient inflows and vertical mixing of nutrients) and changes in remineralisation rates (with related changes in nutrient concentrations in near bottom waters) associated with near bottom changes in seawater temperature associated with the brine discharge plume.

Additional engineering design considerations, not strictly constituting issues to be considered within the EIA, include the following:

• structural integrity of the intake and outfall pipelines (e.g. related to shoreline movement);

• potential re-circulation of brine effluents if intakes and discharges are situated in close proximity to one another. The model results however indicate that this will not be a concern for all of the intake and dsicharge options considered in this study. Only at site 2 do the salinity and seawater temperatures from the discharge plumes result in salinity of seawater temperature elevations of approximately 0.5 psu and 0.5 to 1.0 ºC above ambient conditions at the intake and then only for short durations (see model results in Appendix C).

• the permeability and particle size distributions of the sands (should beach intake and/or discharge wells be considered), as a high proportion of fines in the beach sand could jeopardise the durability of the intake wells and effectiveness off discharge wells; and

• water quality of feed-waters that should include consideration of possible deteriorating water quality (particularly sediments that may be stirred up during normal port operations, capital and/or maintenance dredging within the port, or large-scale hypoxia of bottom waters), that may require specific mitigation measures or planned flexibility in the operations of the RO Plant.



6.3 Decommissioning Phase

The minimum anticipated life of the RO plant is 25 years. The individual RO modules will be replaced as and when required during this period. No decommissioning procedures or restoration plans have been compiled at this stage, as it is envisaged that the plant will be refurbished rather than decommissioned after the anticipated 25 years, as the beach wells/boreholes (if used) are expected to have a considerably longer lifespan than the RO modules. The potential impacts during the decommissioning phase are expected to be minimal in comparison to those occurring during the operational phase, and no key issues related to the marine environment are identified at this stage. As full decommissioning will require a separate EIA, potential issues related to this phase will not be dealt with further in this report.






7 MODELLED CHANGES IN THE MARINE ENVIRONMENT DUE TO THE RO PLANT EFFLUENT DISCHARGES TO THE MARINE ENVIRONMENT

7.1 Introduction

The brine and co-discharges have been modelled using the Delft3D modelling suite that has been set up to simulate both the three dimensional hydrodynamics and water quality in Saldanha Bay. The model set-up and calibration is described in detail in Appendix B. The model simulations have been undertaken for five intake and discharge combinations associated with 3 RO plant locations proposed.

7.2 Assumed intake and discharge locations

A detailed description of the three proposed RO Plant sites are given in Section 3.2 of this report (see Figure 3.2). Due to the specific constraints at the various sites, the conceptual intake and discharge designs for the various sites differ substantially. The assumed location of the various intake and discharge locations are given in Table 7.1 below and are indicated in Figures 7.1a to d. A detailed description of the various intake and discharge combinations is given in Section 3.5 of this report where all of the options considered are summarised in Table 3.1. The detail of the conceptual designs considered in the modelling study are provided below. The intake and discharge combinations assessed in the modelling study are summarised in Section 7.5 and Table 7.5. In the modelling study it is not possible to explicitly simulate either beach well intakes or discharges. Thus only pipeline intakes have been simulated. In terms of the assessment of the proposed RO Plant discharges into the marine environment, the only difference to the modelling outcomes due to the assumption of pipeline intakes rather than beach well or borehole intakes is that there will be a reduction in CIP chemicals and backwash sediments for the beach wells and borehole intake options compared to the pipeline intakes. Provided that the intake waters for beach well and boreholes do not have a significantly different temperature and salinity characteristics to those from pipeline intakes, the modelling of plume dynamics and the subsequent assessment potential impacts of a RO Plant discharge undertaken will remain valid and conservative (i.e. “worst case”) in that the CIP chemicals and backwash sediments for the beach wells and borehole intake options are likely to be significantly reduced compared to the pipeline intakes.



Figure 7.1a: Proposed location of intake and discharge structures for Site 1.

Figure 7.1b: Proposed location of intake and discharge structures for Site 2. The yellow dotted line encloses the potential area within which beach wells may be located.



Figure 7.1c: Proposed location of intake and discharge structures for pipeline discharges into Small and Big Bay for Site 3.

Figure 7.1d: Proposed location of borehole intakes (adjacent to stockyard and MPT) discharge location for a pipeline discharge from Site 3 at Caisson 3. The intake boreholes will be located within the linear distance marked by the white lines in the figure opposite the existing iron-ore stockyard and the MPT (see Figures 3.5 and 3.6 in Section 3 for more detail).






Table 7.1: Location of the intake and discharges for the various proposed sites.

Development Option 1a 1b 2a 2b 3a 3b 3c & 3d

Intake Location

Site 1 Beach well

intake (1a)

Site 1*2

Surf-zone (sump) intake*1

Site 2

Beach well intake

Site 2*3

Surf-zone (sump) intake*1

Site 3*3

Pipeline intake

Site 3*45 Borehole

intake (Stockyard)

Site 3*5 Borehole

intake (MPT)

Latitude 33º 00’ 29.58’’ S 33º 00’ 35.31’’ S 32º 59’ 56.63’’ S

(nominal) 32º 59’ 56.62’’ S 33º 01’ 07.89’’ S

33º 00’ 07.41’’ S to 33º 00’ 26.90’’ S

33º 00’ 56.22’’ S to 33º 001 11.77’’ S

Longitude 18º 00’ 33.17’’ E 18º00’ 36.54’’ E 17º 59’ 46.54’’ E

(nominal) 17º 59’ 38.84’’ E 17º 59’ 14.58’’ E

17º 59’ 46.69’’ E to 17º 59’ 41.11’’ E

17º 59’ 28.42’’ E to 17º 59’ 21.41’’ E

Discharge Location

Site 1 Pipeline outfall

Site 2*3

Pipeline outfall

Site 3*4 Pipeline outfall

(Small Bay)


(Big Bay)


(Caisson 3)

Latitude 33º 00’ 34.11’’ S to 33º 00’ 33.377’’ S

33º 00’ 02.42’’ S

33º 01’ 19.26’’ S (33º 01’ 06.43’’ S)*4 33º 01’ 39.70’’ S

Longitude 18º 22’ 22.61’’ E to 18º009’ 23.33’’ E

17º 59’ 45.59’’ E

17º 59’ 13.82’’ E (17º 59’ 29.09’’ E) 17º 59’ 08.62’’ E

*1 The option of a surf-zone intake via a sump and delivery of intake waters via pipeline to the RO Plant is identified as a pipeline intake in Table 3.1. *2 For Site 1 the intake structures considered are either beach wells or a possible pipeline intake with a possible sump located in the surf zone in an approximate –1

m CD water depth (see Table 7.2). Given that strong wind-driven currents flow from SE to NW in summer, the surf-zone intake structure is located to the SE of the proposed discharge structure comprising a pipeline discharge at the surface over an approximate 30 m distance along the revetment structure of the existing reclamation dam (Figure 7.1a). Given that it is a surf-zone intake, sediment loading may be an issue and sump structures will need to be considered. This has been the approach for the Marine Growers abalone farm just NE of the Port of Coega.

*3 Given that the current flow is predominantly clockwise around the eastern side of Small Bay, it is envisaged that the sea water intake would be situated in or near the surf-zone in an approximate -1 m CD water depth at a location west of the discharge. The discharge is likely to be a surf-zone discharge or a single port discharge in a very shallow water depth, also assumed here to be – 1 m CD. The discharge position is located in shallower water just inshore of the edge of deeper dredged area in this section of the port.

*4 Given that the current moves clockwise around the eastern side of Small Bay it is envisaged that the sea water intake would be situated north of the proposed discharge location in Small Bay in an approximate -1 m CD water depth. The discharge is likely to be a surf-zone discharge of single port discharge in a very an approximate 8 m water depth for a discharge into Small Bay or a shallower (approximately 4 m) water depth if discharged into Big Bay.

*5 The boreholes along the quayside are located at two possible locations along the causeway, namely opposite the existing iron-ore stockyard and opposite the MPT.



Table 7.2: Main characteristics and conceptual design of intake structures for the RO plant.

Site 1 Site 2 Site 3 (a&b) Site 3 (c&d)

Intake Design *1

Beach wells (10 worst case) or surf-zone sump intake via

pipeline.

A surf-zone (sump) intake via pipeline has been assumed for the purposes of the modelling

study *2.

10 Beach wells or a surf-zone (sump) intake via pipeline.

A surf-zone (sump) intake via pipeline has been assumed for the purposes of the modelling study *2

Pipeline intake in quayside

Up to 10 borehole intakes located on the causeway alongside the iron-ore stockpiles or the MPT

A pipeline intake alongside the quayside has been assumed in the

modelling study *2

Co-ordinates of intake structures

33º 00’ 29.58’’ S 18º00’ 36.54’’ E

32º 59’ 56.57’’ S 17º 59’ 38.64’’ E

33º 01’ 07.96’’ S 17º 59’ 14.62’’ E

Stockpile 33º 00’ 07.41’’ S to 33º 00’ 26.90’’ S 17º 59’ 46.69’’ E to 17º 59’ 41.11’’ E

MPT 33º 00’ 56.22’’ S to 33º 001 11.77’’ S 17º 59’ 28.42’’ E to 17º 59’ 21.41’’ E

Distance from shoreline

Approximately 65 m from the shoreline (exact distance depends on bathymetry)

Approximately 75 m from shoreline (exact distance depends on

bathymetry) Intake alongside quay wall Borehole intakes located along

causeway

Distance from discharge point 350 m 250 m

350 m (for discharge to Small Bay)

Other side of the causeway (for discharge into Big Bay)

Approximately 900 m to the closest borehole (for boreholes located

alongside the MPT) Approximately 2400 m to the

closest borehole (for boreholes located alongside the iron-ore

stockyard)

The distance between the Site 3 intake alongside the quay wall as

assumed in the modelling study and the discharge at Caisson 3 is

approximately 1000 m



Site 1 Site 2 Site 3 (a&b) Site 3 (c&d)

Intake Depth Note 3

-1 m CD (approximately 1.865 m below MSL)

This allows for a minimum water depth of approximately 1 m at all

times



times



times

Not relevant for borehole intakes -1 m CD (approximately 1.865 m

below MSL) for intake alongside the quay wall as assumed in the

modelling study.


times

Intake flow rates 8000 m3/day or

92.6 ℓ. s-1 for 3 RO unit plant (assumed to be continuous)

8000 m3/day or 92.6 ℓ. s-1 for 3 RO unit plant (assumed to be continuous)



Intake velocity Velocity is estimated to be

approximately 0.15 m.s-1 at the intake screens and 1.0 m.s-1 in

the intake pipeline

Velocity is estimated to be approximately 0.15 m.s-1 at the

intake screens and 1.0 m.s-1 in the intake pipeline

Velocity is estimated to be approximately 0.15 m.s-1 at the

intake screens and 1.0 m.s-1 in the intake pipeline

Not of relevance for a borehole intake

*1 The assumed specifications for the intake beach well design are that supplied to date by Transnet (see Section 3.3.1).

*2 Beach wells are not explicitly considered in the modelling study. In the modelling study only surf-zone sump intakes have been considered for Sites 1 and 2 and only an intake in the quayside for Site 3 discharges. This implies that an explicit assumption has been made that the temperature of the intake waters from the beach wells will not differ significantly from that for a surf-zone sump intake at Sites 1 and 2. Similarly it is assumed that the temperature of the intake waters from boreholes situated along the causeway will not differ significantly from that of a near-surface intake at the quay wall when considering a discharge at Site 3 discharge at Caisson 3. For surf-zone sump intakes, this is a reasonable assumption. For borehole intakes along the causeway it is not clear how different the temperature of the borehole water would be compared to a near-surface intake at the quay wall. It is likely that the water temperature from the boreholes would be close to that of the seasonal mean near-surface seawater temperatures. Preliminary data from boreholes (D. Visser, pers comm.) seem to indicate that this is not an unreasonable assumption however, this cannot be confirmed without data on the borehole water temperatures. The brine and biocide concentrations in the discharge will not differ according to the nature of the intake other than that the likely biocide concentrations for beach wells/boreholes could be significantly less than those for pipeline intakes. In the modelling we have assumed oxidising biocide concentrations of 0.1 mg. ℓ-1 NaOCl or a non-oxidising concentrations ranging between 1.15 mg.ℓ-1 and 2.475 mg.ℓ-1 DBNPA.

*3 It is assumed that the intake would need to be in a water depth such that the intake remains below water under all tidal, wind and wave conditions. The intake depth is located near the surface to obtain warmest water possible but sufficiently away from the surface to avoid entrainment of debris and possible oils. Intakes at greater depth will have a greater chance of entraining sediments, low oxygen waters or H2S (if present). There is a possibility of a significant temperature difference between the intake waters and the ambient temperatures at the discharge depth and this needs to be considered when assessing the discharge. Transnet has indicated that there may be a rise in temperature of the seawater/brine discharge between the intake and discharge points, however Transnet (through their contractor) has indicated that a typical value for such a temperature increase is a maximum of one degree Celsius. The implication of this is discussed in detail in Section 4.2..






7.3 Intake Characteristics

Important specifications for the seawater intakes include:

• the location of the discharge: - is specified in terms of actual co-ordinates and the distance from both the shoreline and the discharge location. The exact location determines other characteristics such as the water quality, temperature and salinity at the intake as well as the design of the intake;

• the nature and design of the intake: - it is assumed that the intake infrastructure is likely to comprise a surf-zone intake or beach wells at Site 1 and Site 2 and intake pipelines or boreholes at Site 3, where there is more limited potential for installing beach wells. It is assumed that all intakes will be designed to minimise entrainment of biota and will have self-cleaning screens where appropriate/possible. Obviously this will not be a requirement for the beach well/ borehole intake structures or sumps located in the surf-zone;

• the intake depth: - that influences the intake temperature. It is assumed that the intake waters should be as warm as possible as this increases RO plant efficiency. This implies that the intake should be located near the surface (approximately -1 m CD) where the waters are warmer than at depth and also where there is less likelihood of the entrainment of sediments, low oxygen water and possibly H2S if present (although considered unlikely). However the intake should be located sufficiently deep to avoid entrainment of surface debris and possible oil spills;

• the nature of the flows (Intermittent or continuous). Transnet has specified all flow to be continuous as at the steady rates given in Table 7.2.. However shock dosing by biocides has been proposed. The implications of this are discussed in Section 4.2)

The main characteristics of the intake structures at each site is summarised in Table 7.2. There are water quality requirements for the intake waters and it is assumed that pre-treatment

processes may be needed for one or more of the following:

− biofouling prevention;

− control of biological activity (disinfection as well as dechlorination if chlorine is used);

− prevention of scaling and inorganic precipitation, including metals removal, and

− removal of other elements such as sulphur (H2S) and silica.

The use of beach wells and/or boreholes will minimise or, in some cases, remove the requirements for pre-treatment. The assumed specifications for the RO intake waters are (Appendix III, Wetland Consulting Services,

2007):

− 34.9 psu (actual anticipated salinity in Saldanha Bay – changed from the 35.5 psu

assumed by Wetland Consulting Services (2007);



− Suspended solids = 8-25 mg.ℓ-1 after primary screening;

− Sea water temperatures ranging from 12-22 ºC;

− Oil: Negligible

− Algae bloom: Negligible.

The seawater temperatures, depending on the intake depth can be as cold as 9°C, however, surface

waters are expected to remain within the range specified above. It should be noted that the operating

costs of plants will increase with colder intake waters (Swartz et al., 2004) so it is anticipated that the

intakes will be located as near to the sea surface as possible where the water is warmer. The

temperature of waters from the beach wells or boreholes will be determined by the exact location of

the intakes and flow rates possible for each beach well/borehole.

Phytoplankton blooms (see Section 5.2.3) do occasionally occur in Saldanha Bay and could thus be a factor in intake water quality. Similarly, under turbulent conditions in the bay (high wind and wave conditions) significant turbidity due to the stirring up of bottom sediments is often visible suggesting reasonably high turbidity conditions in the water column. However, previous dredge monitoring studies indicated that the discolouration of the water in the bay is often associated with relatively low suspended sediment concentrations in the water column (perhaps a consequence of the fact that the sediments are white and highly visible). These impacts on intake waters would be significantly reduced for beach wells and minimal for boreholes. Typically oil in the water column should be negligible, however, Saldanha Bay is an oil terminal and consequently oil spills are a possibility. It thus is prudent to locate the intake one or two metres below the sea surface. These impacts on intake waters would be somewhat reduced for beach wells and boreholes, however should an oil spill occur that results in significant quantities of dissolved components in the water column or oiling beaches, these impacts would be significant. Given the low oxygen conditions in Small Bay, H2S may be an issue if the intake draws on the bottom waters within the bay, but should not be an issue if water is abstracted on the eastern side of Small Bay or away from the seabed elsewhere in Small Bay. The bottom waters in Big Bay are considered to be fairly well oxygenated (within the context of the large scale low oxygen water fluctuations occurring along the West Coast), precluding risk of poor water quality at the intake as described above. The oxygen concentrations in beach well and borehole intake waters is uncertain.

7.4 Discharge Characteristics

The specific environmental and engineering constraints at the various sites result in substantially different conceptual designs at the various discharge locations. The main discharge characteristics and conceptual designs are described in Table 7.3 below.



Table 7.3: Main characteristics and conceptual design of discharge structures for the RO plant.

Development Phase

Site 1

Site 2

Site 3 (a) (discharge into Small

Bay)

Site 3 (b) (alternate discharge

into Big Bay)

Site 3 (c&d)

(discharge at Caisson 3)

Discharge Design

Discharge at the sea surface along a 30 m length of the

revetment of the reclaim dam at a location approximately 80 m

from low water mark of adjacent shoreline with an adjacent water

depth of approximately 1.5 m through multi-port diffuser

Surf-zone discharge*1 or

pipeline discharge through single port diffuser in an approximate -

0.5 to -1m CD water depth

Single port diffuser in an approximate -8 m CD water

depth

Single port diffuser in an approximate -4 m CD water

depth

Single port diffuser in an approximate -16 to -18 m CD

water depth

Behaviour

Brine discharge sinks to bottom one third of the depth of the water column upon being

discharged (approx 0.5 m layer thickness)

Brine discharge sinks to bottom half of the depth of the water

column upon being discharged (approx 0.7 m layer thickness)

Brine discharge sinks to bottom one third of the depth of the water column upon being

discharged (approx 3.0 m layer thickness)

Brine discharge sinks to bottom approximately one third

of the depth of the water column upon being discharged (approx 2.1 m layer thickness)

Brine discharge sinks to bottom approximately one third

of the depth of the water column upon being discharged (approx 2.7 m layer thickness)

Co-ordinates of intake

structures

33º 00’ 34.11’’ S to 33º 00’ 33.377’’ S

18º 22’ 22.61’’ E to 18º009’ 23.33’’ E

33º 00’ 02.42’’ S 17º 59’ 45.59’’ E

33º 01’ 19.33’’ S 17º 59’ 13.84’’ E

33º 01’ 06.50’’ S 17º 59’ 28.94’’ E

33º 01’ 06.50’’ S 17º 59’ 28.94’’ E

Distance from shoreline

80m from shoreline but along the edge of the revetment 20 to 25 m 90 m 40 m 65 m (from end of causeway)

Distance from intake 350 m 250 m 350 m Other side of causeway 1 000 m

Discharge Depth

adjacent water depth of approximately 1.5 m

approximate -0.5 to -1m CD water depth

approximate -8.4 m CD water depth



Discharge flow rates

4400 m3/day or 50.9 ℓ. s-1 for 3 RO unit plant

(assumed to be continuous 24/7 discharge)


(assumed to be continuous24/7 discharge)







Discharge velocity

Unknown at present but will be as specified required for an

optimal discharge diffuser design







Unknown at present but will be as specified required for an optimal discharge diffuser

design *1 A surf-zone discharge is unlikely to be used here as the wave action to assist dispersion is limited. A single-port discharge may be used where the water is slightly

deeper. This area has not been surveyed. However, based on Google Earth imagery we have assumed a depth of at least -0.5 m CD. A single port discharge in shallow water was used in the model simulations






Anticipated discharges from the RO plant comprise mainly a brine. The volumes of brine, the salinity of the brine and associated co-discharges are tabulated below (Table 7.4) for an assumed 45% recovery of rates of the RO modules (i.e. a brine to sea water ratio of 1:1.55).

Table 7.4: Discharge rates and salinity of the brine discharge for the various development phases and plant capacities.

Salinity Temp Change Biocide Other

Development Phase

Number of 1200

kℓ RO Units Required

Potable Water

Produced(m3/day)

Brine Discharge Volumes (m3/day) (psu)*1 (ºC)*2 mg.ℓ-1*3 mg.ℓ-1*4

Phase 1A/1B 1 1200 1467 63.5 +0 0.1 see Table 8.3

Phase 2A/2B 3 3600 4 400 63.5 +0 0.1 See Table 8.3

*1 A recovery rate of 45% of freshwater from the RO modules has been assumed here. *2 It has been specified that there is no temperature elevation in the intake waters before discharge due to

the water passing through the RO plant reticulation system. There may be a slight warming of the water as it is piped to the RO plant, stored in a buffer tank or as it is piped back from the plant to the discharge point. These effects are ignored in the modelling study where the only significant change assumed is the temperature difference between the intake waters and the ambient temperature of the waters into which the brine will be discharged. The implication of this for the impact assessment is discussed in detail in Section 4.2) The cleaning chemical may be heated with a 15kW heater for better chemical activation. The chemicals are recycled to the cleaning tank and finally flushed with permeate – thus the effects of the heating of the cleaning chemicals being flushed out of the system will have negligible effect on the brine discharge.

*3 Biocides comprising NaOCl with a free chlorine residual of 0.1 mg/ ℓ-1 or DBNPA with a residual concentration ranging between 1.15 mg.ℓ-1 and 2.475 mg.ℓ-1 in the discharge has been assumed for the purposes of assessment. The preferred and most likely biocide to be used is DBNPA.

*4 The discharge rates of all co-discharges is assumed to be continuous, i.e. all co-discharges are “bled” into the brine discharge 24 hours a day and 7 days per week. However Transnet (through their contractor) has indicated that shock dosing by biocides is being considered rather than continuous dosing. The implication of this in terms of the model results and associated impact assessments is discussed in greater detail in Section 4.2 that contains the assumptions and limitations of this impact assessment . Note that the water licence application places the legal onus on Transnet to fully specify all potential constituents in the proposed discharge.

The model simulations undertaken here do not explicitly describe the near-field behaviour of the effluent. For the purpose of assessing the far-field impacts using the DELFT3D-FLOW three-dimensional numerical model, we have assumed near-field behaviours of the discharged effluent. As the intake is generally assumed to be near the surface, the effluent when discharged at any location other than at or near the surface, is likely to be significantly warmer than the ambient temperatures at depth at the discharge locations. The buoyancy of the warmer effluent discharged at depth will help with the dispersion of the effluent plume throughout the water column, however, the high salinity content of the brine means that the effluent remains a dense discharge that will tend to sink to the seabed unless vigorously mixed throughout the water column. (A typical conservative design requirement for such a dense brine are that the discharge diffuser is designed to achieve near-field dilutions exceeding 50 as predicted by near field models such as CORMIX (Jirka et al., 1996) and UOUTPLM (Baumgartner et al., 1971).



For each location, either the most conservative discharge design (conservative approach) or the most likely discharge design (based on information from Transnet) has been considered, resulting in a total of five simulated discharge scenarios. We suggest that the most conservative discharge structure designs (typically a single port bottom discharge) be considered to provide a worst case scenario and that thereafter, if required, further simulations be undertaken with more optimal conceptual engineering designs. In the simulations undertaken we have assumed the most conservative discharge designs (typically a single-port discharge angled upwards into the water column). An associated conservative behaviour is assumed whereby the dense brine effluent (no matter what its temperature and despite being jetted into the water column in the near-field) will be confined to between one third and one half of the water depth from the seabed. This provides scope for improvement in the effluent dispersion through the water column by more optimal diffuser designs than those conceptual designs assumed here. The specific near-field behaviours that may be assumed are as follows:

• Single-port surface/bottom discharge: Here we assume dense brine will sink if discharged at the surface through a single port. Consequently for both surface and bottom discharges, in the modelling the full discharge will occur into the layers of the far-field model representing the bottom half to one third of the vertical extent of the local water column;

• Surf-zone discharge: Despite being located in the surf zone, it is assumed that the full discharge will occur into the layer of the far-field model representing the bottom half to one third of the vertical extent of the local water column. This is a conservative assumption that allows for low turbulence conditions and limited mixing of the dense effluent into the water column expected in the surf-zone during calm, low wave periods.

• Optimal (multi-port) diffuser design at depth: While it is possible to design a diffuser that will mix the effluent throughout the water column, this may come at elevated operational costs (e.g. increased pumping costs). (In the model simulations only single port discharges have been assumed where the full discharge is assumed to occur into those layers of the far-field model representing the bottom half to one third of the vertical extent local water column, i.e. a conservative scenario.) It is assumed that the diffuser ultimately will be designed that under all environmental conditions that the effluent is mostly mixed throughout the water column. There will be occasions when the effluent will not mix throughout the water column, however, this is likely only to occur under extreme stratification and/or strong horizontal flows.

A typical near field behaviour of a dense brine discharge (simulated using CORMIX) is given in Figure 7.2 below. In this model output the initial jetting of the dense brine upwards into the water column can be observed. However, depending on the extent of this initial mixing, the volumes of brine being discharged and the local environmental conditions (i.e. extent of water column stratification, the magnitude and vertical shear in these flows), a range of near-field behaviours are possible (NV1 to NV5 in Figure 7.2). However, it is anticipated that the brine, after being jetted up into the water column, will again sink towards the seabed and be advected away from the discharge point in the near-bottom layers of the water column (as indicated in the main schematic in Figure 7.2).



Figure 7.2: Schematic of the near-field behaviour of a dense effluent.

7.5 Scenarios simulated

A single combination of intake and discharge has been simulated for each of the proposed sites, except for Site 3 where three possible intake and discharge combinations have been considered. The discharge scenarios simulated in the model are as detailed in Table 7.5 below.



Table 7.5: The five intake and discharge combinations simulated in the hydrodynamic and water quality modelling.

Model Scenario#

Site alternative

Alternative Intake/Outfall Infrastructure Locations

1 Site 1 1b): Pipeline intake and pipeline outfall (Big Bay) *1

2 Site 2 2b): Pipeline intake and pipeline outfall (Small Bay) *2

3 Site 3 3a): Pipeline intake (Small Bay) and pipeline outfall (Small Bay)

4 Site 3 3b): Pipeline intake (Small Bay) and pipeline outfall (Big Bay)

5 Site 3 3c & 3d): Pipeline intake (Small Bay) and pipeline discharge at Caisson 3 (Big Bay) *3

*1 If the characteristics of the intake waters from beach well intakes at Site 1 are considered to be the same as those originating from a pipeline intake in the surf-zone at Site 1, then this model scenario can be considered to be representative of a beach well intake and pipeline outfall into Big Bay from Site 1 (Site1 option (a) in Table 2.1). There are no model simulations representing beach well discharges at Site 1 as these are considered not to be a feasible option (Visser et al, 2007).

*2 If the characteristics of the intake waters from beach well intakes at Site 2 are considered to be the same as those originating from a pipeline intake in the surf-zone at Site 2, then this model scenario can be considered to be representative of a beach well intake and pipeline outfall into Small Bay from Site 2 (Site2 option (a) in Table 2.1).

*3 If the characteristics of the intake waters from borehole intakes i) on the quay adjacent to existing iron-ore stockpiles or ii) on the quay (adjacent to MPT) are considered to be the same as those originating from a pipeline intake alongside the quay at the southern extremity of the MPT quay at Site 3, then this model scenario can be considered to be representative of borehole intakes on the quay (either adjacent to the existing stockpiles or adjacent to the MPT) and a discharge from Site 3 via pipeline outfall located at Caisson 3 (Site3 options (c & d) in Table 2.1).

Each of these discharge scenarios has been assessed under a range of environmental conditions comprising:

• a late summer period when the bay is most stratified resulting in strong surface flows and generally significantly weaker bottom water flows;

• a winter scenario when the water column is well-mixed and surface and bottom water flows are similar and;

• an unusually calm period the water column is likely to be highly stratified and calm conditions and weak flows prevail resulting in potentially significantly reduced dispersion of effluents.

The summer and winter simulations are of a two-month duration, while the unusually calm period (autumn) simulation is of a one-month duration. All model outputs have been analysed per season and the model outputs scaled accordingly.



The unusually calm period simulated corresponds statistically with the calmest period measured along the west coast in more than 20 years. The calmest periods in Saldanha Bay are typically the months of April to June. The calmest period identified in existing records measured at Cape Columbine was that of May 1986 when the wind speeds were typically only 75% of the mean wind speeds observed in May during other years. The calmest period in Saldanha Bay was June 2003, while the calmest May on record was May 1997. The least interannual variability in wind speeds observed for the months of April to June is for the month of May. In the modelling study, a “calm” period was synthesised by scaling down the magnitude of the winds of May 1999 (that constituted an “average” year in terms of winds speed over Saldanha Bay during May) to 75% of the wind speed measured in Saldanha Bay in May 1999. Whilst the average or median winds speeds in Saldanha Bay are often the lowest in June in any particular year, the month of May has been chosen for synthesising a “calm” period as the stratification in the water column remains significant through April to late May. It is these more stratified conditions and associated limited vertical mixing that lead to weak near-bottom flow velocities and the accumulation of effluent discharged at depth.

The impact assessment reported in Section 8 is based on analyses of the outputs from these model

simulations, a description of the natural environment and beneficial uses that may be impacted upon.

7.6 Model Results

7.6.1 Analysis of results The out puts from the various model simulations have been analysed for each discharge scenario as

follows:

For each of the scenarios, the changes in seawater temperature, salinity and the biocide

concentrations have been analysed per season (90 day period) as indicated, (i.e. for summer, winter

and an “autumn” period) for both surface and bottom waters. Consequently all reported days of

exceedance of a water quality guidelines are for an approximate 90 day period. For example, 45

days of exceedance represents the fact that the water quality guidelines is exceeded for roughly half

of the time in a season. These plotted results for the analyses described below are contained in

Appendix C.

The water quality guidelines and the motivation for their use are given in Section 8 of this report.

Salinity

The changes in salinity have been plotted as:

• 80%, 90%, 95% and 99% exceedance contours which indicate the ΔS values exceeded for a

total of approximately 18 days, 9 days, 5 days and less than 1 day in a season, respectively. In

keeping with a conservative approach we have reported the 99% exceedance contours only.

These 99% exceedance contours more or less represent the maximum values observed



(maximum salinity footprint of the salinity plumes), i.e. values that, at worst, are only exceeded for

a total of 22 hours within any particular season;

• Days of exceedance of the South African Water Quality guideline of ΔS < 1 psu (or S < 36 psu).

This represents a cumulative duration of exceedance as described above;

Time series plots of “before” and after” salinity are available for analysis at a number of locations in a

radius of between 50m and 100m from each discharge and at identified sensitive locations. As for

the results indicating elevations in seawater temperature, time series plots of the predicted changes

in salinity are deemed superfluous as the spatial analyses described above are adequate in

characterising the footprint of the plume. Furthermore, due to the limited “footprint” of the discharge

plumes, in has not been necessary to plot time series at identified sensitive site in the bay. Similarly time series plots of “before” and after” salinity at the proposed intakes are not reported as the influence of the discharge plumes at the proposed intake locations is shown by the spatial analyses (i.e. 99% exceedance plot of temperature and salinity) to be negligible.

Temperature

The changes in seawater temperature have been plotted as:

• 80%, 90%, 95% and 99% exceedance contours which indicate the ΔT values exceeded for a

total of approximately 18 days, 9 days, 5 days and less than 1 day, respectively. In keeping with

a conservative approach we have reported only the 99% exceedance contours that more or less

represent the maximum values observed (maximum thermal footprint of the thermal plumes), i.e.

values that, at worst, are only exceeded for a cumulative total of approximately 22 hours within

any particular season. This period represents a cumulative total that in principle could represent

a single 22 hour exceedance or, at the other extreme, 2 hourly12 exceedances on up to 11

different occasions during the season.

• Days of exceedance of the South African Water Quality guideline of ΔT < 1 ºC. This represents a

cumulative duration of exceedance as described above;

• Exceedance of the ANZECC (2000) temperature guideline that requires that the median

temperature in the environment with an operational discharge should not lie outside the 20 and

80 percentile temperature values for a reference location or ambient temperatures observed prior

to the construction and operation of the proposed discharge

• Time series plots of “before” and after” seawater temperature are available for analysis at a

number of locations in a radius of between 50 m and 100 m from each discharge and at identified

sensitive locations. These are deemed superfluous as the spatial analyses described above are

adequate in characterising the “footprint” of the plume. Furthermore, due to the limited “footprint”

of the discharge plumes, it has not been necessary to plot time series at identified sensitive site

in the bay. Similarly time series plots of “before” and after” seawater temperature at the proposed intakes are not reported as the influence of the discharge plumes at the

12 The temporal resolution of the model output is at 2 hourly intervals except where a specific site has been targeted for

more detailed time series analysis.



proposed intake locations is shown by the spatial analyses (i.e. 99% exceedance plot of temperature and salinity) to be negligible.

Biocides

The biocide (free chlorine or residual DBNPA) concentrations have been plotted as:

• 80%, 90%, 95% and 99% exceedance contours which indicate the μg.ℓ-1 oxidising biocide

concentrations exceeded for a total of approximately 18 days, 9 days, 5 days and less than

1 day, respectively. In keeping with a conservative approach we have reported the 99%

exceedance contours only that more or less represent the maximum values observed (maximum

footprint of biocides in the plumes), i.e. values that, at worst, are only exceeded for a total of 22

hours within any particular season;

• Days of exceedance of the selected Water Quality guideline that the biocide concentrations

should remain below 3 μg.ℓ-1or, in the case of DBNPA, below the tarrget values proposed.

Time series plots of biocide concentration are available for analysis at a number of locations in a

radius of between 50 m and 100 m from each discharge and at identified sensitive locations. As for the results indicating elevation in seawater temperature and salinity, time series plots of the predicted changes in biocides are deemed superfluous as the spatial analyses described above are adequate in characterising the footprint of the plume.

Achievable Dilutions of a generic contaminant

To assist with the assessment of potential co-discharges (see Section 8.1.3.3), achievable dilutions

of a generic contaminant have been plotted as:

• 80%, 90%, 95% and 99% exceedance contours which indicate the achievable dilutions exceeded

for a total of approximately 18 days, 9 days, 5 days and less than 1 day, respectively. In keeping

with a conservative approach we have reported the 99% exceedance contours only that more or

less represent the maximum values observed (maximum salinity footprint of the salinity plumes),

i.e. values that, at worst, are only exceeded for a total of 22 hours within any particular season;

• Days of exceedance of the selected Water Quality guideline that state that the oxidising biocide

(NaOCl) and non-oxidising biocides (DBNPA) concentrations should remain below 3 μg.ℓ-1; and

below 0.035 mg. ℓ-1 (or 0.070 mg. ℓ-1 if a less conservative approach is taken), respectively.

The modelling results have been set-up to simulate the transport and fate of a generic dissolved

contaminant in the water column. From the information on the dispersion of this hypothetical tracer, it

is possible to calculate the achievable dilution of a generic contaminant, provided the contaminant

behaves as a conservative tracer.



The achievable dilution (AD) in the model simulations is calculated as:

To assess compliance of individual co-discharges against the relevant water quality guidelines, the

required dilutions (RD) of any specific pollutant to ensure compliance with its associated water

quality guideline may be determined using appropriate water quality guidelines. The required

dilutions for compliance with the relevant water quality guidelines are calculated as follows:

Note that the

required dilutions to ensure compliance are dependant on the assumed background concentrations

of the relevant co-discharge constituent as these determine how “effective” the dilution process is in

lowering co-discharge constituents in the water column. The relevant water quality guideline is

assumed to have been met if the calculated achievable dilutions in the model exceed the required

dilutions determined from the constituent concentrations at the point of discharge and the relevant

water quality guideline that needs to be met.

7.6.2 Summary of results in terms of exceedance of selected water quality guidelines

The plots referred to in Section 7.6.1 above are contained in Appendix C. Due to the vast number of

plots associated with the five discharge scenarios and the respective parameters of concern (i.e.

salinity, temperature, biocides and potential co-discharges), the plume dimensions at the sea surface

and near the seabed as determined for the various parameters of concern are summarized in Tables

7.6 to 7.9 below. The plume dimension reported is the maximum distance of any plume edge from the

discharge location. The plume edge is delineated by the contour indicating where the relevant guideline has

been exceeded for no more than 6 hours within a season.

Salinity

The model results for elevated salinity are summarised in Table 7.6 below. The dimensions of the plume indicated are the maximum dimension in any direction of the plume “footprint”, where the dimensions of the “footprint” are determined by the exceedance of selected water quality guidelines (1 psu and 4 psu) for periods ranging from 6 hours in a season to approximately 5 days. In the table the figures roughly represent the dimensions of the maximum “footprint” of the effluent plume (the “footprint” where the relevant water quality guideline or target value is exceeded for no more than 6 hour per season), while the figure in brackets indicates the spatial dimensions of the effluent “footprint” that is exceeded for less than approximately 5 days in a season. The column indicating the maximum elevation in salinity (ΔS) of the ambient waters at the intake due to the influence of the discharge plume does not inform the assessment of environmental impacts but rather indicates whether or not “re-circulation” effects are likely to be a concern at the intake, whereby intake waters are contaminated by the discharge plume waters.

ionconcentratBackgroundionconcentratguidelinequalityWaterionconcentratBackgroundionconcentratcolumnwaterSimulatedRD

−−

=

ionconcentratBackgroundionconcentratcolumnwaterSimulatedionconcentratBackgroundionConcentrateDischAssumedAD

−−

=arg



Table 7.6: Summary of effluent plume dimensions around the discharge point (based on exceedances of the salinity water quality guidelines for cumulative periods of 6 hours and approximately 5 days), and the magnitude of the salinity elevation at the intake.

Maximum Discharge Plume Dimensions

(m)

Max ΔS at intake

(psu) Site WQ*1 guideline

Location in water column

Summer Winter Autumn Summer Winter Autumn

surface -*2 -*2 -*2 < 0.10 < 0.20 < 0.20 SAWQ

(ΔS < 1psu) bottom 290 (160)

200 (150)

400 (180)

< 0.10 < 0.20 < 0.20

surface -*2 -*2 -*2

1 (discharge

into Big bay) (ΔS < 4psu)

bottom 160 (100)

105 -*2

150 (100)

surface -*2 50 -*2 < 0.50 0.20 < 0.50 SAWQ

(ΔS < 1psu) bottom 400 (200)

310 (200)

520 (400)

< 0.50 < 0.2 < 0.50

surface -*2 -*2 -*2

2 (discharge into Small

Bay) (ΔS < 4psu)bottom 120

(60) 100 -*2

160 (90)

surface -*2 -*2 -*2 < 0.05 < 0.05 < 0.05 SAWQ (ΔS < 1psu) bottom 50 50 50 < 0.10 < 0.10 < 0.20

surface -*2 -*2 -*2

3 (discharge

into Small Bay) (ΔS < 4psu)

bottom -*2 -*2 -*2

surface -*2 -*2 -*2 < 0.05*3 < 0.05*3 < 0.05*3

SAWQ (ΔS < 1psu) bottom 120

(70) 50

(-*2) 90

(50) < 0.05*3 < 0.05*3 < 0.05*3

surface -*2 -*2 -*2

3 (discharge

into Big Bay) (ΔS < 4psu)

bottom -*2 -*2 -*2

surface -*2 -*2 -*2 < 0.05*3 < 0.05*3 < 0.05*3 SAWQ

(ΔS < 1psu) bottom 40 (-*2) -*2 -*2 < 0.05*3 < 0.05*3 < 0.05*3

surface -*2 -*2 -*2

3 (discharge

at Caisson 3) (ΔS < 4psu)

bottom -*2 -*2 -*2

*1 SAWQ refers to the South African Water Quality guideline of 33 psu < S < 36 psu while the ΔS < 4 psu (i.e. salinity <

39 to 40 psu) guidelines is consistent with salinity thresholds where impacts are likely to be significant (see Section 8.1.3.1 for level of significance).

*2 The modelling undertaken cannot resolve detail within an approximate 50 m radius of the discharge. While the model results indicate no plume in the relevant water depth, a conservative approach would be to assume the possibility that any of the guidelines could be exceeded within a “sacrificial zone” of 50 m around the discharge point. This does not imply that the relevant guideline will be exceeded within this radius, but rather that it remains a possibility that this will be the case due to the limited resolution of the model used to simulate the plume behaviour.

*3 The minimum value resolved in the plots of the model results is 0.05 psu which is a negligible quantity in terms of efficiencies of the RO plant. For the alternate discharge into Bay Bay for Site 3, the separation if the intake and discharge is such that the potential elevation in salinity at the in take due to the brine discharge is negligible and certainly much less than 0.05 psu reported in the Table above.

In terms of the relative dimensions of the plume where relevant salinity guidelines are exceeded (see

Table 7.6 above and Figure 7.3 below), the various combinations of intake and discharge locations



may be ranked as follows (1 indicates the smallest plume dimensions and most likely environmentally

preferable option while 5 indicates the largest plume dimension and likely least desirable option in

terms of environmental impacts):

Site 1 2 3

discharge into Small Bay

3 discharge into

Big Bay

3 discharge at

Caisson 3 Ranking 4 5 1 3 1

Figure 7.3: Comparative maximum dimensions of the elevated salinity “footprint” (ΔS < 1 psu or S < 36 psu) for all discharge sites.


concessions


Existing mariculture

concessions



Temperature

The model results for elevated seawater temperature are summarised in Table 7.7 below. The dimensions of the plume indicated are the maximum dimension in any direction of the plume “footprint”, where the dimensions of the footprint are determined by the exceedance of selected water quality guidelines (1 ºC and ANZECC water quality guideline) for periods ranging from 6 hours in a season to approximately 5 days. In the table the figures roughly represent the dimensions of the maximum “footprint” of the effluent plume (the “footprint” where the relevant water quality guideline or target value is exceeded for no more than 6 hour per season), while the figure in brackets indicates the spatial dimensions of the effluent “footprint” that is exceeded for less than approximately 5 days in a season. The column indicating the maximum temperature elevation (ΔT) of the ambient waters at the intake due to the influence of the discharge plume does not inform the assessment of environmental impacts but rather indicates whether or not “re-circulation” effects are likely to be a concern at the intake, whereby intake waters are contaminated by the discharge plume waters.

Table 7.7: Summary of effluent plume dimensions around the discharge point (based on exceedances of the temperature water quality guidelines for cumulative periods of 6 hours and approximately 5 days), and the magnitude of the temperature elevation at the intake.

Max Discharge Plume Dimensions*2

(m)

Max ΔT at intake

(°C) Site WQ*1 guideline


Summer Winter Autumn Summer Winter Autumn

surface -*2 -*2 -*2 < 0.25 < 0.25 < 0.25 SAWQ (ΔT < 1 ºC) bottom 220

(50) 100*4

-*2 300

(170) < 0.25 < 0.25 < 0.25

surface -*2 -*2 -*2

1 (discharge

into Big Bay) ANZECC bottom -*2 -*2 120

surface -*2 50 -*2 < 0.50 < 0.50 < 0.75 SAWQ (ΔT < 1 ºC) bottom 650

(300) 300 -*2

650 (570) < 0.75 < 0.50 < 0.75

surface -*2 -*2 -*2


Bay) ANZECC bottom -*2 -*2 250

surface -*2 -*2 -*2 < 0.05 < 0.05 < 0.05 SAWQ (ΔT < 1 ºC) bottom 500

(280) 180 (-*2)

440 (250) < 0.50 < 0.25 < 0.75

surface -*2 -*2 -*2


Bay) ANZECC bottom -*2 -*2 -*2 surface -*2 -*2 -*2 < 0.25*3 < 0.25*3 < 0.25*3 SAWQ

(ΔT < 1 ºC) bottom 440 (240)

170 (110)

430 (240) < 0.25*3 < 0.25*3 < 0.25*3

surface -*2 -*2 -*2

3 (discharge

into Big Bay) ANZECC bottom -*2 -*2 -*2

surface -*2 -*2 -*2 < 0.25*3 < 0.25*3 < 0.25*3 SAWQ (ΔT < 1 ºC) bottom 440

(170) 180 (-*2)

240 (170) < 0.25*3 < 0.25*3 < 0.25*3

surface -*2 -*2 -*2

3 (discharge

at Caisson 3) ANZECC bottom -*2 -*2 -*2

*1 SAWQ refers to the South African Water Quality guideline of ΔT < 1 ºC while ANZECC refers to the

ANZECC (2000) guideline that the median temperature in the environment with an operational discharge should not lie outside the 20 an 80 percentile temperature values for a reference location or ambient temperatures observed prior to the construction and operation of the proposed discharge.



*2 The modelling undertaken cannot resolve detail within an approximate 50 m radius of the discharge. While the model results indicate no plume in the relevant water depth, a conservative approach would be to assume the possibility that any of the guidelines could be exceeded within a “sacrificial zone” of 50 m around the discharge point. This does not imply that the relevant guideline will be exceeded within this radius, but rather that it remains a possibility that this will be the case due to the limited resolution of the model used to simulate the plume behaviour.

*3 The minimum value resolved in the plots of the model results is 0.25 ºC which is a negligible quantity in terms of efficiencies of the RO plant. For the alternate discharge into Bay for Site 3, the separation if the intake and discharge is such that the potential elevation in temperature at the in take due to the brine discharge is negligible and certainly much less than 0.25 ºC reported in the Table above.

*4 The plume dimension is only 40 m in diameter but is located some 100 m distant of the discharge location.

Figure 7.4: Comparative maximum dimensions of the elevated temperature “footprint” (ΔT < 1 ºC) for all discharge sites.

In terms of the relative dimensions of the plume where relevant temperature guidelines are exceeded

(see Table 7.7 and Figure 7.4 above), the various combinations of discharge locations and locations

of the discharge may be ranked as follows (1 indicates the smallest plume dimensions and most


concessions



concessions



likely environmentally preferable option while 5 indicates the largest plume dimension and likely least

desirable option in terms of environmental impacts):

Site 1 2 3


3 discharge into

Big Bay

3 discharge at


Biocides

The model results for biocide concentrations are summarised in Table 7.8 below. The dimensions of

the plume indicated are the maximum dimension in any direction of the plume “footprint”, where the

dimensions of the footprint are determined by the exceedance of selected water quality guidelines

(free residual chlorine concentration < 3 µg.ℓ-1) for periods ranging from 6 hours in a season to

approximately 5 days. In the table the figures roughly represent the dimensions of the maximum

“footprint” of the effluent plume (the “footprint” where the relevant water quality guideline or target

value is exceeded for no more than 6 hour per season), while the figure in brackets indicates the

spatial dimensions of the effluent “footprint” that is exceeded for less than approximately 5 days in a

season.

Table 7.8: Summary of effluent plume dimensions around the discharge point (based on exceedances of the biocide water quality guidelines) for cumulative periods of 6 hours and approximately 5 days, respectively.

Max Discharge Plume Dimensions*2

(m) Site WQ*1 guideline


Summer Winter Autumn

surface -*2 -*2 -*2 1 (discharge into

Big Bay) ANZECC (< 3 µg.ℓ-1) bottom 300

(200) 210

(150) 400

(230) surface -*2 50 -*2 2

(discharge into Small Bay)

ANZECC (< 3 µg.ℓ-1) bottom 400

(250) 360

(270) 570

(470) surface -*2 -*2 -*2 3

(discharge into Small Bay)


(-*2) -*2 -*2


Big Bay) ANZECC (< 3 µg.ℓ-1) bottom 120

(-*2) 80

(-*2) 90

(-*2) surface -*2 -*2 -*2 3

(discharge at Caisson 3)


(-*2) -*2 -*2

*1 ANZECC refers to the ANZECC (2000) guideline that the biocide concentration (FRC) should not exceed

3 µg.ℓ-1 or the water quality guideline adopted for DBNPA, i.e. residual DBNPA concentrations in the effluent should not exceed 0.035 µg.ℓ-1 or 0.070 µg.ℓ-1, depending which target value is assumed to be the most appropriate given the possible residual concentrations of DBNPA in the discharge.

*2 The modelling undertaken cannot resolve detail within an approximate 50 m radius of the discharge. While the model results indicate no plume in the relevant water depth, a conservative approach would be to assume the possibility that any of the guidelines could be exceeded within a “sacrificial zone” of 50 m



around the discharge point. This does not imply that the relevant guideline will be exceeded within this radius, but rather that it remains a possibility that this will be the case due to the limited resolution of the model used to simulate the plume behaviour.

Figure 7.5: Comparative maximum dimensions of the biocide “footprint” (oxidising biocide concentration <3 µg.ℓ-1 or DBNPA residual concentrations less than the target values assumed to be appropriate) for all discharge sites.

In terms of the relative dimensions of the plume where relevant biocide guidelines are exceeded (see

Table 7.8 above and Figure 7.5 above), the various combinations of discharge locations and

locations of the discharge may be ranked as follows (1 indicates the smallest plume dimensions and

most likely environmentally preferable option while 4 indicates the largest plume dimension and likely

least desirable option in terms of environmental impacts):

Site 1 2 3


3 discharge into

Big Bay

3 discharge at



concessions



concessions



Achievable Dilutions (Co-discharges)

The model results for a “generic” pollutant reported in terms of achievable dilutions are summarised

in Tables 7.9 and 7.10 below. The dimensions of the plume indicated are the maximum dimension in

any direction of the plume “footprint”, where the dimensions of the footprint are determined by the

non-exceedance of selected achievable dilutions (50 times and 100 times dilution factor) for periods

ranging from 6 hours in a season to approximately 5 days.

In the tables the figures roughly represent the dimensions of the maximum “footprint” of the effluent

plume, i.e. the “footprint” where the required dilution of 50 times (Table 7.9) and 100 times (Table

7.10) are not achieved for a cumulative total duration of no more than 6 hours per season), while the

figure in brackets indicates the spatial dimensions of the effluent “footprint” where the required

dilution of 50 times (Table 7.9) and 100 times (Table 7.10) is not achieved for a cumulative total

duration of no more than approximately 5 days in a season.

Table 7.9: Summary of effluent plume dimensions around the discharge point (based on non-exceedances of a required dilution of 50 times) for cumulative periods of less than 6 hours and less than approximately 5 days, respectively.

Discharge Plume Dimensions*2




surface 80 (-*2) -*2 -*2 1

(discharge into Big Bay)

Dilution factor < 50 bottom 350

(200) 260

(180) 430

(360)

surface 240 (-*2)

150 (-*2)

150 (-*2)

2 (discharge into

Small Bay) Dilution factor

< 50 bottom 560 (330)

500 (390)

670 (500)


Small Bay)


(100) 50

(-*2) 150 (90)


Big Bay)


(140) 130 (80)

230 (160)

surface -*2 -*2 -*2 3 (discharge at

Caisson 3)


(130) 140 (90)

170 (140)

*1 The assumed generic water quality guideline assumed in the table above is that the achievable dilution of

any co-discharge needs to exceed 50 times. *2 The modelling undertaken cannot resolve detail within an approximate 50 m radius of the discharge.

While the model results indicate no plume in the relevant water depth, a conservative approach would be to assume the possibility that any of the guidelines could be exceeded within a “sacrificial zone” of 50 m around the discharge point. This does not imply that the relevant guideline will be exceeded within this radius, but rather that it remains a possibility that this will be the case due to the limited resolution of the model used to simulate the plume behaviour.



The required dilution target values of 50 and 100 times dilution are merely nominal conservative

required dilution that provides indicative results for potential co-discharges. The assumption here is

that the respective water quality guidelines will be sufficiently stringent for required dilutions of 50 to

100 to be necessary. The model outputs however can be re-processed assuming any specified

thresholds deemed to be representative of the pollutant of concern. In that sense the modelling

approach utilised here is entirely generic and scalable. The footprints in Figure 7.6 are based on 50

times dilution contours.

Table 7.10: Summary of effluent plume dimensions around the discharge point (based on

non-exceedances of a required dilution of 100 times) for cumulative periods of less than 6 hours and less than approximately 5 days, respectively.

Discharge Plume Dimensions*1




surface -*1 280 (-*2)

210 (60) 1 Dilution factor

< 100 bottom 500 (260)

400 (220)

580 (470)

surface 550 (380)

520 (190)

460 (260) 2 Dilution factor

< 100 bottom 750 (450)

800 (550)

830 (700)

surface -*1 -*1 -*1 3 (discharge

into Small Bay)


(280) 190

(140) 400

(260)


into Big Bay)


(340) 280

(240) 460

(310)


at Caisson 3)


(180) 280

(220) 320

(260)

*1 The modelling undertaken cannot resolve detail within an approximate 50 m radius of the discharge.

While the model results indicate no plume in the relevant water depth, a conservative approach would be to assume the possibility that any of the guidelines could be exceeded within a “sacrificial zone” of 50 m around the discharge point. This does not imply that the relevant guideline will be exceeded within this radius, but rather that it remains a possibility that this will be the case due to the limited resolution of the model used to simulate the plume behaviour.



In terms of the relative dimensions of the plume where there is a non-exceedance of a required

dilution of 50 or 100 (see plume dimensions in Tables 7.9 and 7.10 and Figure 7.6 below), the

various combinations of discharge locations and locations of the discharge may be ranked as follows

(1 indicates the smallest plume dimensions and most likely environmentally preferable option while 5

indicates the largest plume dimension and likely least desirable option in terms of environmental

impacts):

Site 1 2 3


3 discharge into

Big Bay

3 discharge at


Figure 7.6: Comparative maximum dimensions of the plume “footprint” (achievable dilution < 50) for all discharge sites.


concessions



concessions






8 ASSESSMENT OF ENVIRONMENTAL IMPACTS

8.1 Identification of Potential Environmental Impacts

The anticipated environmental impacts of potential concern to the marine environment from the

proposed Reverse Osmosis Desalination plant in Saldanha Bay identified and summarised in Section

6 primarily include those associated with construction activities in the coastal zone and ongoing

discharge of brine to the marine environment. These are described in more detail below.

8.1.1 Construction of Intake and Discharge Structures

Pipelines The use of intake and discharge pipelines in the engineering designs for the RO Plant will most likely

involve considerable disturbance of the high shore, intertidal and shallow subtidal beach habitats

during the construction and installation process. In the absence of engineering specifications

provided by the client, the following assessment will assume that HDPE pipelines (or similar) not

exceeding a diameter of 0.36 m will be used for both intakes and discharges.

Individual pipeline sections of 10 - 24 m are usually fabricated by the supplier and transported to the

site where they are subsequently butt-welded together into long strings. This will require a sufficiently

large and relatively flat onshore area where the pipes can be stockpiled and prepared. However

potential impacts associated with this construction area will not be further assessed here as it will

most likely either be located within the existing port area or in the dunes and well above the high

water mark (Site 1).

Depending on the oceanographic conditions on site, concrete weight collars may then be placed

around the strings to provide stability on the seabed. The pipe sections are dragged down the shore

by dozer and capped at either end before being floated out to sea by means of a tug. The air is

released from the pipe, and as it fills with water it sinks to the bottom. Depending on the required

overall length of the pipeline, the different strings making up the total pipeline are placed one behind

the other and connected by means of spool pieces. This is usually undertaken by commercial divers.

Special attention has to be paid to nearshore beach crossings where currents and breaking waves

may prevail. To avoid exposure on the beach and nearshore area, and to avoid damage by wave

forces in the surf-zone, the pipeline needs to be buried below the seabed. This usually requires the

construction of a temporary jetty to provide a stable work platform from which a trench (protected

between rows of sheetpiles) can be excavated. Excavation to a suitable depth to accommodate the

pipeline may potentially require blasting. However, at this stage the use of blasting is not part of the

project description and impacts associated with blasting are thus not assessed in this study. If

blasting becomes an option for trench-excavation a separate assessment is necessary. The pipe is

then placed in the trench and subsequently buried by earth-moving machinery. Further offshore, the

pipe is left lying on the seabed, and with time will settle into the sandy substrate.



Pipeline launching and entrenchment will involve extensive traffic on the beach (site 1 and 2) by

heavy vehicles and machinery, as well as the potential for hydrocarbon spills. Although the activities

on the intertidal beach will be localised and confined to within a hundred metres of the construction

site, the beach sediments will be completely turned over in the process and the associated

macrofauna will almost certainly be entirely eliminated. Any shorebirds feeding and/or roosting in the

area will also be disturbed and displaced for the duration of construction activities. The invertebrate

macrofaunal species inhabiting these beaches are all important components of the detritus / beach-

cast seaweed-based food chains, being mostly scavengers, particulate organic matter and filter-

feeders (Brown and McLachlan, 1994). As such, they assimilate food sources available from the

detritus accumulations typical of this coast and, in turn, become prey for surf-zone fishes and

migratory shorebirds that feed on the beach slope and in the swash zone. By providing energy input

to higher trophic levels, they are all important in nearshore nutrient cycling, and the reduction or loss

of these macrophyte assemblages may therefore have cascade effects through the coastal

ecosystem (Dugan et al., 2003).

Jetty construction (if necessary) will most likely require the use of pile drivers, and trench excavation

will necessitate the removal of significant volumes of beach sand (i.e. a trench width of approximately

1 m width and the length of the pipeline from the beach crossing to the intake or discharge point. This

trench material is typically placed on a floating pontoon ready for backfill. This excavation process

will result in increased suspended sediments in the water column, potentially affecting light

penetration and thus phytoplankton productivity and algal growth. The impact of the sediment plume,

however, is expected to be relatively localised and of short duration (only for the duration of the

trench excavation). Settlement of the suspended material on adjacent areas of seabed may result in

smothering of the resident biota. However, some mobile benthic animals are capable of migrating

vertically through more than 30 cm of deposited sediment (Maurer et al., 1981a,b; 1982). The biota

that inhabited the excavated sediments will most likely perish. Once the pipeline has been laid and

sufficient sediments have accumulated around the pipe, the affected seabed areas with time will be

recolonised by benthic macrofauna.

Provided the construction activities are all conducted concurrently, the duration of the disturbance

should be limited to a few months. Studies on the disturbance of beach macrofauna communities on

the West Coast by beach mining activities have ascertained that, provided physical changes to beach

morphology are kept to a minimum, and sediment characteristics on the beach are not severely

altered, biological "recovery" of disturbed areas will occur within 2-5 years (Nel et al., 2003, 2004).

Disturbed subtidal communities within the wave base (<40 m water depth) might recover even faster

(Newell et al., 1998). Recovery of beach macrofaunal assemblages occurs primarily through

immigration from adjacent areas. Mitigation measures should therefore include rehabilitation of the

disturbed area immediately following construction, by removing all artificial constructions or beach

modifications created during construction from above and within the intertidal zone after completion



of construction activities. No accumulations of excavated beach sediments should be left above the

high water mark, and any substantial sediment accumulations below the high water mark should be

levelled.

Furthermore, an adjacent portion of undisturbed beach should be allocated where populations of

macrofaunal species can survive and supplement recolonisation in impacted areas.

Beach Wells The installation of intake beach wells will involve substantial disturbance of the high and mid-shore

beach habitats during the construction phase. The sinking of wells requires the use of a drilling rig,

which excavates the sediments to the required depth, and sleeves the cylindrical pit with PVC or

stainless steel to prevent the walls from collapsing. The stainless steel well is subsequently sunk into

the prepared excavation. Drainage pumps are most likely required to keep the pit free of water

during construction. The water is likely to be pumped across the beach into the surf-zone resulting in

localised inputs of fresh/brackish water and possibly increased suspended sediments in the water

column. This may potentially affect light penetration and thus phytoplankton productivity and algal

growth. The associated impacts are, however, expected to be relatively localised and of short

duration (only for the duration of the well construction).

Well construction will involve extensive traffic by heavy vehicles and machinery on the upper and

mid-sections of beach, as well as the potential for hydrocarbon spills. These activities will be

relatively localised and confined to within a few hundred metres of the construction site, although

their extent will ultimately depend on the number of wells to be installed. Dune vegetation and

associated fauna at the excavation sites will almost certainly be entirely eliminated, and high and

mid-shore beach macrofauna severely disturbed.

Boreholes on the Causeway The intakes boreholes proposed to be located along the causeway will be drilled on existing

reclaimed areas and therefore are expected to have negligible impacts on the marine environment.

8.1.2 Permanent Intake and Discharge Structures Intake of water directly from the ocean usually results in loss of marine species as a result of

impingement and entrainment. Impingement refers to injury or mortality of larger organisms (e.g. fish)

that collide with and are trapped by intake screens, and entrainment refers to smaller organisms that

slip through the screens and are taken into the plant with the feed water. Entrained material includes

holoplanktonic organisms (permanent members of the plankton, such as copepods, diatoms and

bacteria) and meroplanktonic organisms (temporary members of the plankton, such as juvenile

shrimps and the planktonic eggs and larvae of invertebrates and fish). Most studies and findings

related to entrainment have been done in association with the effects of power plant once-through

cooling systems. In these systems, entrained organisms are killed or injured due to the high pressure



or temperatures, and in the case of RO desalination plants, when water is forced against the filters or

membranes. While some studies consider a 100% mortality rate of entrained organisms in power

plant cooling systems (California Coastal Commission, 2004), other study results suggest that the

majority of the individuals would survive passage through such a system with mortalities ranging from

10 - 20% (Bamber and Seaby, 2004). These authors noted, however, that the rate of survival is

species-specific and that generalizations from the responses of one species to those of another are

not valid. It is likely that when compared to once-through cooling systems, mortality rates in RO

plants are greater (potentially 100%) since the seawater is forced at high pressures through filters or

membranes to remove particles, including the small organisms that are taken in with the feed-water.

While the significance of both impingement and entrainment is related to the location of an intake, the

first is primarily a function of intake velocity, and the latter is more related to the overall volume of

water drawn into the RO plant. Impingement can be mitigated through structural or operational

designs to open water intakes (see Section 9.1.1). The project description specifies that the intakes

will be designed to minimise impingement of biota and will have self-cleaning screens.

Transnet Projects has indicated an estimated 0.15 m.s-1 velocity at the intake screens and 1.0 m.s-1

in the intake pipelines. The intake volumes (i.e. 8000 m3.day-1 or 92.6 ℓ.s-1) are relatively low and the

effect of entrainment is thus assessed as minor, especially if care is taken to ensure that the intake

design minimises potential entrainment impacts.

The potential of scouring of sediment around the discharge outlet is a serious design issue for an

effluent discharging-system discharging into a shallow receiving water body (Carter and van

Ballegooyen, 1998). However, the discharge volumes being considered here (i.e. 4400 m3.day-1 or

50.9 ℓ.s-1) are low and the configuration being assessed (i.e. single or multi-port diffusers jetting water

upwards into the water column) is such that the potential impacts on bottom sediments are limited.

Should such impacts occur they will be confined to the immediate vicinity of the discharge point.

The location of a discharge or intake pipeline on the shoreline, or even the shore-crossing of the

pipeline may distort sediment transport pathways in the nearshore environment (and may even

extend to the distortion of aeolian sand transport pathways if the infrastructure associated with the

discharge is located in, or extends through, the mid- and upper shore) and consequently will alter the

natural environment to some degree. This will be limited should the pipeline be trenched through the

shore-crossing. The proposed expansion of the iron-ore export facilities may further pose a problem

for the location of intake/discharge structures on the shoreline (Site 1) as accretion or erosion of the

beach may occur (Smith et al., 2007), resulting in damage or dysfunction of the intake/discharge

structures unless these considerations are included in the engineering designs.

The use of intake beach wells/boreholes provides a natural pre-filter system, resulting in fewer

chemicals required for the pre-treatment process. Beach wells also reduce issues relating to

impingement and entrainment, as well as diminishing potentially harmful algal bloom effects on the

water quality of the intake waters. This intake option as a pre-treatment strategy is thus attractive



because of the reduction in chemical use and the potentially lower operation and maintenance costs

compared to other pre-treatment options, including media or cartridge filtration (California

Desalination Task Force, 2003a, Campbell and Jones, 2005). The use of traditional vertical beach

wells, however, is limited to smaller systems due to the large number of wells that would need to be

drilled to fulfil the pre-treatment requirements (i.e. to meet the intake volumes required).

Beach wells should only be used in areas where the impact on aquifers has been studied and saltwater intrusion of freshwater aquifers will not occur (Younos, 2005). Although such a

situation is unlikely if the plant is situated near the seashore, special care needs to be taken to

ensure that such a risk is minimised. For example, intake beach wells for a desalination plant on the

KwaZulu-Natal north-coast situated on the beach drew fresh water from the dune area instead of

from the sea, causing serious destabilisation problems for the sensitive dune vegetation (DWAF,

2007). These potential impacts (saltwater intrusion and effects on groundwater) are not considered

to be part of the scope of this assessment.

To determine the efficacy of beach wells/boreholes for a particular site, exploratory drilling to

determine the aquifer characteristics (depth, strata) and pump tests on boreholes should be

conducted. A preliminary idea of required depth, potential discharge and number of required beach

wells/boreholes in the site area can be obtained to match the demands of the proposed plant.

Spacing between wells/boreholes at the proposed withdrawal rates is another important factor in

design planning. Surface seawater RO systems operate at a conservative flux of 7 to 9 gfd (gallons

per square foot per day). Beach well systems can be operated at 9 -10 gfd. The difference in flux

rate results from reductions in the membrane-fouling rate due to the better water quality obtained

from beach wells (California Desalination Task Force, 2003a). The results from the geotechnical

groundwater study have shown that by installing large diameter (300 – 500 mm) boreholes equipped

with continuous slot-screens the borehole yields can reach at least 10 L/s (Visser et al., 2007).

Numerical modelling indicated that ~10 vertical intake wells drilled 50 m apart parallel to the shoreline

above the high water mark at Site 1 would be required to abstract the anticipated raw water demand

of 8000 m3/day for the RO plant. Alternatively, if horizontal collector wells are decided on, at least

two with ~100 m of horizontal collector pipes each would be required. Although no similar detailed

studies have taken place to date for the proposed beach well intakes at Site 2, up to 10 intake wells,

spaced ~50 m apart have been proposed for this site. Based on yield information to date between 5

and 10 boreholes will need to be located along the causeway (D. Visser, SRK, pers. comm.) to

provide the requisite intake waters (Site 3).

As noted above, the proposed expansion of the iron-ore export facilities may further pose a problem

for the location of beach well discharge structures on the shoreline (Site 1). Studies indicate that the

accretion in the vicinity of Site 1 will be slightly less than occurs at present (Smith et al., 2007).

These changes in shoreline may result in damage or dysfunction of the discharge structures unless

these considerations are included in the engineering designs.



8.1.3 RO Plant Effluents The effluent water discharged from the desalination plant will constitute a high-salinity brine

(expected salinity 63.5 psu) that has been treated with a biocide, and which contains other chemical

residuals from RO membrane cleaning processes. Under current design specifications, the feed-

waters will be drawn from ~ - 1 m CD or approximately 1.9 m below mean sea level (except for the

beach well intake scenario at Site 1 and boreholes at site 3), and consequently from above the

thermocline present in the bay. The level of nitrate in these waters is assumed to be low if not

negligible, and will thus not be discussed further. Although storage of the feed-water in the buffer

tank prior to it entering the RO plant may potentially result in a slight elevation in temperature, this

increase is assumed to be negligible. (It should, however, be noted that a similar plant on the

Bushman’s River in the Eastern Cape discharged a brine more than 2ºC warmer than the seawater

temperature in the receiving environment.) For the purpose of this assessment it is thus assumed

that the discharged water will not be significantly warmer in relation to the temperature of the intake

waters (see Section 4.2 for a more detailed discussion of this assumption) However, it is possible

that the receiving water masses may potentially have lower temperatures than that of the brine

effluent, as it is expected that the high density brine will remain at the same temperature as the near

surface intake waters but will sink to the bottom in the vicinity of the discharge location (see Section

7.4). In Saldanha Bay these bottom waters may be significantly colder than the near surface waters

being drawn into the intake.

The approach taken in this study is consistent with the approach taken by the decision-making

authorities. Based on identified sensitive ecosystems and beneficial use areas and the likely

environmental impacts of the discharge, water and sediment quality target values have been

specified comprising general recommended targets as set out in water quality guideline documents,

such as those of South Africa, Australia/New Zealand and the World Bank (RSA DWAF, 1995;

ANZECC, 2000; World Bank, 1998). In the SBQWQFT Water Quality Management Plan, Monteiro

and Kemp (2004) have proposed the three ecosystem state variables (over and above the specified

water quality criteria in Table 8.1) for assessing general thresholds of ecosystem response in

Saldanha Bay. These are phytoplankton biomass, as mg C/m3, dissolved oxygen as ml O2 /ℓ and

particulate organic carbon (POC) in deposition areas as mg C/m2. These reflect the overall

productivity of the Saldanha Bay system and its biogeochemical status13.

13 Threshold levels for these variables are set at changes of 10% (or more) over the ambient condition that extend for a

period of 7 days or longer; the purpose being to be able to reliably show whether there is any net change in the system due to the proposed discharge. The 10% change is predicated on typical measurement ‘uncertainty’ for field investigations of the distributions of the variables considered (Monteiro and Kemp, 2004). The 7 day period derives from measurements showing that exposures of organisms to e.g. increased or decreased nutrient concentrations in the case of Saldanha Bay macroalgae, need to persist for at least this period prior to there being any measurable effect in biomass or physiological function. An additional reason is that Saldanha Bay is exposed to effects of upwelling on the adjacent continental shelf which, in spring/summer, typically has a 6-10 day periodicity. Thus effects that persist for periods shorter than the upwelling cycle are considered to be transient in nature; effects persisting for longer than 7 days may run across two or more upwelling cycles and may become cumulative. (For example, there was a drop in primary production in December 1999 for three weeks that resulted in mussels stopping growing for this period.) Under this system, if the analyses of the proposed discharge indicate that the identified system indicator levels are be exceeded the potential ecological consequences would need to be assessed in more detail, i.e. the exceedance of a 10% change for more that 7 days acts as a “red flag”.



However given the scale of the discharge and associated “footprint” or impacted area is relatively

modest, it is deemed that a detailed study, where bay productivity, oxygen concentrations and

nutrient dynamics are explicitly simulated, is not warranted here. The present assessment is based

on three-dimensional modelling of plume extent followed by an ecological risk assessment as

described in Section 4.1.2 and 4.1.3).

For the purpose of this assessment the dimensions of the discharge plume in the receiving

environment are defined here by the spatial extent of exceedance of established water quality

guidelines or target values for constituents of concern. Table 8.1 provides recommended target

values derived from the national and international guidelines and pertaining to the proposed

hypersaline discharge.

As guideline or target values for the non-oxidising biocide dibromonitrilopropionamide (DBNPA)

proposed for use in the RO Plant could not be located in either the South African Water Quality

guidelines (DWAF, 1995) nor the ANZECC guidelines (ANZECC, 2000), a literature search was

undertaken to locate a suitable water quality guideline. The most sensitive marine species for which

toxicological data exists is the Eastern Oyster (Crassostrea virginica). A 48 hour exposure LC50

(median lethal concentration) of 0.72 mg.ℓ-1 of DBNPA has been reported (Kaine et al., 1996; Dow

Chemical Fact Sheet No. 253-01464-06/18/02, 34pp.) where the LC50 refers to the percentage of

abnormally developed Eastern Oyster larvae and not mortality as is normally the case. Klaine et al. (1996) also report an 96-hour flow through study performed with Eastern Oyster that resulted in

significant reduction of shell deposition at < 0.07 mg.ℓ-1 that was considered to represent a Lowest

Observed Effect Concentration (LOEC) for Eastern Oyster. US EPA (1994) report Levels of Concern

(LOC)14 for Eastern oysters of 0.035 mg. ℓ-1 (typical species) and 0.004 mg/l (endangered species).

In this study appropriate water quality guidelines are assumed to range between 0.035 mg. ℓ-1 and

0.070 mg. ℓ-1 for DBNPA in the brine discharge.

14 Typical species LOC values are defined as ½ LC50, while LOC values for endangered species are defined as ½0 LC50.



Table 8.1: Water quality guidelines for the discharge of a high-salinity brine into the marine environment.

VARIABLE SOUTH AFRICA (DWAF 1995)

AUSTRALIA/NEW ZEALAND

(ANZECC 2000) WORLD BANK a

(World Bank 1998)

Zone of impact /

mixing zone

To be kept to a minimum, the acceptable dimensions of this zone informed by the

EIA and requirements of licensing authorities, based

on scientific evidence.

- 100 m radius from point

of discharge for temperature

Temperature

b The maximum acceptable

variation in ambient temperature is ± 1°C

Where an appropriate reference system(s) is

available, and there are sufficient resources to collect the necessary information for

the reference system, the median (or mean) temperature

should lie within the range defined by the 20%ile and

80%ile of the seasonal distribution of the ambient

temperature for the reference system,.

< 3°C above ambient at the edge of the zone

where initial mixing and dilution take place.

Where the zone is not defined, use 100

meters from the point of discharge when there

are no sensitive aquatic ecosystems within this

distance.

Salinity b 33 – 36 psu

Low-risk trigger concentrations for salinity are that the median

(or mean) salinity should lie within the 20%ile and 80%ile

of the ambient salinity distribution in the reference system(s). The old salinity

guideline (ANZECC 1992) was that the salinity change should

be < 5% of the ambient salinity.

-

Total residual

Chlorine d

no guideline, however deleterious effects recorded for concentrations as low as

2 – 20 μg. ℓ-1. A very conservative trigger value

thus is < 2 μg CL . ℓ-1.

3 µg Cl. ℓ-1 measured as total residual chlorine (low reliability trigger value at 95% protection

level, to be used only as an indicative interim working level) (ANZECC 2000) c

0.2 mg. ℓ-1 at the point of discharge prior to

dilution

Total residual DBNPA

No guideline exists, suggest values ranging between

0.035 mg. ℓ-1 and 0.070 mg. ℓ-1No guideline found No guideline found

Dissolved oxygen

For the west coast, the dissolved oxygen should not

fall below 10 % of the established natural variation.

For the south and east coasts the dissolved oxygen should not fall below 5 mg/ℓ

(99 % of the time) and below 6 mg/ℓ (95 % of the time)

Where an appropriate reference system(s) is

available, and there are sufficient resources to collect the necessary information for

the reference system, the median lowest diurnal DO

concentration for the period for DO should be > the 20%ile of the ambient dissolved oxygen concentration in the reference system(s) distribution. Where

possible the trigger value should be obtained during low

flow and high temperature periods when DO

concentrations are likely to be at their lowest.

-



VARIABLE SOUTH AFRICA (DWAF 1995)

AUSTRALIA/NEW ZEALAND

(ANZECC 2000) WORLD BANK a

(World Bank 1998)

Nutrients

Waters should not contain concentrations of dissolved nutrients that are capable of

causing excessive or nuisance growth of algae or

other aquatic plants or reducing dissolved oxygen concentrations below the target range indicated for

dissolved oxygen (see above)

Default trigger values of PO4-P: 100 µg. ℓ-1 NOx-N: 50 µg. ℓ-1 NH4

+-N: 50 µg. ℓ-1 for the low rainfall southern Australian region (Table 3.3.8 in ANZECC 2000)

-

Chromium 8 μg. ℓ-1 (as total Cr)

Marine moderate reliability trigger value for chromium (III)

of 10 μg. ℓ-1 with 95% protection

Marine high reliability trigger

value for chromium (VI) of 4.4 μg. ℓ-1 at 95% protection.

0.5 mg. ℓ-1 (total Cr) for effluents from thermal

power plants

Iron -

Insufficient data to derive a reliable trigger value. The current Canadian guideline

level is 300 μg. ℓ-1

1.0 mg. ℓ-1 for effluents from thermal power

plants

Molybdenum -

Insufficient data to derive a marine trigger value for

molybdenum. A low reliability trigger value of 23 μg. ℓ-1 was

adopted to be used as indicative interim working

levels.

-

Nickel 25 μg. ℓ-1 (as total Ni)

7 μg. ℓ-1 at a 99% protection level is recommended for

slightly-moderately disturbed marine systems.

a The World Bank guidelines are based on maximum permissible concentrations at the point of discharge and do not explicitly take into account the receiving environment, i.e. no cognisance is taken of the fact of the differences in transport and fate of pollutants between, for example, a surf-zone, estuary or coastal embayment with poor flushing characteristics and an open and exposed coastline. It is for this reason that we include in this study other generally accepted Water Quality guidelines that take the nature of the receiving environment into account.

b Both in the case of temperature and salinity using the maximum ΔT and ΔS measured at any location at any time during the model simulations constitutes an extremely conservative approach. We nevertheless have used the dimension of the 99% exceedance of the ΔT = +1ºC contour and the ΔS = 36 psu contour as representing the “footprint” of the potential impact. This constitutes a total of approximately 6 hours per season (or one day per annum) of exposure to conditions exceeding the stated ΔT or ΔS values.

c The ANZECC (2000) Water Quality guideline for salinity is less stringent than, but roughly approximates, the South African Water Quality guideline that requires that salinity should remain within the range of 33 psu to 36 psu. Similarly the ANZECC (2000) Water Quality guideline for water temperature is less stringent than the South African Water Quality guideline that requires that water temperature does not vary by more than 1 ºC from the ambient water temperature. The ANZECC (2000) Water Quality guideline for water temperature is likely to be particularly relevant in the bottom waters of Saldanha Bay where there is significant natural temperature variability due to upwelling cycles as the ANZECC guideline explicitly takes into account this variability and the fact that the marine ecology of the region is likely to be adapted to such variability. At these depths the South African Water Quality guideline of temperature variability of < 1 ºC is likely to be very conservative (possibly overly conservative) in these circumstances.

d Chlorine “shocking” may be preferable in certain circumstances. This involves using high chlorine levels for a few seconds rather than a continuous low-level release. In this case the target value is a maximum value of 2 mg. ℓ-1 for up to 2 hours, not to be repeated more frequently than once in 24 hours, with a 24-hour average of 0.2 mg. ℓ-1 (The same limits would apply to bromine and fluorine.).

The individual constituents of the effluent and their potential impacts on the marine environment are

discussed in more detail under separate headings below.



8.1.3.1 Salinity All marine organisms have a range of tolerance to salinity, which is related to their ability to regulate

the osmotic balance of their individual cells and organs to maintain positive turgor pressure. Aquatic

organisms are commonly classified in relation to their range of tolerance as stenohaline (able to

adapt to only a narrow range of salinities) or euryhaline (able to adapt to a wide salinity range), with

most organisms being stenohaline.

Salinity changes may affect aquatic organisms in two ways:

• direct toxicity through physiological changes (particularly osmoregulation), and

• indirectly by modifying the species distribution.

Salinity changes can also cause changes to water column structure (e.g. stratification) and water

chemistry (e.g. dissolved oxygen saturation and turbidity). For example, fluctuation in the salinity

regime has the potential to influence dissolved oxygen concentrations, and changes in the

stratification could result in changes in the distribution of organisms in the water column and

sediments. Significant effects on stratification and oxygen concentrations (if at all) are only expected

in the very near-field. Behavioural responses to changes in salinity regime can include avoidance by

mobile animals, such as fish and macrocrustaceans, by moving away from adverse salinity and

avoidance by sessile animals by reducing contact with the water by closing shells or by retreating

deeper into sediments.

However, in marine ecosystems adverse effects or changes in species distribution are anticipated

more from a reduction rather than an increase in salinity (ANZECC, 2000). Very little information

exists on the effect of an increase in salinity on organisms in coastal marine systems, most studies

being done either on effects of a decline in salinity due to an influx of freshwater, or on salinity

fluctuations in estuarine environments, where most of the fauna can be expected to be of the

euryhaline type.

Sub-lethal effects of changed salinity regimes (or salinity stress) can include modification of metabolic

rate, change in activity patterns or alteration of growth rates (McLusky, 1981). The limited data

available include a reported tolerance of adults of the mussel Mytilus edulis of up to 60 psu (Barnabe,

1989), and successful fertilization (Clark, 1992) and development (Bayne, 1965) of its larvae at a

salinity of up to 40 psu. The alga Gracilaria verrucosa can tolerate salinity ranges from 9-45 psu

(Engledow and Bolton, 1992). The shrimp Penaeus indicus was capable of tolerating a salinity range

of 1 to 75 psu if allowed an acclimation time of around 48 hours (McClurg, 1974), the oyster

Crassostrea gigas tolerated salinities as high as 44 psu (King, 1977), and the shrimp Penaeus

monodon survived in 40 psu saline water (Kungvankij, et al. 1986a, b, cited in DWAF 1995). Chen et

al. (1992) reported a higher moulting frequency in juveniles of the prawn Penaeus chinensis at a



salinity of 40 psu. Lethal effects were reported for seagrass species: for example, salinities of 50 psu

caused 100% mortality of the Mediterranean seagrass Posidonia oceanica, 50% mortality at 45 psu,

and 27% at 40 psu. Salinity concentrations above 40 psu also stunted plant growth and no-growth

occurred at levels exceeding 48 psu (Latorre, 2005). The high saline concentration can also lead to

an increase of water turbidity, which is likely to reduce light penetration, an effect that might disrupt

photosynthetic processes (Miri and Chouikhi, 2005). The increased salt concentration can reduce

the production of plankton, particularly of invertebrate and fish larvae (Miri and Chouikhi., 2005). One

of the main factors of a change in salinity is its influence on osmoregulation, which in turn affects

uptake rates of chemical or toxins. In a review on the effects of multiple stressors on aquatic

organisms Heugens et al. (2001) summarize that in general metal toxicity increases with decreasing

salinity, while the toxicity of organophosphate insecticides increases with increasing salinity. For

other chemicals no clear relationship between toxicity and salinity was observed. Some evidence,

however, also exists for an increase in uptake of certain trace metals with an increase in salinity

(Roast et al., 2002, Rainbow and Black, 2002).

Very few ecological studies have been undertaken to examine the effects of high salinity discharges

from desalination plants on the receiving communities. One example is a study on the macrobenthic

community inhabiting the sandy substratum off the coast of Blanes in Spain (Raventos et al., 2006).

The brine discharge from the desalination plant was approximately 12 Hm3/year (or 33 707 m3. day-

1). Visual census of the macrobenthic communities were carried out at two control locations (away

from the discharge outlet) and one impacted (at the discharge outlet) location several times before

and after the plant began operating. No significant variations attributable to the brine discharges from

the desalination plant were found. This was partly attributed to the high natural variability that is a

characteristic feature of seabeds of this type, and also to the rapid dilution of the hypersaline brine

upon leaving the discharge pipe. Other studies, however, indicated that brine discharges have led to

reductions in fish populations, and to die-offs of plankton and coral in the Red Sea (Mabrook, 1994),

and to mortalities in mangrove and marine angiosperms in the Ras Hanjurah lagoon in the United

Arab Emirates (Vries et al., 1997).

The South African Water Quality guidelines (DWAF, 1995) set an upper target value for salinity of 36

psu. The paucity of information on the effects of increased salinity on marine organisms makes an

assessment of the high salinity plume difficult. However, this guideline seems sufficiently

conservative to suggest that no adverse effects should occur for salinity < 36 psu. At levels

exceeding 40 psu, however, significant effects are expected, including possible disruptions to

molluscan bivalves (e.g. mussels/oysters/clams) and crustacean (and possibly fish) recruitment as

salinities >40 psu may affect larval survival (e.g. Bayne, 1965; Clarke, 1992). This applies

particularly to the larval stages of fishes and benthic organisms in the area, which are likely to be

damaged or suffer mortality due to osmotic effects, particularly if the encounter with the discharge

effluent is sudden.



The model results have shown that the brine sinks to the bottom due to its greater density and is

primarily affecting the bottom third of the water column. The surface waters are not affected by the

brine.

An assessment of the effect of the brine discharge in terms of the modelling results, and for each

Alternative Site, is provided below. As a conservative measure, the assessment is based on the

‘worst-case-scenario’, which for all three sites almost exclusively appears to be during the extremely

calm weather scenario in autumn when the impacts in the bottom waters are the greatest due to

limited vertical mixing of surface and bottom waters due to primarily wind but also waves in shallow

waters.

Site 1:

At Site 1 the discharge is planned to occur via a pipeline along a 30 m length of the revetment of the

reclaim dam at a location approximately 80 m from the low water mark of the adjacent shoreline

where the water depth adjacent to the revetment is ~ 1.5 m.

The model results for the pipeline discharge indicate that the maximum extent of exceedance of the

36 psu target value is 200 m by 400 m from the brine outlet (Figure C1.2f, Appendix C), but this is

expected to occur only for a total of 0.25% (i.e. a cumulative 6 hours in a 90 day autumn-period) of

the calm weather period. Within a 90 x 50 m area around the outlet, the target value will be

exceeded >50% of the time (i.e. a cumulative total of 45 days in a 90 day autumn-period). The

maximum extent of the footprint 4 psu target value is approximately 160 m x 60 m (Figure C1.3b).

During periods of little or no stratification (e.g. winter) the plume footprint is significantly smaller due

to greater mixing of the plume throughout the vertical extent of the water column. Due to the

relatively small footprint of the plume, even under the ‘worst-case-scenario’, this impact is assessed

to be of low significance.

Site 2:

The discharge alternatives at Site 2 constitute either a surf-zone discharge or a single port diffuser at

between -0.5 m to -1 m below chart datum and 20-25 m from the shoreline. From the modelling

study it can be expected that the plume will have a maximum extent of 250 x 520 m around the

discharge outlet for a total of 0.25% (or a total duration of approximately 6 hours) of the calm weather

period (Figure C2.1f). Within 100 m of the outlet, the target value of 36 psu will be exceeded for

>50% of the time in autumn. The maximum extent of the 4 psu target value is approximately 150 m x

160 m (Figure C2.3f).

Again, during winter the plume footprint is notably smaller due to greater mixing of the plume

throughout the vertical extent of the water column. The impact is regarded as having a low

significance due to the relatively small plume dimensions. However it should be recognised that the

discharge for Site 2 is occurring into the relatively poorly-flushed Small Bay system and would



inherently pose greater risks than the other sites considered. It could also place significant

constraints on any future increases in plant capacity should they ever be required.

Site 3:

The discharge alternatives at Site 3 constitute a single port diffuser at either 1) approximately -8 m

CD into Small Bay, 2) at -4 m CD into Big Bay, or 3) at approximately -16 to -18 m CD at Caisson 3.

These will be located alongside the iron-ore causeway at distances ranging from approximately 50 to

70 m from the causeway. In the modelling study the discharge alternatives are defined as Site 3(a)

(discharge into Small Bay), Site 3(b) (discharge into Big Bay) and Site 3(c&d) (discharge at Caisson

3).

1) In the case of a discharge into Small Bay (Site 3a), the model results indicate that even for

the worst-case-scenario the 36 psu guideline value will only be exceeded for <50 m (i.e. the

assumed resolution of the modelling study) around the outlet structure for 0.25% of the time

(i.e. a cumulative total of 6 hours within a season). Given the limitations in the resolution of

the modelling study, a conservative assumption would be to assume that the 36 psu

guideline value could be exceeded regularly within this radius although this is not necessarily

the case.

2) The modelling results indicate that the discharge into Big Bay (Site 3b) will result in a

maximum plume dimension of approximately 120 m in diameter at 0.25% of the time. The

impact is regarded as having a low significance due to the relatively small plume dimensions

3) For discharge at Caisson 3 (Site 3c&d) the model results are similar but slightly less than

those indicated for a discharge into Small Bay (Site 3a). The model results indicate that even

for the worst-case-scenario the 36 psu guideline value will only be exceeded for <50 m (i.e.

the assumed resolution of the modelling study) around the outlet structure for 0.25% of the

time (i.e. a cumulative total of 6 hours within a season). Given the limitations in the

resolution of the modelling study, a conservative assumption would be to assume that the 36

psu guideline value could be exceeded regularly within this radius although this is not

necessarily the case. The potential mitigating effects of propeller wash at Caisson 3 have

not been taken into account in the model simulations. It is expected that this will further mix

the plume with the surrounding waters and limit the dimension of the plume exceeding 36

psu.

All of the discharge options have small saline footprints, and their impacts are therefore considered to be of low significance. The detailed assessment of the elevated salinity in the

discharge brine is summarised in Section 8.2.2.



8.1.3.2 Temperature The increase in temperature of the feed-water during its progress through the RO plant is expected

to be limited (i.e. specified to be less than one degree Celsius) and consequently the thermal

characteristics of the discharged effluent will be largely similar to the ambient seawater temperature

at the intake. During the summer months (or during very calm weather periods), significant

stratification of the water column occurs in Saldanha Bay, where the temperature of the shallow

surface layer (<5 m) in Small Bay can be as high as 20°C, and can be underlain by a strong

thermocline (>1 °C m-1) and a bottom layer with temperatures of about 11°C (Monteiro et al., 1999).

At Site 3, the discharge will be at -8 m CD into Small Bay, -4 m CD into Big Bay or at approximately -

16 to -18 m CD at Caisson 3.. The temperature of the discharged brine can thus be significantly

warmer than that of the ambient water at these depths. A similar scenario is assumed for Sites 1 and

2, as the high density brine is expected to sink to the bottom.

Bamber (1995) defined four categories for direct effects of thermal discharges on marine organisms:

• Increases in mean temperature

• Increases in absolute temperature

• High short term fluctuations in temperature

• Thermal barriers

Increased mean temperature Changes in water temperature can have a substantial impact on aquatic organisms and ecosystems,

with the effects being separated into two groups:

• influences on the physiology of the biota (e.g. growth and metabolism, reproduction timing and

success, mobility and migration patterns, and production); and

• influences on ecosystem functioning (e.g. through altered oxygen solubility).

The impacts of increased temperature has been reviewed in a number of studies along the West

Coast of South Africa, some in Saldanha Bay (e.g. Luger et al., 1997; van Ballegooyen and Luger,

1999; van Ballegooyen et al., 2004, 2005). A synthesis of these findings is given below.

Most reports on adverse effects of changes in seawater temperature on southern African West Coast

species are for intertidal (e.g. the white mussel Donax serra) or rocky bottom species (e.g. abalone

Haliotis midae, kelp Laminaria pallida, mytilid mussels, Cape rock lobster Jasus lalandii). Cook

(1978) specifically studied the effect of thermal pollution on the commercially important rock lobster

Jasus lalandii, and found that adult rock lobster appeared reasonably tolerant of increased

temperature of +6°C and even showed an increase in growth rate. The effect on the reproductive

cycle of the adult lobster female was, however, more serious as the egg incubation period shortened

and considerably fewer larvae survived through the various developmental stages at +6°C above



ambient temperature. Zoutendyk (1989) also reported a reduction in respiration rate of adult J.

lalandii at elevated temperatures.

Other reported effects include an increase in biomass of shallow water hake Merluccious capensis

and West Coast sole Austroglossus microlepis at 18°C (MacPherson and Gordoa, 1992) but no

influence of temperatures of <17.5°C on chub-mackerel Scomber japonicus (Villacastin-Herroro et

al., 1992). In contrast, 18°C is the lower lethal limit reported for larvae and eggs of galjoen Distichius

capensis (Van der Lingen, 1994).

Internationally, a large number of studies have investigated the effects of heated effluent from coastal

power stations on the open coast. These concluded that at elevated temperatures of <5°C above

ambient seawater temperature, little or no effects on species abundances and distribution patterns

were discernable (van Ballegooyen et al. 2005). On a physiological level, however, some adverse

effects were observed, mainly in the development of eggs and larvae (e.g. Cook, 1978, Sandstrom et

al., 1997; Luksiene et al., 2000). Other effects observed include alterations in the photosynthesis

behaviour of algal assemblages (Martinez-Arroyo et al. 2000), decreases in the duration of larval

development (e.g. barnacle larvae, Thiyagarajan et al. 2003), and suppressed growth in the post

larvae of the spiny lobster Panulirus argus due to prolonged intermoulting periods and reduced size

increments with each moult (Lellis and Russel, 1990). For a temperature increase of approximately

5°C an increased photosynthetic rate and biological community metabolism (Q10 ~ 2.5) is reported

(Parsons et al., 1977).

In Saldanha Bay, increased mean temperatures may lead to Mytilus galloprovincialis outcompeting

Choromytilus meridionalis where these species already occur (e.g. iron ore jetty and causeway) as

the juveniles of the former species grow faster in temperatures between 17 and 22°C (van Erkom

Schurink and Griffiths, 1993). Mussel farming originally harvested Choromytilus meridionalis until

Mytilus galloprovincialis started to dominate.

The brine is assumed to have low levels of nitrate as the feed-water was drawn from the surface

waters. There is thus a potential of displacement of naturally high nitrate waters by nitrate deficient

waters at depth. However, the increased mineralization of nutrients in the sediments due to

increased water temperatures will offset or partially offset this reduction in nutrients.

The South African Water Quality Guidelines recommend that the maximum acceptable variation in

ambient temperature should not exceed 1°C, which is an extremely conservative value in view of the

negligible effects of thermal plumes on benthic assemblages reported elsewhere for a ΔT of +5°C or

less. The greatest extent of the thermal plume is expected to be during the autumn calm weather

period, and the following thus focuses on these results.



Site 1:





1°C above ambient temperature will occur for 300 m around the outlet for (Figure C1.5f) a total of

0.25% of the time (i.e. 6 hours in a season). Closer to the outfall, the 1°C threshold will be exceeded

for 20% of the time in a 80 m x 100m area offshore of the discharge and for 5% of the time in an 150

m x 170 m area offshore of the discharge.

Site 2:


between -0.5 m to -1 m below chart datum and 20 to 25 m from the shoreline. The thermal footprint

(i.e. the maximum extent of exceedance of the 1°C above ambient temperature) is expected to

extend over a maximum area of 650 x 450 m (Figures C2.5b and f) for a total of 0.25% of both the

summer and the autumn calm weather period (i.e. 6 hours in a season), and cover an area of 480 x

350 m for approximately 20% of the time and 570 m x 350 m for approximately 5% of the time during

the calm autumn period (Figure C2.5f). These results indicate that the discharge at this location will

be vulnerable to potential thermal impacts during calm periods.

In addition to the exceedance of the SAWQ guideline of a maximum allowable deviation of water

temperature of less than 1°C, the near bottom water temperature at this site also exceeds the less

conservative ANZECC guideline that the change in water temperature should not lie outside the 20

and 80 percentiles of the natural undisturbed temperature variability at the site. With the proposed

discharge the median bottom water temperature exceeds the 80 percentile of the natural temperature

variability at this site during calm (autumn) conditions over an approximate area of 150 m x 250 m.

This exceedance of the ANZECC water quality guideline occurs under summer or winter conditions at this site but does not occur during any season (or “worst case” calm

conditions) at any of the other discharge sites.

Site 3:


below CD into Small Bay, 2) at -4 m below CD into Big Bay, or 3) at approximately -16 to -18 m below

CD at Caisson 3. These will be located alongside the iron-ore causeway at distances ranging

between approximately 50 to 70 m from the causeway. In the modelling study these are defined as

Site 3(a) (discharge into Small Bay), Site 3(b) (discharge into Big Bay) and Site 3(c&d) (discharge at

Caisson 3).



1) In the case of a discharge into Small Bay, the 1°C threshold will be exceeded in an area of

maximal 500 x 230 m for 0.25% (i.e. a total duration of 6 hours during summer) 280 m x

120m for approximately 5% of the time, and for approximately 180 x 100 m around the outlet

for <20% of the time (Figure C3.5b). The thermal footprint could be reduced if the seawater

intake was located deeper in the water column where the temperature difference between

the intake location and the discharge depth could be significantly reduced.

2) The discharge into Big Bay will result in a maximum thermal plume dimension of 440 x 250 m

at 0.25% of the time during summer (Figure C4.5b), 240 x 150 m for approximately 5% of the

time during the calm autumn period (Figure 4.5f) and 180 m for approximately 20% of the

time during the calm autumn period.

3) For discharge at Caisson 3 (Site 3c&d) the model results indicate that the discharge into Big

Bay will result in a maximum thermal plume dimension of 440 x 240 m at 0.25% of the time

during summer (Figure C5.5b), 160 x 140 m for approximately 5% of the time during the calm

autumn period (Figure 4.5f) and 70 m for approximately 20% of the time during the calm

autumn period.

All benthic species have preferred temperature ranges and it is reasonable to expect that those

closest to their upper limits (i.e. boreal as opposed to temperate) would be negatively affected by an

increase in mean temperature. However, in the case of Saldanha Bay two facts have to be noticed.

Firstly, an increase in mean temperature compared to natural ambient bottom temperatures will only

occur seasonally during periods of high stratification (i.e. strong upwelling and/or calm wind

conditions). Secondly, the sessile biota in Saldanha Bay are naturally exposed to wide temperature

ranges due to surface heating and rapid vertical mixing of the water column and intrusions of cold

bottom shelf water into the system. This can lead to rapid variations in temperature of as much as

10°C over less than 12 hours, the absolute range being approximately 10 to 22°C (Monteiro et al.

1999), on occasion in shallower regions reaching temperatures above this. It can thus be assumed

that the biota in these waters are relatively robust and well-adapted to substantial natural variations in

temperature. In fact, while the addition of the warmer brine effluent may at times significantly

increase bottom temperatures in the vicinity of the discharge, the effluent is likely to decrease

temperature variability in the lower water column in the region of impact (albeit with a bias towards

higher temperatures), rather than increasing temperature variability.

The application of the ANZECC water quality guideline (which requires that the median temperature

in the environment with an operational discharge should not lie outside the 20 and 80 percentile

temperature values for a reference location or ambient temperatures observed prior to the

construction and operation of the proposed discharge), is more appropriate to the high temperature

variability conditions in the bottom waters of Saldanha Bay. There is compliance with this guideline

at all discharges sites except for the discharge into the NE corner of Small Bay from Site 2 during



autumn when an area of approximately 100 m around the discharge is predicted to exceed the

ANZECC seawater temperature guideline.

The impact of the elevated temperature of the brine in relation to the receiving waters, is thus

considered to be of low significance for all sites. The discharge from Site 2, while it exceeds the

ANZECC water quality guideline, does so only over a limited area and thus also is deemed to have

an impact of low significance.

Increased absolute temperature The maximum observed sea surface temperature in the region typically is < 22°C. Strong wind

events are likely to mix the water column to such an extent that the bottom waters, at times, have

similar water temperatures to the surface waters. Only at the proposed discharge site in Small Bay

and at Caisson 3 for Site 3, is it less likely that this occurs. The discharged brine will not be heated

above this naturally occurring maximum temperature and therefore an increase in absolute

temperature is not expected and is not further assessed here.

Short term fluctuations in temperature and thermal barriers Temperature fluctuations are typically caused by variability in flow or circulation driven by frequently

reversing winds or tidal streams. For example, Bamber (1995) described faunal impoverishment in a

tidal canal receiving hot water effluent where the temperature variability was ~12°C over each tidal

cycle. As noted above, the bottom waters in the bay may vary rapidly in temperature, i.e. by as much

as 10°C over less than 12 hours. Thus the ecological effects of brine-induced rapid changes in

temperature are not further assessed.

For thermal barriers to be effective in limiting or altering marine organism migration paths they need

to be persistent over time and cover a large cross-sectional area of the water body. The predictions

for the brine plume distributions indicate that neither condition will be met in Saldanha Bay. Over and

above this there are no known migration pathways in the system. This effect can therefore be

considered insignificant.

8.1.3.3 Dissolved Oxygen Dissolved oxygen (DO) is an essential requirement for most heterotrophic marine life. Its natural

levels in seawater are largely governed by local temperature and salinity regimes, as well as organic

content. Coastal upwelling regions are frequently exposed to hypoxic conditions owing to extremely

high primary production and subsequent oxidative degeneration of organic matter. Along the

southern African coast, low-oxygen waters are a feature of the Benguela system, and thus of

Saldanha Bay. Small Bay does experience a fairly regular oxygen deficit during late summer and

winter months, whilst Big Bay experiences less frequent and lower magnitude oxygen deficits

(Atkinson et al., 2006). In addition, existing discharges into Saldanha Bay exacerbate these naturally



occurring low oxygen conditions in certain areas, e.g. in the vicinity of the fish factory discharges in

Small Bay.

Hypoxic water (<2 ml O2. ℓ-1) has the potential to cause mass mortalities of benthos and fish (Diaz

and Rosenberg, 1995). Marine organisms respond to hypoxia by first attempting to maintain oxygen

delivery (e.g. increases in respiration rate, number of red blood cells, or oxygen binding capacity of

haemoglobin), then by conserving energy (e.g. metabolic depression, down regulation of protein

synthesis and down regulation/modification of certain regulatory enzymes), and upon exposure to

prolonged hypoxia, organisms eventually resort to anaerobic respiration (Wu, 2002). Hypoxia

reduces growth and feeding, which may eventually affect individual fitness. The effects of hypoxia on

reproduction and development of marine animals remains almost unknown. Many fish and marine

organisms can detect, and actively avoid hypoxia (e.g. rock lobster “walk-outs”). Some

macrobenthos may leave their burrows and move to the sediment surface during hypoxic conditions,

rendering them more vulnerable to predation. Hypoxia may eliminate sensitive species, thereby

causing changes in species composition of benthic, fish and phytoplankton communities. Decreases

in species diversity and species richness are well documented, and changes in trophodynamics and

functional groups have also been reported. Under hypoxic conditions, there is a general tendency for

suspension feeders to be replaced by deposit feeders, demersal fish by pelagic fish and

macrobenthos by meiobenthos (see Wu, 2002 for references). Further anaerobic degradation of

organic matter by sulphate-reducing bacteria may additionally result in the production of hydrogen

sulphide, which is detrimental to marine organisms (Brüchert et al. 2003).

Because oxygen is a gas, its solubility in seawater is dependent on salinity and temperature, whereby

temperature is the more significant factor. Increases in temperature and/or salinity result in a decline

of dissolved oxygen levels. The temperature in the effluent is not significantly elevated in relation to

the intake water temperature, and a reduction in dissolved oxygen is thus only expected as a result of

the elevated salinity (63.5 psu) of the brine. For example, saturation levels of dissolved oxygen in

seawater decrease with rising salinity from 5.84 ml. ℓ-1 at 15˚C and 35 psu, to 4.90 ml. ℓ-1 at 63.5 psu

(DWAF, 1995), not taking into account any biological use of oxygen due to respiration, oxidation and

degradation. In summer months the surface water can reach temperatures of 20 °C (Atkinson et al.

2006), and the DO in the brine at this temperature would decline from 5.30 ml. ℓ-1 at 35 psu to 4.49

ml. ℓ-1 at 63.5 psu. These approximate calculations translate into a 15-16% reduction of DO in the

brine. The South African Water Quality Guidelines for Coastal Marine Waters (DWAF 1995) state

that for the west coast, the dissolved oxygen should not fall below 10 % of the established natural

variation. A potential difference in DO concentration of 15 % is well within the natural variability

range of the waters in Saldanha Bay (Atkinson et al., 2006), and the potential for a reduction in

dissolved oxygen levels will also drastically reduce within a few meters of the outlet as the effluent

mixes with the receiving waters.



As the receiving waters can be colder than the brine effluent they can potentially have a higher

solubility potential and thus potentially a higher DO concentration. However, near-bottom waters on

the West Coast are often characterised by hypoxic conditions as a result of decomposition of organic

matter and low-oxygen water generation processes. In particular, the bottom waters of Small Bay

often experience oxygen deficits (Atkinson et al., 2006), particularly during thermal stratification. It is

thus unlikely that the receiving bottom water will have notably higher oxygen concentrations than the

surface waters. Descending brine plumes may therefore act as localised suppliers of oxygenated

water to the seabed, rather than contributing to oxygen depletion. A decrease in DO levels in the

discharged brine is thus not of critical concern, and is assessed as being of low significance.

A critical factor that needs to be observed is that oxygen depletion in the brine might also occur

through the addition of sodium metabisulfite, an oxygen scavenger, which is commonly used as a

neutralizing agent for chlorine (http://www.paua.de/Impacts.htm). Should oxidising biocides be used

in pumping systems (e.g. intake pipe and RO membranes) to ensure that they are remain free of

biofouling organisms, the use of sodium metabisulfite as a neutralizing agent for chlorine may be

required, however, presently the project description is such that only non-oxidising biocides are

specified for use in pumping systems (e.g. intake pipe and RO membranes) to ensure that they are

remain free of biofouling organisms. Chlorine will only be used to treat the product (fresh water) and

thus is not expected to be discharged to the marine environment. Should sodium metabisulfite be

used as a neutralizing agent for chlorine, aeration of the effluent is recommended prior to discharge

(http://www.paua.de/Impacts.htm) However, it should be noted that there will be only a limited

requirement for the use of chlorine as a biocide in the case of beach well intakes and consequently

the use of sodium metabisulfite in a system based on beach well intakes is unlikely. It should be emphasized here that the use of an Oxidising biocide and thus also sodium metabisulfite as a neutralizing agent presently is not proposed for the Saldanha Bay RO plant, however the impacts of an oxygen scavenger have been assessed in this study for completeness and to allow for the use of chlorine as a biocide of choice should it be considered a viable alternative to non-oxidising biocides..

As discussed above, the expected changes in dissolved oxygen are associated with both direct

changes in dissolved oxygen content due to the difference between the ambient dissolved oxygen

concentrations and those in the effluent being discharged. However, indirect changes in dissolved

oxygen content of the water column and sediments due to changes in hydrodynamic and ecosystem

functioning in the bay are also possible.

For example, oxygen concentrations may change (particularly in the bottom waters and in the

sediments) due to:

i) changes in phytoplankton production as a result of changes in nutrient dynamics (both in

terms of changes in nutrient inflows and vertical mixing of nutrients) and subsequent

deposition of organic matter, and



ii) changes in remineralisation rates (with related changes in nutrient concentrations in near

bottom waters) associated with near bottom changes in seawater temperature

associated with the brine discharge plume.

Several of the waste water constituents have the potential to act as nutrients for plants (e.g. Sodium

tripolyphosphate and trisodium phosphate). In principle the phosphate can act as a plant nutrient and

thus increase algal growth, however, phosphate generally is not limiting in marine environments such

as Saldanha Bay, unless there are significant inputs of nitrogen (nitrates, ammonia), which is the

limiting nutrient in such systems.

On the other hand, the discharge of the high saline brine may result in increased turbidity, potentially

reducing the light penetration for algal growth (Miri et al., 2005). Such indirect changes are difficult to

quantify, but as a pro-active measure it is recommended that pollutants in the brine that may result in

increased phytoplankton production, be reduced as much as possible.

8.1.3.4 Biocides The use of a biocide in the intake water is undertaken to ensure that the pumping systems (e.g.

intake pipe and RO membranes) are maintained free of biofouling organisms. For example, larvae of

sessile organisms (e.g. mussels, barnacles) can grow in the intake pipe, and impede the intake flow

of the feed-water. Biofouling by algae, fungi and bacteria can rapidly lead to the formation and

accumulation of slimes and biofilms, which can increase pumping costs, plant operation costs and

can lead to the proliferation of sulphate reducing bacteria. This can ultimately result in the production

of hydrogen sulphide causing metallic corrosion problems.

There are two main groups of biocides: the oxidising biocides and the non-oxidising biocides. The

classification is based on the mode of biocidal action against biological material. Oxidising biocides

include chlorine and bromine-based compounds and are non selective with respect to the organisms

they kill. Non-oxidising biocides are more selective, in that they may be more effective against one

type of micro-organisms than another. A large variety of active ingredients are used as non-oxidising

biocides, including quaternary ammonium compounds, isothiazolones, halogenated bisphenols,

thiocarbamates as well as others. The non-oxidising biocide proposed for use in the RO Plant is 2.2

Dibromo-3-nitrilopropionamide (DBNPA) commercially identified as Hydrex 4202. DBNPA has

extremely fast antimicrobial action and rapid degradation to relatively non-toxic end products. The

ultimate degradation products formed from both chemical and biodegradation processes of DBNPA

include ammonia, carbon dioxide, and bromide ions. DBNPA also degrades with sunlight (with the

formation of inorganic bromide ion) with a reported half-life of approximately 7 days (Dow Chemicals,

Fact Sheet No. 253-01464-06/18/02). Water quality guidelines identified as being appropriate to

Saldanha Bay are indicated in Table 8.1. Based on literature seemingly degradation end products

(e.g. ammonia) will not be problematic in the marine environment, however it is the specific biocidal

action of residual DBNPA in the effluent streams to the marine environment that is the major concern.



Transnet Projects (through their contractor) have indicated that only non-oxidising biocides (DBNPA)

will be used in any system that will result in an effluent stream entering the marine environment.

Oxidising biocides, while a viable alternative, will only be used in the treatment of the product (i.e. the

freshwater produced by the RO Plant) and will consequently not result in a discharge of residual

biocide to the marine environment. The reason for this choice is the sensitivity of the RO Plant

membranes to chorine residuals in the intake waters. In this report we have assessed both

alternatives (i.e. oxidising and non-oxidising biocides) in systems that will result in residual

concentrations of biocides being discharged into the marine environment.

It is proposed that DBNPA will be used as filtration shock dosing injected before the dual medium

filters (i.e. added to the feed tank) in a nominal shock dosing regime of 10 ppm (i.e. 10 mg.ℓ-1) for a

10 minute duration every 4 hours. This non-oxidising biocide may also on occasion (e.g. once per

month) be added at the intake to prevent fouling of the intake systems. An alternative to this would

be the manual cleaning of such intake systems (e.g. “pigging”). The issue of concern here is the

residual biocide concentration entering the marine environment at the point of discharge into the

marine environment. In the absence of the contractor being able to indicate the residual

concentrations entering the marine environment, data was obtained from a risk assessment for

cooling towers (Klaine et al., 1996) that indicated a profile of residual concentration of DBNPA at

discharge as indicated below.

Table 8.2: Likely profile of residual concentrations of DBNPA in discharges to the marine environment from the RO Plant. (after Klaine et al., 1996)

Percentile Residual concentration (mg. ℓ-1)

0 0.00 10th 0.01 20th 0.11 30th 0.27 40th 0.46 50th 0.65 60th 0.85 70th 1.05 80th 1.26 90th 1.53 100th 2.24

Although these concentrations are for cooling waters suggesting that these residuals may be

underestimated (DBNPA biodegrades more rapidly with higher temperatures), these concentrations

are likely to be representative of the RO Plant discharges to be assessed here as the initial dosing

assumed by Klaine et al. (1996) in developing these residual concentrations was 24 ppm compared

to the more modest 10 ppm proposed for the RO Plant.



The range of expected discharge concentrations indicated above, together with the guidelines

proposed in Table 8.1, suggest that the required dilution of the effluent containing non-oxidising

biocides in the marine environment would be in the same range as that required for oxidising biocides

(NaOCl), namely approximately 33 dilutions. For example, the highest residual concentration (2.34

mg.ℓ-1) and the upper water quality target value (0.07 mg.ℓ-1) suggest a required dilution of 32.

Similarly, the 80th percentile discharge concentration (1.26 mg.ℓ-1) and the more conservative of

water quality target value (0.035 mg.ℓ-1) suggest a required dilution of 36 to meet the water quality

target values.

In the assessment that follows, the assessment thus is based on an achievable dilution of

approximately 33. This means that all of the information and analyses for oxidising biocides (based

on an initial residual chlorine discharge concentration of 0.1 mg. ℓ-1 that is diluted by a factor of 33 to

meet the proposed water quality guideline of 3 μg. ℓ-1) are also of direct relevance to the assessment

of the proposed non-oxidising biocide (DBNPA). In all of the text that follows, where reference is

made to the water quality guideline for residual chlorine of 3 μg.ℓ-1 being achieved, it can be assumed

that it also implies that the proposed water quality guidelines for DBNPA have also been achieved.

For this reason any reference to chlorine as a biocide and the achievement of the 3 μg. ℓ-1 water

quality target value for residual chlorine can be assumed to refer to, and be relevant to, the

assessment of both oxidising biocides (chlorine and the non-oxidising biocide (DBNPA) proposed for

use in the RO Plant.

Note that the validity of the assessment for DBNPA necessarily implies that the dosing of DBNPA will

be controlled and adjusted to ensure that the residual DBNPA concentrations meet those suggested

in the Table 8.2 above, i.e. for assumed water quality guidelines or target values of of 0.035 mg. ℓ-1

and 0.07 mg. ℓ-1 (and assuming that the biocide plume dimension delineated by the 33 x dilution

contour are acceptable), the residual concentration of DBNPA in the effluent should not exceed

figures of between 1.15 mg.ℓ-1 and 2.475 mg.ℓ-1). However due to the uncertainties in the exact

residual concentration of DBNPA at the point of discharge it is assumed that there is an undertaking

that the dosing levels and dosing regimes will be adjusted to ensure that any potential environmental

impacts of significance will be avoided. The monitoring and toxicity testing proposed (Section 9) is

designed to provide the requisite information to ensure this. If required, it is possible to reduce the

residual DBNPA concentrations by designing the brine basin so as to ensure greater and sufficient

dilution of the DBNPA residuals in the effluent stream before discharge or to revert to the use of an

oxidising biocide (chlorine) in this role.

As stated above, although not presently considered an option for the RO Plant, we have also

considered and assessed the use of chlorine as a biocide. For this purpose we have assumed that

sodium hypochlorite (NaOCl) as an oxidising biocide will be added at a constant concentration of

approximately 1.0 ppm in the inlet water and that it is proposed to discharge the brine directly into the

sea.



There is thus the potential for residual quantities of chlorine and its degradation/transformation

products to be present in the effluents. The concentration of freely available chlorine (FAC) or

residual chlorine at the outlet is specified to be approximately 0.1 mg.ℓ-1 or 0.1 ppm. However, in

extreme cases, outlet concentrations may attain 2.0 mg.ℓ-1 if shock doses are used to clean the plant

(e.g. Stenton-Dozey and Brown, 1994). (It should be noted that modern plants often operate on

polyamide membranes, which are sensitive to Oxidising chemicals such as chlorine. Neutralization is

then typically required before the feedwater enters the RO unit. Under these circumstances it can be

assumed that the brine does not contain free residual chlorine).

The chemistry associated with seawater chlorination is complex and only a few of the reactions are

given below, summarized from White (1986), DWAF (1995) and ANZECC (2000). Chlorine does not

persist for extended periods in water but is very reactive. Its by-products, however, can persist for

longer.

The addition of sodium hypochlorite to seawater results in the formation of hypochlorous acid:

NaOCl + H2O → HOCl + Na+ + OH-

Hypochlorous acid is a weak acid, and will undergo partial dissociation as follows:

HOCl → H+ + OCl-

In waters of pH between 6 and 9, both hypochlorous acid and hypochlorite ions will be present; the

proportion of each species depending on the pH and temperature of the water. Hypochlorous acid is

significantly more effective as a biocide than the hypochlorite ion. Seawater chlorination differs

greatly from that of fresh water primarily due to the high bromide concentration of seawater (average

bromide concentration in seawater is 67 mg.ℓ-1). In the presence of bromide, chlorine instantaneously

oxidises bromide to form hypobromous acid:

HOCl + Br- → HOBr + Cl--

Hypobromous acid is also an effective biocide. It is worth noting that, for a given pH value, the

proportion of hypobromous acid relative to hypobromite is significantly greater than the corresponding

values for the hypochlorous acid - hypochlorite system. Thus, for example, at pH 8 (the pH of

seawater), hypobromous acid represents 83% of the bromine species present, compared with

hypochlorous acid at 28%. Hypobromous acid can also disproportionate into bromide and bromate

which is accelerated by sunlight.

Naturally occurring organic substances (e.g. ammonia) contribute to a major part of oxidant

consumption, i.e. chlorine reacts readily with nitrogenous substances (e.g. ammonia) to form N-

chlorinated compounds, which constitute the combined chlorine. These compounds are more

persistent than the free chlorine. The reaction of hypochlorous acid with ammonia results in the



formation of chloramines. The formation of chloramine species is dependent on pH, temperature,

contact time and the relative concentrations of chlorine and ammonia. Essentially, any free chlorine

will be converted to monochloramine at pH 7 to 8 when the ratio of chlorine to ammonia is equimolar

(5:1 by weight) or less. At higher chlorine to ammonia ratios, or lower pH values, dichloramine and

trichloramine will be formed. When ammonia is present, the competing reactions of chlorine with

bromide and ammonia are likely to result in the rapid formation of both monochloramine and

hypobromous acid. In coastal seawater, ammonia concentrations are usually lower (typically less

than 28 µg N.ℓ-1) and therefore hypobromous acid is the main species. However, when ammonia

increases, bromamines (tri-and di-bromamines) may be formed. These bromamines are highly

oxidising species and disappear rapidly to form organic bromamines. Chlorine can also react with

nitrogen-containing organic compounds, such as amino acids to form organic chloramines. Little is

known about the biocidal properties of these compounds.

In natural waters, chlorine can undergo a range of reactions in addition to those discussed above. It

will react with inorganic constituents of water such as iron (II), manganese (II), nitrite and sulphide.

The reaction of chlorine with organic constituents in aqueous solution can be grouped into several

types:

(a) Oxidation,

where chlorine is reduced to chloride ion, e.g. RCHO + HOCl → RCOOH + H+ + Cl-

(b) Addition,

to unsaturated double bonds, e.g. RC = CR' + HOCl → RCOHCClR'

(c) Substitution,

to form N-chlorinated compounds, e.g. RNH2 + HOCl → RNHCl + H2O

or C-chlorinated compounds, e.g. RCOCH3 + 3HOCl → RCOOH + CHCl3 + 2H2O

Chlorine substitution reactions can lead to the formation of halogenated compounds, such as

chloroform (e.g. reaction c), and, where HOBr is present, mixed halogenated and brominated organic

compounds. Although such reactions are significant in terms of the resultant halogenated by-

products, it has been estimated that only a few percent of the applied chlorine ends up as

halogenated organic products. Chlorine is a powerful oxidant, and a significant proportion of the

applied chlorine is likely to be consumed in reactions such as (a), leading to the formation of non-

halogenated organic products, with chlorine being reduced to chloride.

A number of other source water characteristics are likely to have an impact on the concentrations of

organic by-products present in brine water discharges: natural organic matter in water is the major

precursor of halogenated organic by-products, and hence the organic content of the source water

(often measured as total organic carbon, TOC) may affect the concentration of by-products formed.

In general, the higher the organic content of the source water, the higher the potential for by-product

formation. The ammonia concentration is likely to affect the extent of by-product formation, through



reaction with chlorine to form chloramines. Although seawater generally contains lower

concentrations of ammonia than freshwater, under certain conditions (dependent on chlorine

dose:ammonia nitrogen concentration) it can compete with bromide for the available chlorine to form

monochloramine. In addition, hypobromous acid can react with ammonia to form bromamines.

Although the sequence of reactions is complex, it is likely that the reaction of either hypochlorous or

hypobromous acid with ammonia to form halamines will reduce organic by-product formation during

the chlorination of seawater. The pH of the incoming feed-water could also affect the nature of the

by-products formed. In general, while variations in pH are likely to affect the concentrations of

individual by-products, the overall quantity formed is likely to remain relatively constant.

The presence of certain pollutants in source waters could lead to an increase in the levels of certain

halogenated organics. The presence of phenol, for example, can lead to the formation of

chlorophenols, some of which can taint fish flesh at concentrations as low as 0.001 mg.ℓ-1 (DWAF

1995). Currently, there is no evidence of contamination of the source waters with phenol (Atkinson et

al., 2006).

Paradoxically, chlorine chemistry thus establishes that no free chlorine is found in chlorinated

seawater where bromide oxidation is instantaneous and quantitative. However, the chlorinated

compounds, which constitute the combined chlorine, are far more persistent than the free chlorine.

After seawater chlorination, the sum of free chlorine and combined chlorine is referred to as total

residual chlorine (TRC).

Marine organisms are extremely sensitive to residual chlorine, making it a prime choice as a biocide

to prevent the fouling of marine water intakes. Many of the chlorinated and halogenated by-products

that are formed during seawater chlorination (see above) are also carcinogenic or otherwise harmful

to aquatic life (Hoepner and Lattemann, 2002; Einav et al., 2002). Values listed in the South African

Marine Water Quality Guideline (DWAF 1995) show that 1500 µg.ℓ-1 is lethal to some phytoplankton

species, 820 µg.ℓ-1 induced 50% mortality for a copepod and 50% mortality rates are observed for

some fish and crustacean species at values exceeding 100 µg.ℓ-1 (see also ANZECC, 2000). The

lowest values at which lethal effects are reported are 10 – 180 µg.ℓ-1 for the larvae of a rotifer,

followed by 23 µg.ℓ-1 for oyster larvae (Crassostrea virginica). Sublethal effects include valve closure

of mussels at values <300 µg.ℓ-1 and inhibition of fertilisation of some urchins, echiuroids, and

annelids at 50 µg.ℓ-1. Eppley et al. (1976) showed irreversible reductions in phytoplankton

production, but no change in either plankton biomass or species structure at chlorine concentrations

greater than 10 µg.ℓ-1. Bolsch and Hallegraeff (1993) showed that chlorine at 50 µg.ℓ-1 decreased

germination rates in the dinoflaggelate Gymnodinium catenatum by 50% whereas there was no

discernable effect at 10 µg.ℓ-1. This indicated that particularly the larval stages of some species may

be vulnerable to chlorine pollution. The minimum impact concentrations reported in the South African

water quality guidelines are in the range 2 to 20 µg.ℓ-1 at which fertilisation success in echinoderm

(e.g. sea urchin) eggs is reduced by approximately 50% after 5 minute exposures.



For the assessment that follows a conservative water quality guideline or target value of < 3 µg.ℓ-1

(ANZECC, 2000) is used for oxidising biocides (NaOCl) and the guidelines indicated in Table 8.1

used for the non-oxidising biocide DBNPA). In all text that follows, reference to biocides implies

either chlorine or DBNPA and the reference to target values or water quality guidelines constitutes a

reference to the respective guidelines for either chlorine or DBNPA. Similarly while reference in the

Figures contained in Appendix C refers specifically to the target value of 3 µg.ℓ-1 for chlorine, it should

be taken also to refer to the relevant target values for DBNPA.

Site 1:

At Site 1 there the discharge is planned to occur via a pipeline along a 30 m length of the revetment

of the reclaim dam at a location approximately 80 m from the low water mark of the adjacent

shoreline where the water depth adjacent to the revetment is ~ 1.5 m.


target value for biocide is restricted to an area of ~160 m to 180 m radius around the discharge outlet

(Figure C1.7b and f), but this is expected to occur only for a total of 0.25% of the summer and/or calm

weather period (i.e, a cumulative total of 6 hours within a season comprising 90 days). The target

value could however be exceeded within a 120 m radius of the discharge point for up to 5 days within

all seasons (Figures C1.7b,d and f).

During periods of little or no stratification (e.g. winter) the biocide plume footprint is significantly

smaller due to greater mixing of the plume throughout the vertical extent of the water column. Due to

the restricted size of the plume footprint, this impact is assessed to be of low significance, even under

the ‘worst-case-scenario’.

Site 2:


between -0.5 m to -1 m below chart datum and 20-25 m from the shoreline. Modelling results

indicate that the biocide plume will be restricted to an area of ~400 to 550 m radius around the

discharge outlet (Figure C2.7b and f), but this is expected to occur only for a total of 0.25 % (i.e. a

cumulative duration of approximately 6 hours of the season) of the calm weather period. The target

value could however be exceeded within a 220 to 450 m radius of the discharge point for up to 5

days within the summer season and or calm autumn period (Figure C2.7b and f). Given the

limitations in the resolution of the modelling study, a conservative assumption would be to assume

that the target value for residual concentrations of biocide could be exceeded regularly within a

minimum radius of 50 m of the discharge point, although this is not necessarily the case.

The size of the plume footprint thus is relatively constrained. Consequently, the impact can be

regarded as being of low significance. However it should be recognised that the discharge for Site 2

is occurring into the relatively poorly-flushed Small Bay system that supports a number of mariculture



activities and would inherently pose greater risks than the other sites considered. This is particularly

true given the many potential biocide-related compounds that could be formed and the high nutrient

wastes (nitrogeous compounds) from these mariculture and fish processing activities in Small Bay.

Site 3:






Caisson 3). In all cases, the model results indicate that the plume is restricted to within 120 m of the

outlet near the seabed in summer (when the water column is highly stratified and vertical mixing

limited) and a nominal 50 m radius in the other seasons. Only in the case of the Big Bay discharge

does the plume footprint marginally exceed a nominal 50 m radius of the discharge in seasons other

than summer.

All of the discharge options have negligible biocide footprints when compared to those at the

Alternative Sites 1 and 2, and are therefore considered to be of low significance.

8.1.3.5 Co-discharged Waste Water Constituents

Table 8.3 provides a list of chemicals used in the pre-treatment, or cleaning (CIP = Clean in Place)

process of the RO membranes. Some of the pre-treatment chemicals will be co-discharged with the

brine. This list is compiled from the information provided in the Tender applications for the

construction of the RO plant. The potential impact of biocides is discussed in the previous section;

this section deals with the other potentially co-discharged constituents of the effluent.



Table 8.3: Potential chemicals used for the Reverse Osmosis process - information as supplied by Transnet on 5 December 2007. Quantities based on intakes flows of 8000 m3.day-1.

Substance Quantity Method of Storage Additional Notes

Sodium hypochlorite (NaOCl)

333 kg.day-1 (12%) 40 kg.day-1 (100%)

58 kg.day-1 (12%) 7 kg.day-1 (100%)

3333 ℓ storage tank


Direct intake cleaning only & discharged with brine (presently not under consideration as it is proposed to used a non-oxidising biocide for this purpose) Treating the Permeate (Fresh Water) after RO membranes

Non-oxidising biocide DBNPA – 2.2 Dibromo-3-nitrilopropionamide (Hydrex 4202)

Design: 4 kg.day-1

Max: 4 kg.day-1

98% solution storage capacity of

100 litres

Filtration shock dosing Injected before DMF& discharged with brine

Flocculant Ferric Chloride (FeCl3)

Design: 65 kg.day-1 (40%) or 26 kg.day-1 (100%)

Max: 108 kg.day-1 (40%)

or 43 kg.day-1 (100%)

40 % solution storage capacity of

100 litres

Filtration dosing Injected before DMF & discharged with brine

Anti-scalant (SWRO) Hydrax 4105

Design: 22 kg.day-1

Max: 32 kg.day-1

100 ℓ storage tank Dosed before the RO membranes DMF & discharged with brine

Calcium Hydroxide or Sodium Hydroxide (Caustic soda)

Maximum of 0.5 kg.day-1 1000 ℓ polypropylene tank

Product water stabilization, pH Control, dissolve organic substances and silica

Limestone (Calcium Carbonate)

Design: 39 kg.day-1 (92%) or 39 kg.day-1 (100%)

Max: 47 kg.day-1 (92%) or 43 kg.day-1 (100%)


Water remineralisation i.e. treating the

Permeate (Fresh Water) after RO

membranes

Citric acid H3C6H5O7.H2O

Design: 2.5 kg.day-1 (100%) Max: 9.9 kg.day-1 (100%) 100 ℓ storage tank

CIP Chemicals (low pH)

Direct intake cleaning Discharged to sewage

or waste system

Sulphuric acid Design: 0.3 kg.day-1 (98%) or 0.3 kg.day-1 (100%) 100 ℓ storage tank

CIP Chemicals (high pH)


or waste system

Ethylenediaminetetraacetic Acid (EDTA)




or waste system

Sodium tripolyphosphate (STPP)




or waste system



Substance Quantity Method of Storage Additional Notes

Trisodium phosphate (TSP)




or waste system

Sodium lauryl sulphate (SLS)




or waste system Coagulants like ferric chloride (FeCl3) are used as part of the pre-treatment process to cause

particles in feed-water to form larger masses that can be more easily removed with filters before the

water passes through to the RO membranes. Coagulant aids (organic substances with high

molecular masses that bridge particles further together) and pH control (Calcium Hydroxide or

Sodium Hydroxide) are supplementary methods to enhance coagulation. These chemicals generate

a by-product of coagulated/flocculated materials that become entrapped within the filter-medium and

can be removed only by water-backwashing. The pre-treatment filters are backwashed with filtered

seawater every few days, producing a sludge that contains mainly sediments and organic matter, and

filter coagulant chemicals. According to the project description, a continuous dosage of approximately

8.13 mg.ℓ-1 is anticipated for the intake flow rates of 8000 m3.day-1. The discharge volume of the sludge

discharge is 336 m3.day-1 however the concentration of the material in the backwash presently is

unknown. However, it is reasonable to assume that the full 65 kg.day-1 of flocculant added will be

discharged together with the brine discharge at a nominal flocculant concentration of approximately

14.8 mg.ℓ-1. Acute toxic effects of the sludge are generally not expected, but some evidence suggests

that chronic effects could occur (Sotero-Santos et al., 2007, e.g. reduction of fecundity in Daphnia).

However, ferric chloride may cause a discoloration of the receiving water, and the sludge discharge

may lead to increases in turbidity and suspended matter. Impacts such as reduced primary

production or burial of sessile organisms by increased turbidity in the discharge may thus occur

(Sotero-Santos et al., 2007, http://www.paua.de/Impacts.htm). A process for removal of the particles

is thus recommended before discharge of the sludge. The proposed RO plant design specifications

indicate that the DMF backwash (sludge) will be blended with the brine, a viable but less desirable

alternative.

Scaling on inside tubes or on RO membranes impairs plant performance. Anti-scalants are

commonly added to the feed-water in RO plants to prevent scale formation. The main

representatives of anti-scalants are organic, carboxylic-rich polymers such as polyacrylic acid and

polymaleic acid. Acids and polyphosphates are still in use at a limited scale but on an increasingly

limited scale as they can cause eutrophication through formation of algal blooms and macro algae

(DWAF, 2007). According to the initial list of chemicals provided, potential anti-scalant products are

Avista Vitec 3000, and Flocon 135, which is a phophinocarboxylic acid. Both products are certified

by the National Sanitation Foundation (NSF) under ANSI/NSF Standard 60 for use in producing



potable water. Polymer antiscalants have similar properties to natural humic matter or Gelbstoffe,

which are common seawater constituents (Hoepner and Lattemann, 2002). They have high molecular

weight, multiple carboxylic groups, metal-ion binding capacity and a high stability. LC50 values are

relatively high (generally exceeding 1,000 ppm, see Hoepner and Lattemann, 2002), which indicates

a low toxicity and is far above the typical dosage in desalination plants. An acute environmental risk

associated with their release into the marine environment is thus expected to be relatively low. Due

to a poor degradability, however, dispersal and relatively long residence times must be expected,

during which it cannot be excluded that antiscalants will limit the availability of biologically essential

trace metal ions (Hoepner and Lattemann, 2002). It is unlikely (particularly in the quantities used in

the RO Plant) that anti-scalants will promote the accumulation of metals in the sediments.

The cleaning intervals (CIP) of the RO membranes are typically three to six months depending on the

quality of the plant's feed-water (Einav et al., 2002). The chemicals used in cleaning are mainly

weak acids and detergents. In RO plants, alkaline cleaning solutions (pH 11-12) are used for

removal of silt deposits and biofilms, whereas acidified solutions (pH 2-3) remove metal oxides and

scales. Further chemicals are often added to improve the cleaning process of RO membranes, such

as detergents, oxidants, complexing agents or biocides for membrane disinfection. Neutralization of

the extremely alkaline or acidic solutions and treatment of additional cleaning agents typically is

recommended before discharge to the ocean to remove any potential toxicity

(http://www.paua.de/Impacts.htm). The chemicals Sulphuric acid, Ethylenediaminetetraacetic acid

(EDTA), Sodium tripolyphosphate (STPP), Trisodium phosphate (TSP), and Sodium lauryl sulphate

(SLS) were specified as being used in the RO membrane cleaning (Table 8.3). Although it is specified that these chemicals will not be discharged to the marine environment but rather will enter the sewage system or be transported off-site, for completeness, a short summary of the environmental fates and effects of these chemicals is given below.

Sulphuric acid (H2SO4) is used for pH adjustment in the desalination process to reduce the pH for the

acid wash cycle. It is a strong mineral acid that dissociates readily in water to sulphate ions and

hydrated protons, and is totally miscible with water. At environmentally relevant concentrations,

sulphuric acid is practically totally dissociated, sulphate is at natural concentrations and any possible

effects are due to acidification. This total ionisation will imply also that sulphuric acid, itself, will not

adsorb on particulate matters or surfaces and will not accumulate in living tissues

(http://www.chem.unep.ch/irptc/sids/oecdsids/7664939.pdf). Sulphuric acid can be acutely toxic to

aquatic life via reduction of water pH. Most aquatic species do not tolerate pH lower than 5.5 for any

extended period. No guideline values are available for this substance but No Observed Effect

Concentration (NOEC) values were developed from chronic toxicity tests on freshwater organisms

and range from 0.058 mg.ℓ-1 for fish populations to 0.13 mgℓ-1 for phytoplankton and zooplankton

populations, respectively (http://www.chem.unep.ch/irptc/sids/oecdsids/7664939.pdf). As seawater is

highly buffered, the limited sulphuric acid discharges proposed (0.3 kg.day-1) are not expected to

have significant impacts in the marine environment.



EDTA is an aminopolycarboxylic salt that is used as a chelating agent to bind or capture trace

amounts of iron, copper, manganese, calcium and other metals. In water treatment systems, EDTA

is used to control water hardness and scale-forming calcium and magnesium ions to prevent scale

formation (http://www.dow.com/productsafety/finder/edta.htm). Because of the ubiquitous presence

of metal ions, it has to be assumed that EDTA is always emitted as a metal complex, although it

cannot be predicted which metal will be bound. EDTA will biodegrade very slowly under ambient

environmental conditions but does photodegrade. EDTA is not expected to bioaccumulate in aquatic

organisms, adsorb to suspended solids or sediments or volatilize from water surfaces (European

Union Risk Assessment Report 2004). Toxicity tests on aquatic organisms have shown that adverse

effects occur only at higher concentrations (the lowest concentrations at which an adverse effect was

recorded is 22 mg.ℓ-1) (European Union Risk Assessment Report, 2004). On the other hand, if trace

elements like Fe, Cu, Mn, and Zn are low in the natural environment, an increased availability of

essential nutrients caused by the complexing agent EDTA is able to stimulate algal growth. Heavy

metal ions in the water are complexed by free EDTA, and a comparison of the toxicity of those

compared to the respective uncomplexed metals and free EDTA have shown a reduction in toxicity

by a factor of 17 to 17000 (Sorvari and Sillanpää, 1996). Experiments (albeit with significantly higher

trace metal concentrations than are typically observed in the environment) indicate that EDTA

decreases the accumulation of metals such as Cd, Pb and Cu, however the absorption of Hg by

mussels is seemingly promoted through complexation with EDTA (Gutiérrez-Galindo, 1981, as cited

in the European Union Risk Assessment Report, 2004). Potential promotion of the accumulation of

metals in sediments is unlikely to be a concern as in high concentrations EDTA prevents the

adsorption of heavy metals onto sediments and even can remobilise metals from highly loaded

sediments (European Union Risk Assessment Report, 2004).

Within the framework of marine risk assessment, the European Union has published a risk

assessment report in which a Predicted No Effect Concentration (PNEC) of 0.64 mg.ℓ-1 was

calculated (European Union Risk Assessment Report, 2004).

Sodium tripolyphosphate (STPP, Na5P3O10) is the sodium salt of triphosphoric acid, and is a typical

ingredient of household cleaning products, and is thus present in domestic waste waters. STPP is an

inorganic substance that when in contact with water (waste water or natural aquatic environment) is

progressively hydrolysed by biochemical activity, finally to orthophosphate. Acute aquatic ecotoxicity

studies have shown that STPP has a very low toxicity to aquatic organisms (all EC/LC50 are above

100 mgℓ-1) and is thus not considered as environmental risk (HERA, 2003). The final hydrolysis

product of STPP, orthophosphate, however, can lead to eutrophication of surface waters due to

nutrient enrichment, phosphate as a nutrient is not limiting in marine environments such as Saldanha

Bay unless there are significant inputs of nitrogen (nitrates, ammonia) that is the limiting nutrient in

such systems. In addition to the hydrolysis into orthophosphate, STPP can, depending on the



presence of cationic ions, precipitate in the form of insoluble calcium, magnesium or other metal

complex species (HERA, 2003).

Trisodium phosphate (TSP) (Na3PO4) is a highly water-soluble cleaning agent. When dissolved in

water it has an alkaline pH. The phosphate can act as a plant nutrient, and can thus increase algal

growth, however, as noted above, phosphate as a nutrient is not limiting in marine environments such

as Saldanha Bay unless there are significant inputs of nitrogen (nitrates, ammonia) which is the

limiting nutrient in such systems. Phosphate is classified as not acutely toxic to aquatic organisms

(http://www.pesticideinfo.org/Detail_Chemical.jsp?Rec_Id=PC34419).

Sodium lauryl sulphate (SLS) (C12H25NaO4S) is an anionic surfactant, which is a class of chemicals

used for their detergent properties. SLS is biodegradable in surface waters, and biodegradation

ranges from 45 to 95% within 24 hours. Products of SLS biodegradation are carbon dioxide or

saturated fatty acids. SLS is classified as a substance of low environmental toxicity with low

bioaccumulation (OECD, 1997).

Typically the rinse water is kept in a titration container and after being treated (titration, neutralization

of the cleaning materials), it is disposed off by transporting it in closed containers to an authorized

salt disposal site. Alternatively, it is injected by continuous flow of small quantities into the brine and

discharged into the sea in the effluent (Einav et al., 2002). Transnet has undertaken to not co-

discharge these waters with the brine discharge.

In summary, the toxicity of the various chemicals used in the pre-treatment and CIP process (aside from biocides) is relatively low, and none of the products are listed as tainting substances (DWAF, 1995). Of concern is more the potential increase in turbidity (backwash-sludge). To reduce the impact on the marine environment, it is recommended that particulates are removed from the backwash sludge. Whether this is necessary and/or the extent to which this is necessary at the various proposed discharge locations, will be informed by the results of the monitoring programme proposed for the brine discharge.

The waste brine often contains low amounts of heavy metals that pass into solution when the plant's

interior surfaces corrode. In RO plants, non-metal equipment and stainless steels are typically used.

The RO brine may therefore contain traces of iron, nickel, chromium and molybdenum, but

contamination levels are generally low (Hashim and Hajjaj, 2005; http://www.paua.de/Impacts.htm).

Heavy metals tend to enrich in suspended material and finally in sediments, so that areas of

restricted water exchange (e.g. the NE corner of Small Bay) and soft bottom habitats impacted by the

discharge could be affected by heavy metal accumulation. Many benthic invertebrates feed on this

suspended or deposited material, with the risk that metals are enriched in their bodies and passed on

to higher trophic levels. At this stage, no assessment of the potential concentration of heavy metals



can be provided, as it is an incidental by-product of RO plant processes and is likely to be low. It is

therefore recommended that limits are established for heavy metal concentrations in the brine

discharges (see Table 8.1), and the brine regularly monitored to avoid exceedance of these limits.

As the effluent is likely to consist of a combination of co-discharge constituents, namely Ferric

chloride and anti-scalants (and biocides that in this assessment have been considered separately

from other co-discharges), the approach taken in assessing their potential impacts has been to model

dilution rates. The specific concentrations of all of the co-discharges at the point of discharge are,

however, not known, so dilution factors of 50 and 100 were chosen as a conservative target value for

modelling exceedances. For example, the total residual chlorine added as a biocide at a

concentration of 100 µg.ℓ-1 (0.1 mg.ℓ-1) will need to be diluted by a factor of 33 to achieve a target

value of 3 µg.ℓ-1. The dilution threshold of 50 and 100 assumed for the co-discharge assessment

would be the equivalent of requiring that the biocides in the effluent discharge be reduced to 2 µg.ℓ-1

or less and 1 µg.ℓ-1 or less, respectively. The achievable dilution thresholds of 50 and 100 therefore

is considered a suitably conservative dilution ratio for assessing potential footprints of co-discharges

(even potential tainting constituents or by-products) in the effluent. It may, however, be that for

specific co-discharge constituents significantly greater or lesser dilutions are required. The exact

required dilution will depend on the discharge concentration and the relevant water quality target

value considered appropriate for the receiving environment. Presently, none of the co-discharges

that have been specified require such high dilutions.

All proposed intake points at the three alternative sites are located in the vicinity of the iron ore

causeway and its associated ship traffic, and there is therefore a risk of oil contaminating the feed-

water. Hydrocarbon contamination of the sediment in Saldanha Bay were measured in 1999 and

found to be very low, and were not considered to pose an ecological risk (Atkinson et al., 2006).

Similar, but more extensive sampling and analyses in Big Bay (CSIR, 2007) confirm that there are

limited hydrocarbons present in the sediments. However, the potential for an oil spill accident or

contamination has to be taken into account. Oil contamination can not only affect the quality of the

potable water but can also foul the seawater intake filter, hence limiting the amount of water intake.

Internal membranes may also be fouled and disrupt the reverse osmosis process (Al Malek and

Mohamed, 2005). The direct intake structure (pipeline) will be ~ -1 m CD (i.e. on average

approximately 1.9 m below MSL) and since spilled oil is mostly confined to surface waters, the risk of

an intake of oil is reduced. However, depending on the oil type and weather conditions, significant

quantities of oil may be dispersed into the water column. For beach wells to be affected, significant

contamination of beach sediments would have to occur, thus intake waters from beach wells are

unlikely to be subject to levels of hydrocarbon contamination that would be of concern.



Site 1:




At Site 1 the model results for this discharge configuration indicate that during the periods when there

is the least dispersion of effluent in the bottom layers, i.e. during summer (Figure 1.10b) or a calm

weather period (Figure 1.10f), a dilution of 50 will be exceeded beyond an area of 360 x 230 m

around the outfall for 99% of the time (all but 6 hours during a season) and beyond an area of 190 m

x 120 m for 95% of the time (i.e. all but approximately 5 days during a season). For an achievable

dilution less than 100 this will be restricted to a footprint of 120 x 50 m or less for 50% of the time.

Similarly during the same periods (Figure C1.9b and 1.9f), a dilution of 100 will be exceeded beyond

an area of 500 x 350 m around the outfall for 99% of the time (all but 6 hours during a season) and

beyond an area of 430 m x 250 m for 95% of the time (i.e. all but approximately 5 days during a

season). An achievable dilution less than 100 this will be restricted to a footprint of 130 x 60 m or

less for 50% of the time.

During periods of little or no stratification (e.g. winter) the co-discharge plume footprint is significantly

smaller due to greater mixing of the plume throughout the vertical extent of the water column. Unlike

for other water quality parameters such as salinity, the spatial extent of the “footprint” of the

discharge plume (i.e. non-exceedance of an achievable dilution of 50) in the surface waters is largely

similar to that at depth however the dilutions in the surface water significantly exceed those at depth.

Due to the restricted size of the plume footprint, this impact is assessed to be of low significance,

even under the ‘worst-case-scenario’. The assessment of low significance is particularly applicable as the CIP cleaning chemicals will not be discharged with the brine.

Site 2:


between -0.5 m to -1 m below chart datum and 20 to 25 m from the shoreline. Modelling results

indicate that the co-discharge plume will be diluted by a factor of 50 or more beyond an area

extending 640 m offshore and 320 m during calm periods (Figure C2.10f) and beyond an area of

550 m alongshore and 500 m offshore during summer (Figure C2.10b). A dilution of 50 will be

exceeded for 95% of the time beyond an area of 300 x 300 m around the outfall during summer and

beyond an area of 550 m x 280 m during the calm autumn period. Achievable dilutions exceeding 50

will be achieved 50% of the time beyond a radius of ~300 m of the outlet Furthermore, modelling

results indicate that the co-discharge plume will be diluted by a factor of 100 or more beyond an area

extending 800 m offshore and 700 m during calm periods (Figure C2.9f) and beyond an area of

600 m alongshore and 700 m offshore during summer (Figure C2.9b). A dilution of 100 will be

exceeded for 95% of the time beyond an area of 450 x 600 m around the outfall during summer and

beyond an area of 700 m x 500 m during the calm autumn period. Achievable dilutions exceeding

100 will be achieved 50% of the time beyond a radius of ~350 m of the outlet. Unlike for other water



quality parameters such as salinity, the spatial extent of the “footprint” of the discharge plume (i.e.

non-exceedance of an achievable dilution of 50 or 100 in the surface waters is largely similar in

shape to that at depth, however the dilutions in the surface water significantly exceed those at depth

The potential impacts of co-discharges can be regarded as being of low significance, provided that an

achievable dilution of 50 to 100 is sufficiently conservative for all co-discharge constituents and

potential by-products. This is the case for all co-discharges presently specified (See Table 8.3). The assessment of low significance is particularly applicable as the CIP cleaning chemicals will not be discharged with the brine.

Site 3:






Caisson 3). 1) For a discharge into Small Bay, the model results indicate that a dilution of 50 will be

exceeded beyond an area of 240 x 100 m around the outfall for 99% of the time and beyond

an area of 100 m x 100 m for 95% of the time during the summer period (Figure C3.10b).

The plume dimension are similar (but slightly smaller) for the calm autumn period when the

water column is more stratified (Figure C3.10f). A dilution of 100 will be exceeded beyond an

area of 450 x 200 m around the outfall for 99% of the time and beyond an area of 260 m x

120 m for 95% of the time during the calm autumn period (Figure C3.9f). The results are

similar for the summer period when the water column is more stratified (Figure C3.9b). 2) For a discharge into Big Bay, the model results indicate that during summer (the worst case

result) a dilution of 50 will be achieved beyond an area 280 m x 150 m around the outfall for

99% of the time and beyond an area of 130 m x 100 m around the outfall for 95% of the time

(Figure C4.10b). During calm periods (the worst case result), a dilution of 100 will be

achieved beyond an area 420 m x 250 m around the outfall for 99% of the time and beyond

an area of 300 m x 210 m around the outfall for 95% of the time (Figure C4.9f).

3) For discharge at Caisson 3 (Site 3c&d) the model results indicate that the worst case results

occur for the calm autumn period. A dilution of 50 exceeded beyond an area 180 m x 170 m

around the outfall for 99% of the time and beyond an area of 120 m x 120 m around the

outfall for 95% of the time (Figure C4.10bf). Dilutions exceeding 50 will be achieved beyond

a radius of 50 m of the discharge for 50% of the time. A dilution of 100 will be achieved

beyond an area 330 m x 320 m around the outfall for 99% of the time and beyond an area of

230 m x 230 m around the outfall for 95% of the time (Figure C4.9f). Dilutions exceeding 100

will be achieved beyond a radius of 140 m of the discharge for 50% of the time.



All of the discharge options have negligible co-discharge footprints when compared to those at the

Alternative Sites 1 and 2, and are therefore considered to be of low significance. The assessment of low significance is particularly applicable as the CIP cleaning chemicals will not be discharged with the brine.

8.1.3.6 Tainting Effects

Certain chemical constituents are known to taint seafood which may greatly influence the quality

and/or market price of cultured products. Possible tainting effects may affect mariculture activities,

gillnet fisheries and fish factory processing in the bay. Especially the mariculture industry is

extremely sensitive to tainting or even the perception of tainting, and there has thus to be absolute

certainty that such effects are unlikely to occur. Presently the project description as supplied by the

TNPA does not indicate the presence of any such tainting substances. The spatial extent of the

“footprint” of the discharge plumes is relatively localised, however, and in no case do they extend as

far as proposed mariculture activities in either Big Bay or Small Bay (Figure 5.10), or to seawater

intakes for fish factory processing, which are located in the west of Small Bay (Figure 5.12). For the

discharge at depth at Caisson 3, the high density brine is expected to sink to the bottom.

Consequently, mussel spat harvested in surface waters off the caissons during years of poor

recruitment are unlikely to be affected by the discharge. Although gillnet permit holders often fish in

the shallow surf-zone waters immediately adjacent to the iron-ore causeway (S. Lamberth, MCM,

pers. comm.), the limited extent of the discharge footprints at Sites 1 and 2 are unlikely to significantly

affect these activities.

8.2 Assessment of Potential Environmental Impacts

The potential impacts associated with construction activities, the operation of the plant (brine and co-

discharges) and the long-term impacts associated with the intake and discharge structures as

assessed below.

8.2.1 Assessment of Impacts During Construction In the absence of engineering specifications detailing the construction of the intake and discharge

structures (both pipeline or beach wells, boreholes), the following assessment is based on generic

assumptions for such constructions. While beach macrofaunal communities may recover within 2

years, the precautionary principle has been adopted, and the duration of the impact has been rated

as medium. As a result, the outcome of the consequence and significance ratings become “Medium”.

Should the final construction specifications be sufficiently different to those described above, the

impacts associated with the construction phase will need to be re-evaluated. For the intake structure

along the quayside for Site 3 and the proposed discharge structure at Caisson 3, the impacts will be

minimal as these construction activities will be utilising existing infrastructure as their basis and

construction activities will not be extensive. The construction of the borehole intakes along the

causeway are assumed to have no impact on the marine environment



8.2.1.1 Construction of Intake and Discharge Infrastructure

Construction of Intake Beach Wells (Site 1 and Site 2, i.e. development options 1a & 2a)

Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local High Short -

Medium Medium Definite Medium -ve high provided assumptions

correct With mitigation No significant mitigation possible other than avoiding beach well construction

Construction of Intake boreholes (Site 3, i.e. development options 3c & 3d))

Extent Intensity Duration Consequence Probability Significance Status Confidence

Without mitigation Local Low Medium Low Definite Very low -ve

high provided assumptions

correct With

mitigation No mitigation required

Construction of Intake pipeline (Site 1 & Site 2, i.e. development options 1b & 2b)


Without mitigation Local High Short -


correct With

mitigation Limited mitigation is possible using “best practise” mitigation measures during construction. It is not possible to

propose specific mitigation measures based on existing known detail on construction activities

Construction of Intake pipeline (Site 3, i.e. development options 3a & 3b)


Without mitigation Local Low Short Very low probable Very low -ve


correct With

mitigation Limited mitigation is possible using “best practise” mitigation measures during construction. It is not possible to

propose specific mitigation measures based on existing known detail on construction activities

Construction of Discharge Pipeline (Site 2 and Site 3 for discharges into Small Bay and Big Bay only, i.e. development options 2a, 2b, 3a and 3b)


Without mitigation Local Low Short -


correct With

mitigation Limited mitigation is possible using “best practice” mitigation measures during construction. It is not possible to

propose specific mitigation measures based on existing known detail of construction activities

Construction of Discharge infrastructure (Site 1 and Site 3 for discharge at Caisson 3. i.e. development options 1a, 1b, 3c and 3d)


Without mitigation Local Low Short Very low probable Very low -ve


correct With

mitigation No mitigation deemed necessary other than the use of “best practice” mitigation measures during construction. It is not possible to propose specific mitigation measures based on existing known detail of construction activities.



8.2.2 Assessment of Impacts Associated with Brine Discharge

8.2.2.1 Alternative Site 1 At Site 1 several intake and discharge scenarios are under scrutiny. These include intake via

pipelines and beach wells but discharge via a pipeline. These different scenarios are assessed

separately below:

Salinity

A. Pipeline discharge with intake of feed-water from intake beach wells (development option 1a) Extent Intensity Duration Consequence Probability Significance Status Confidence

Without mitigation Local Low Long-term Low Definite Low -ve High

With mitigation No mitigation considered, other than optimal discharge diffuser design

B. Pipeline discharge with intake through intake pipeline (development option 1b) Extent Intensity Duration Consequence Probability Significance Status Confidence

Without mitigation Local Low Long-term Low Definite Low -ve High


Temperature A. Pipeline discharge with intake of feedwater from intake beach wells (development option 1a)

Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local Low Long-term* Low Definite Low -ve High


* The impact occurs during periods of moderate to strong water column stratification, i.e. ~ 9 months per annum.

B. Pipeline discharge with intake through intake pipeline (development option 1b) Extent Intensity Duration Consequence Probability Significance Status Confidence

Without mitigation Local Low Long-term* Low Definite Low -ve High



Dissolved Oxygen The use of sodium metabisulfite as a neutralizing agent for chlorine presently is not proposed for the

Saldanha Bay RO plant as the use of chlorine as a biocide will be limited to the product (i.e.

freshwater produced). Nevertheless the consequences of its potential use, should chlorine ultimately

be considered as the biocide of choice, is assessed As its application is independent of the intake

and discharge scenarios being considered, the assessments below evaluate the potential impacts

both with the addition of an oxygen-scavenging compound as well as without the addition of the

substance.



A. No addition of oxygen-scavenging compounds (development options 1a & 1b) Extent Intensity Duration Consequence Probability Significance Status Confidence


With mitigation Mitigation not deemed necessary unless oxygen scavenging compounds are added to the feed-water

B: With addition of oxygen-scavenging compounds Extent Intensity Duration Consequence Probability Significance Status Confidence

Without mitigation Local Medium Long-term Medium Definite Medium -ve High

With mitigation* Local None Long-term Very Low Definite Very Low -ve High

* Mitigation in this case would constitute aeration of the effluent prior to discharge Oxidising Biocide (NaOCl) A. Pipeline discharge with intake of feed-water from intake beach wells (development options 1a &

1b) Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local Low Long-term Low Definite Low -ve High

With mitigation

If biocide dosage is as specified, no mitigation is likely to be required, other than optimal discharge diffuserdesign

B. Pipeline discharge with intake through intake pipeline (development options 1a & 1b) Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local Medium Long-term Medium Definite Medium -ve High


* The mitigation proposed (if necessary and to be applied to the extent required) is de-chlorination by the use of an oxygen scavenger such as Sodium metabisulphate, however this does potential exacerbate impacts associated with low concentrations of dissolved oxygen in the marine environment

Non-oxidising Biocide (DBNPA) A. Pipeline discharge with intake of feed-water from intake beach wells (development options 1a &

1b) Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local Low Long-term Low Definite Low -ve High

With mitigation

If biocide dosage is as specified, no mitigation is likely to be required, other than optimal discharge diffuser design

B. Pipeline discharge with intake through intake pipeline (development options 1a & 1b) Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local Medium Long-term Medium Definite Medium -ve High

With mitigation* Local None Long-term Low Definite Low -ve Medium

* The proposed mitigation to reduce DBNPA impacts in the marine environment is to design the brine basin so as to ensure greater and sufficient dilution of the DBNPA residuals in the effluent stream before discharge or to revert to the use of an oxidising biocide (chlorine) in this role.



Co-discharged Constituents A. Pipeline discharge with intake of feed-water from intake beach wells (development option 1a)

Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local Low* Long-term Low Definite Low -ve High**

With mitigation*** Local None Long-term Very Low Definite Very Low -ve High**

* Intensity is low in the marine environment provided that the undertaking that the CIP chemicals are disposed off to the sewage (waste treatment) system or to a waste disposal site. Particulates in backwash should be greatly reduced by the use of beach wells.

** High as long as the list of chemicals is adhered to and the the CIP chemicals are disposed off to the sewage (waste treatment) system or to a waste disposal site as stated in the project description

*** Mitigation in unlikely to be required except for perhaps Site 2, should monitoring studies indicate the accumulation of sediments to the extent that it becomes a concern. The assessment with mitigation assumes mitigation measures whereby particulates are removed from the flocculant/backwash sludge to the extent required and disposed elsewhere than the marine environment

B. Pipeline discharge with intake through intake pipeline (development option 1b) Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local Low* Long-term Low Definite Low -ve High**

With *** mitigation Local None Long-term Very Low Definite Very Low -ve High**

* Intensity and the ultimate impact significance is low in the marine environment provided that there is adherence to the undertaking that CIP chemicals are disposed off to the sewage (waste treatment) system or to a waste disposal site is adhered to.

** High as long as the list of chemicals is adhered to and the the CIP chemicals are disposed off to the sewage (waste treatment) system or to a waste disposal site as stated in the project description.

*** The mitigation is to remove at least large particles from the flocculant sludge/backwash to the extent that may be required, should monitoring studies indicate the accumulation of sediments to the extent that it becomes a concern.

8.2.2.2 Alternative Site 2 At Site 2 several intake and discharge scenarios are considered. These include intake via beach

wells or pipelines with discharge via a pipeline. These different scenarios are assessed separately

below:

Salinity A. Pipeline discharge with intake of feed-water from intake beach wells (development option 2a)

Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local Low Long-term Low Definite Low -ve High


B. Pipeline discharge with intake through intake pipeline (development option 2b)

Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local Low Long-term Low Definite Low -ve High




Temperature A. Pipeline discharge with intake of feedwater from intake beach wells (development option 2a)

Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local Low/Medium* Long-term** Low/Medium Definite Low/Medium -ve High

With mitigation No mitigation specified/considered, other than optimal discharge diffuser design

* Intensity could be considered to be medium in that it does exceed the ANZECC guidelines over a small area around the discharge. Based on the South African Water quiality guideline of Δ< 1ºC the plume impinges marginally on eastern boundary of the area demarcated for seaweed harvesting and then only for a total duration of approximately 24 hours per season. Note that the intensity of the thermal impacts at Site 2 may be reduced and assumed to be low i) given appropriate engineering design mitigation (i.e. assuming that it would be possible to locate the pipeline intake in a manner that would ensure sufficiently lower seawater temperatures at the intake) or ii) if it is assumed that beach well intake water temperatures are lower that the near surface water temperatures assumed in the modelling.

** The impact occurs during periods of moderate to strong water column stratification, i.e. ~ 9 months per annum. B. Pipeline discharge with intake through intake pipeline (development option 2b)

Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local Low/Medium* Long-term** Low/Medium Definite Low/Medium -ve High

With mitigation

No mitigation specified/considered, other than optimal discharge diffuser design, however the thermal footprint for this site is significantly larger than the other sites**

* Intensity could be considered to be medium in that it does exceed the ANZECC guidelines over a small area around the discharge. Based on the South African Water quiality guideline of Δ< 1ºC the plume also impinges marginally on eastern boundary of the area demarcated for seaweed harvesting and then only for a total duration of approximately 24 hours per season. * With engineering design mitigation (i.e. assuming that it would be posible to locate the pipeline intake in a manner that would ensure suffieciently lower seawater temperatures at the intake) or assuming that beach well intake water temperatures will be lower that the near surface wtaer temperatures assumed in the modelling, the intensity of the thremal impacts may be reduced and assumed to be low.

** The impact occurs during periods of moderate to strong water column stratification, i.e. ~ 9 months per annum. The

significance of this impact is low, however should mitigation be deemed to be necessary at Site 2, the mitigation would be to locate the intake to ensure lower seawater temperatures at the intake. This would, however, most likely lead to lower plant efficiencies.

Dissolved Oxygen The use of sodium metabisulfite as a neutralizing agent for chlorine presently is not proposed for the






substance.

A. No addition of oxygen-scavenging compounds (development options 2a & 2b) Extent Intensity Duration Consequence Probability Significance Status Confidence



B: With addition of oxygen-scavenging compounds

Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local Medium Long-term Medium Definite Medium -ve High


* Mitigation in this case would constitute aeration of the effluent prior to discharge



Oxidising Biocide (NaOCl) A. Pipeline discharge with intake of feed-water from intake beach wells (development option 2a) Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local Low Long Low Definite Low -ve High

With mitigation


B. Pipeline discharge with intake through intake pipeline (development option 2b) Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local* Medium Long-term Medium Definite Medium -ve High

With mitigation** Local None Long-term Very Low Definite Very Low -ve High

* Although local the achievable dilutions are significantly lower than the other sites. This poses greater risks in terms of potential impacts by cumulative, synergistic effects of pollutants as well as potential by-products of biocides.

** The mitigation proposed (if deemed necesarry) is de-chlorination by the use of an oxygen scavenger such as Sodium metabisulphate, however this does potentially exacerbate impacts associated with low concentrations of dissolved oxygen in the marine environment

Non-oxidising Biocide (DBNPA) A. Pipeline discharge with intake of feed-water from intake beach wells (development option 2a) Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local Low Long Low Definite Low -ve High

With mitigation


B. Pipeline discharge with intake through intake pipeline (development option 2b) Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local* Medium Long-term Medium Definite Medium -ve High

With mitigation** Local None Long-term Low Definite Low -ve Medium

* Although local the achievable dilutions are significantly lower than the other sites. This poses greater risks in terms of potential impacts by cumulative, synergistic effects of pollutants as well as potential by-products of biocides.

** The proposed mitigation to reduce DBNPA impacts in the marine environment is to design the brine basin so as to ensure greater and sufficient dilution of the DBNPA residuals in the effluent stream before discharge or to revert to the use of an oxidising biocide (chlorine) in this role.

Co-discharged Constituents A. Pipeline discharge with intake of feed-water from intake beach wells (development option 2a)

Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local* Low** Long Low Definite Low -ve High***

With mitigation**** Local None Long Very Low Definite Very Low -ve HIgh***

* Although local the achievable dilutions are significantly lower than at the other sites. This poses greater risks in terms of potential impacts by cumulative, synergistic effects of pollutants as well as potential by-products of biocides or co-discharges

** Intensity is low in the marine environment provided that the undertaking that the CIP chemicals are disposed off to the sewage (waste treatment) system or to a waste disposal site. Particulkates in backwash should be greatly reduced by the use of beach wells.

*** High as long as the list of chemicals is adhered to and the the CIP chemicals are disposed off to the sewage (waste treatment) system or to a waste disposal site as stated in the project description

**** The mitigation is to remove at least large particles from the flocculant sludge, however this is unlikely to be necessary unless the discharge of the backwash materials proves prolematic.



B. Pipeline discharge with intake through intake pipeline (development option 2b) Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local* Low/Medium** Long Low/Medium Definite Low/Medium -ve High***

With **** mitigation Local None Long Very Low Definite Very Low -ve High***

* Although the impact remains local, the achievable dilutions are significantly lower than at the other sites. This poses greater risks in terms of potential impacts by cumulative, synergistic effects of pollutants as well as potential by-products of biocides or co-discharges

** Intensity is low in the marine environment provided that the undertaking that the CIP chemicals are disposed off to the sewage (waste treatment) system or to a waste disposal site. However, in the relatively quiescent environment of Site 2 however, backwash and flocculant material will not be dispersed to the same extent at other sites and thus pose greater water and sediment quality risks than at other proposed sites, suggesting a medium intensity rating.

*** High as long as the list of chemicals is adhered to and the the CIP chemicals are disposed off to the sewage (waste treatment) system or to a waste disposal site as stated in the project description

***** The mitigation is to remove at least large particles from the flocculant sludge/backwash to the extent that may be required, should monitoring studies indicate the accumulation of sediments to the extent that it becomes a concern.

8.2.2.3 Alternative Site 3 At Site 3 several intake and discharge scenarios are under scrutiny. These include intake and

discharge via pipelines, with two discharge options being considered, namely a bottom discharge

though a single or multi-port diffuser at ~8 m depth into Small Bay and a discharge though a single or

multi-port diffuser at ~4 m depth into Big Bay. An additional scenario considers intake of the feed-

water through two alternative boreholes situated on the quay, with a subsequent pipeline discharge at

-17 m into Big Bay at Caisson 3. These different scenarios are assessed separately below:

Salinity A. Pipeline intake and discharge into Small Bay

Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local Low Long Low Definite Low -ve High


B. Pipeline intake and discharge into Big Bay Extent Intensity Duration Consequence Probability Significance Status Confidence

Without mitigation Local Low Long Low Definite Low -ve High


C. Borehole intake and discharge at Caisson 3





Temperature A. Pipeline intake and discharge into Small Bay

Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local Low Long-term* Low Definite Low -ve High







C. Borehole intake and discharge at Caisson 3 Extent Intensity Duration Consequence Probability Significance Status Confidence

Without mitigation Local Low Long Low Definite Low -ve Medium*


* The groundwater report seems to confirm the assumption that the water temperature from the boreholes would be close to that of the seasonal mean sea surface temperatures.

Dissolved Oxygen No addition of oxygen scavenging compounds

A. Pipeline intake and discharge into Small Bay Extent Intensity Duration Consequence Probability Significance Status Confidence









Dissolved Oxygen With addition of oxygen scavenging compounds

The use of sodium metabisulfite as a neutralizing agent for chlorine presently is not proposed for the








substance.

A. Pipeline intake and discharge into Small Bay Extent Intensity Duration Consequence Probability Significance Status Confidence

Without mitigation Local Medium Long Medium Definite Medium -ve High

With mitigation* Local None Long Very Low Definite Very Low -ve High

* Mitigation in this case would constitute aeration of the effluent prior to discharge B. Pipeline intake and discharge into Big Bay

Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local Medium Long Medium Definite Medium -ve High


* Mitigation in this case would constitute aeration of the effluent prior to discharge C. Borehole intake and discharge at Caisson 3

Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local Medium Long Medium Definite Medium -ve High


* Mitigation in this case would constitute aeration of the effluent prior to discharge Oxidising Biocide (NaOCl) A. Pipeline intake and discharge into Small Bay

Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation

Local Low Long Low Definite Low -ve High


* The mitigation proposed (if necesarry) is de-chlorination by the use of an oxygen scavenger such as Sodium metabisulphate, however this does potential exacerbate impacts associated with low concentrations of dissolved oxygen in the marine environment


Without mitigation



* The mitigation proposed (if necesarry) is de-chlorination by the use of an oxygen scavenger such as Sodium metabisulphate, however this does potential exacerbate impacts associated with low concentrations of dissolved oxygen in the marine environment

C. Borehole intake and discharge at Caisson 3 Extent Intensity Duration Consequence Probability Significance Status Confidence

Without mitigation


With mitigation*


* It is not expected that mitigation will be required due to a reduced requirement for a biocide in intake waters from boreholes



Oxidising Biocide (NaOCl) A. Pipeline intake and discharge into Small Bay



With mitigation* Local None Long Very Low Definite Low -ve Medium


B. Pipeline intake and discharge into Big Bay



With mitigation* Local None Long Low Definite Low -ve Medium





With mitigation*


* It is not expected that mitigation will be required due to a reduced requirement for a biocide in intake waters from boreholes

Co-discharged Constituents A. Pipeline intake and discharge into Small Bay

Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local Low* Long Low Definite Low -ve High**

With mitigation Local None Long Very Low Definite Very Low -ve HIgh**

* *Intensity and the ultimate impact significance is low in the marine environment provided that there is adherence to the undertaking that CIP chemicals are disposed off to the sewage (waste treatment) system or to a waste disposal site is adhered to.



B. Pipeline intake and discharge into Big Bay


With *** mitigation Local None Long Very Low Definite Very Low -ve High**








With mitigation*** Local None Long Very Low Definite Very Low -ve High**



*** The mitigation is to remove at least large particles from the flocculant sludge/backwash to the extent that may be required, should monitoring studies indicate the accumulation of sediments to the extent that it becomes a concern. For boreholes it is highly unlikely to be a requirement.

8.2.3 Assessment of Impacts Associated with Intake and Discharge

Structures

Entrainment of Biota (Beach well intakes, i.e. development options 1a & 2a)

Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local None None Not significant Improbable Insignificant Neutral High

With mitigation No mitigation deemed necessary

Entrainment of Biota (Pipeline Intake, i.e. development options 1b, 2b, 3a & 3b)

Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local Medium Long-term Medium Definite Medium -ve High

With mitigation Local Low Long-term Low Definite Low -ve High

* Mitigation here constitutes appropriate engineering design to avoid entrainment at the intake. Entrainment of Biota (Boreholes intakes, i.e. development options 3c & 3d)

Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local None None Not significant Improbable Insignificant Neutral High

With mitigation No mitigation deemed necessary

Flow Distortion (Pipeline Discharge)

Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local Low Long-term Low Definite Low -ve Low*

With mitigation

Insufficient detail in project description to define specific mitigation measures, however appropriate engineering design should negate any potential impacts

* Low as insufficient detail in project description, however, it should be possible to largely mitigate any potential impacts by appropriate engineering design resulting in a very low to non-existent significance.



Sediment Dynamics (Pipeline Discharge) Site2 and Site 3 with discharges into Small and Big Bay

Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation Local Low Long-term Low Definite Low -ve Low*

With mitigation No mitigation can be recommended as insufficient detail in project description

* Low as insufficient detail in project description, however it should be possible to largely mitigate any potential impacts by appropriate engineering design resulting in a very low to non existent significance.

Site1 and Site 3 with discharges at Caisson 3

Extent Intensity Duration Consequence Probability Significance Status Confidence Without mitigation None* None* None Not significant Improbable insignificant -ve Low**

With mitigation No mitigation required

* Assumes that the discharge at Caisson 3 will be located in a position such that it will not affect bottom sediments

8.3 “No-development” Alternative

The “no-development” alternative implies that the RO Plant and associated infrastructure will not be

commissioned. From a marine perspective this is undeniably the preferred alternative, as all impacts

associated with beach disturbance and effluent discharge will no longer be an issue. This must,

however, be seen in context with existing proposed developments for extensions to the Port, as well

as the use of possible alternative water sources for dust control. Furthermore, it needs to be weighed

up against the potential positive socio-economic impacts undoubtedly associated both with the RO

Plant project itself, as well as the Port extension.

8.4 Project Impacts and Environment Interaction Points

Figures 7.3 – 7.6 illustrate the maximum extents of the brine, thermal and biocide plumes, and

achievable dilutions exceeding 50 times of potential co-discharges, respectively, at the three

alternative sites (and the three alternative discharge scenarios at Site 3) under the ‘worst-case-

scenario’ conditions of predominantly extended calm periods during autumn but in many cases also

during the more highly stratified conditions prevailing during the summer months. These are shown

in relation to areas of potentially threatened or sensitive habitats and other beneficial users of the

marine environment in Saldanha Bay, which were pointed out in Section 5.2.7.

From these figures it can be seen that the plume footprints do not extend as far as any existing or

proposed mariculture activities, seawater intakes for fish processing factories, recreational and

commercial gill-netting / trek-netting areas, or Marine Protected Areas and National Parks (i.e.

including the Langebaan Lagoon). At Site 2, however, the plumes do extend over the eastern

boundary of the area demarcated for seaweed harvesting.






9 CONCLUSIONS AND RECOMMENDATIONS

9.1 Environmental Acceptability and Comparison of Alternatives

The environmental acceptability of the development alternatives are outlined below.

Site 1

From a coastal and marine environmental perspective, Site 1 is the most “pristine” of the three

alternative sites. Although some litter is present on the beach and in the dunes, regular pedestrian

and vehicular traffic appears minimal, as access to the site is restricted. As a consequence, the

sheltered cove is used by coastal birds as a resting and feeding place, despite the nearby harbour

activities. From a visual assessment undertaken during the site visit, the beach appears dissipative

to intermediate with low to moderate wave action. Macrofaunal communities inhabiting the beach are

therefore expected to be relatively diverse, and the surf-zone may serve as a nursery area for fish.

Construction activities (beach well and/or pipeline construction) as part of the proposed development

will severely impact the beach and nearshore habitats and their associated communities. However,

provided construction activities are not phased over an extended period, the beach is not repeatedly

disturbed through persistent activities and suitable post-construction rehabilitation measures are

adopted, the macrofaunal communities are likely to recover in the short-to medium-term. The highly

localised, yet significant impacts over the short-term thus need to be weighed up against the long-

term benefits of using subsurface intakes at this site. The beach intake wells or surf-zone infiltration

galleries proposed for this site will result in pre-filtration of feed-water, thereby significantly reducing

(but not eliminating) the need for the extensive use of biocides and co-pollutants (associated with

ensuring appropriate intake water quality and for cleaning of processes resulting from lower quality

intake waters), and eliminating the impact of impingement and entrainment of biota. Furthermore, a

discharge along an extended length (30 m or more) on the revetment, together with the close

proximity to the outer edge of the surf-zone, will ensure effective dispersion of the effluent. It is

important, however, that the discharge does not occur into the surf-zone as surf-zone trapping is

likely to significantly reduce the overall dispersion of the effluent despite the greater likelihood of the

effluent being dispersed throughout the vertical extent of the water column in the surf-zone. Should

the discharge occur into the surf-zone or along the shoreline, the non-thermal impacts are likely to be

more extensive, extending in an alongshore direction away from the reclaim dam. This may

negatively affect gillnet/treknet-fishing activities in the immediate vicinity of the discharge.

Site 2

The shoreline and beach at Site 2 are already severely disturbed and altered by existing port

development and marine litter. While pipeline construction activities will nonetheless impact beach

communities, the effects are likely to be less severe than at Site 1 due to the already severely

disturbed nature of the site.



Perhaps of greatest concern at this site is the issue of water quality. This applies both to the feed-

water, which will be drawn from Small Bay, as well as the effects of the effluent to be discharged

back into Small Bay. Due to reduced flushing rates and discharges of organic-rich effluents from fish

processing factories, eutrophication and anoxia have been reported, and target limits for pathogenic

microorganisms are often exceeded. The clock-wise circulation in Small Bay under both SE and NW

wind conditions (Figures 5.5a & 5.5b) may result in the anoxic waters extending into the eastern

corner of Small Bay, thereby potentially affecting the quality of the feed-water for the RO plant. This

may require the additional implementation of filters and/or use of chemical purifiers in the RO plant. A

continuous discharge of brine into Small Bay is also likely to further aggravate the water quality

situation within the bay, especially when considering the comparatively large extent of discharge

plumes at Site 2 relative to those predicted at the other alternative sites. Reduced circulation within

the eastern corner of Small Bay, particularly the deeper dredged areas in close proximity of the

discharge may also result in accumulation of the high density brine due to insufficient flushing.

Although impacts due to the effluents discharge are deemed not to be significant, the risks of

potentially impacting important activities such as mariculture activities in Small Bay (through potential

eutrophication, biocide and associated break-down products, co-discharges, etc) and surf-zone

gillnet/treknet fisheries adjacent to the causeway would be minimised by avoiding a discharge into

one of the “quieter” corners of the relatively poorly flushed Small Bay (i.e. relatively poorly flushed

compared to Big Bay). The quality of feed-water at the intakes to fish processing factories in the west

of Small Bay is, however, unlikely to be affected.

Site 3

As Site 3 is located on the existing iron-ore causeway, intake and discharge pipelines will be

constructed in an already artificial and altered environment. Intertidal and shallow subtidal biota will

be minimally impacted, and no beaches will be directly disturbed or macrofaunal assemblages

eliminated. Disturbance of shorebirds will also be minimal, and the impacts of construction will be

certainly the least of the three alternative options.

The proximity of shipping traffic and an intake from Small Bay, however, again raises questions as to

the quality of the feed-water.

Discharge into Small Bay at depth (-8 m) is, however, unlikely to contribute to deteriorating water

quality within the bay, as the proposed discharge will be positioned in an area where current speeds

are comparatively high, that is closer to the entrance to Small Bay, and where mixing of the water

column through propeller-wash from shipping activity can be expected. Discharge water from the RO

Plant water discharged into Big Bay (-4 m) or at depth at Caisson 3 is also likely to be rapidly mixed

as a result of exposed nature of the discharge locations in relatively high wave conditions and

increased currents. These alternative discharges into Big Bay and at Caisson 3 from Site 3 are likely

to be the options with the least potential water quality impacts due to them being discharged into the

more dispersive Big Bay environment that also is subject to little water quality concerns compared to



Small Bay. The discharge at Caisson 3 is the preferred discharge location due to being in deep water (allowing more options in terms of discharge design and depth), at a site with strong tidal flows and in a location where any accumulation of brine would be mitigated by propeller wash. Site 3 provides the greatest options in terms of flexibility of determining the depth of discharge with very short pipeline lengths and limited or no disruption of shoreline environments during construction.

Of the three alternative sites, Site 3 is the most environmentally acceptable in terms of construction

activities. This holds also for a comparison of the sites when intake and discharge are done through

pipelines as in most cases the discharge plumes are smallest for Site 3 (particularly the alternate

discharge into Big Bay). This, however, needs to be seen in context with future port developments

and long-term discharges of co-pollutants and biocides in the brine, compared to the option of using

beach well or surf-zone intakes at Sites 1 and 2. The borehole options considered for Site 3 are

likely to have the same advantage in terms of minimising the requirement for the use of biocides to

prevent biofouling and also entrainment of biota.

“No-development” Alternative

In the case of the “no-development” alternative, disturbance and elimination of beach and shallow

subtidal macrofauna, through pipeline and beach well installation will not occur. Anthropogenic

activities on the beach will be limited, shorebirds feeding and nesting in the area will remain

undisturbed and dune vegetation preserved. Likewise, with the “no-development” alternative, no

brine effluent (and associated co-discharges) will be released into the marine environment, and the

risks associated with such a discharge will thus be absent.

Should development proceed, however, the marine biota at the site ultimately selected will be

affected by the hypersaline effluent (albeit spatially constrained) regardless of the final choice of site,

however only to the degree indicated for each of the proposed sites.

9.2 Preferred Alternative

Preliminary results indicate that Site 3 with a discharge at Caisson 3 (and with a borehole intakes to

minimise biocides and entrainment of biota) is the option with the least potential impacts (see Table

9.1), followed by Site 3 (with a discharge into Small Bay or Big Bay), and then Site 1. The least

preferable option is Site 2. Provided that beach wells intakes are installed at Site 1 and that these

beach wells result in significantly lowered use of biocides and reduced co-discharge of cleaning

chemicals, Site 1 is preferred over Site 3 with a discharge into Small Bay (and an assumed pipeline

intake. However given the possibility that intake boreholes are possible for all Site 3 discharge

options, the rankings are as given under the heading “Overall Ranking” at the bottom of Table 9.1. It

should be noted that, despite that fact that Site 3 with a discharge into Small Bay is assessed to not

have significant impacts, the discharge will be into Small Bay, which is perceived to be relatively



poorly flushed and also is perceived to be susceptible to poor water quality due to existing discharges

into Small Bay.

Although Site 2 with its discharge into the NE corner of Small Bay results in the largest effluent plume

“footprints” of all of the proposed sites and associated intake and discharge combinations, it could be

deemed to be acceptable for the present project description. It is, however, not recommended due to

the higher risks associated with a discharge into a relatively quiescent region of the relatively poorly

flushed Small Bay.

The rankings are based on a ranking of one to five. Where the ranking of two options is the same,

i.e. both second best, then a ranking of third best is not allocated. This means that the lowest ranked

option will always have a ranking of 5 unless there is an equally undesirable option (i.e. both will then

be ranked a 4 and there will be no ranking of 5). The ranking themselves do not imply a weighting

indicating the relative magnitude of impacts. The assessment is thus qualitative as is the overall

ranking.



Table 9.1: Summary of potential impacts for the various Sites and associated intake and discharge combinations.

Assessment Criteria Site 1

Site 2

Site 3 (Small Bay)

Site 3 (Big Bay)

Site 3 (Caisson 3)

Dispersive nature of the site

Ranking: 4 Discharge occurs beyond surf-zone into an exposed (but nevertheless relatively quiescent) NW corner of Big Bay.

Ranking: 5 Discharge occurs into shallow waters in a quiescent corner of the relatively poorly flushed Small Bay. Limited wave and current action to disperse any particulates in the discharge

Ranking: 2

Discharge occurs in deeper waters in the relatively poorly flushed Small Bay, but closer to the mouth of the bay and at a location where flows are quite strong

Ranking: 2 Discharge occurs in deeper waters in the relatively more exposed and well-flushed Big Bay but in shallower waters

Ranking: 1 Discharge occurs in relatively deep water in a location with strong tidal flows and potential mitigation factors such as propeller wash to disperse effluents

Design Options

Ranking: 4 Site 1 allows for beach well intakes that have specific advantages in terms of minimizing biocide discharges and co-discharges and eliminating impingement and entrainment Design options are limited (i.e. being a surface discharge no sub-surface multi-port diffuser structure can be considered. This location in unlikely to interfere with future port development options

Ranking: 5 Site 2 allows for beach well intakes that have specific advantages in terms of minimizing biocide discharges and co-discharges and eliminating impingement and entrainment. Design options are limited (discharge in principle could occur in deeper waters than assessed.) The potential limitations imposed by proposed port development need to be considered

Ranking: 1 Pipeline intakes only considered. Could consider borehole intake option that has the same advantages as beach wells in terms of minimizing biocide discharges and co-discharges and eliminating impingement and entrainment Design options for discharge structure could include single- or multi-port diffusers The potential limitations imposed by proposed port development need to be considered

Ranking: 1 Pipeline intakes only considered. Could consider borehole intake option that has the same advantages as beach wells in terms of minimizing biocide discharges and co-discharges and eliminating impingement and entrainment Design options for discharge structure could include single- or multi-port diffusers The potential limitations imposed by proposed port development need to be considered

Ranking: 1 Borehole intakes option considered have the same advantages as beach wells in terms of minimizing biocide discharges and co-discharges and eliminating impingement and entrainment Design options for discharge structure could include single- or multi-port diffusers The potential limitations imposed by proposed port development need to be considered

Sensitivity of discharge location

Ranking: 3

Discharge occurs into Big Bay. Sensitive surf-zone could be excluded. No

proximity to existing mariculture operations

Ranking: 5

Discharge occurs in a quiescent corner of Small Bay that already contains

a number of other discharges and also sensitive mariculture

operations

Ranking: 3

Discharge occurs into Small Bay that already contains a number of

other discharges and also sensitive mariculture

operations

Ranking: 2

Discharge occurs into Big Bay. Sensitive surf-zone could be excluded. No

proximity to existing mariculture operations

Ranking: 1

Discharge occurs in deep water at a highly

dispersive location away from sensitive sites




Site 2

Site 3 (Small Bay)

Site 3 (Big Bay)

Site 3 (Caisson 3)

Construction Impacts

Construction of Beach Well intakes

no mitigation Medium Medium n/a n/a n/a

with mitigation No significant mitigation possible other than avoiding beach well construction n/a n/a n/a

Construction of borehole intakes

no mitigation n/a n/a n/a n/a Very low

with mitigation n/a n/a n/a n/a No mitigation required

Construction of Pipeline intakes

no mitigation Medium Medium Very low Very low n/a

with mitigation Limited mitigation is possible using “best practice” mitigation measures during construction. It is not possible to propose

specific mitigation measures based on existing known detail of construction activities

None deemed necessary other than normal environmental ‘best practice’ n/a

Construction of discharges (Pipelines for Site 2 and Site 3 for discharges into Small Bay and Big Bay and other discharge infrastructure at Site 1 and Site 3 for discharge at Caisson 3) no mitigation Very low Medium Medium Medium Very low

with mitigation

None deemed necessary other than the use of “best

practice” mitigation measures during construction.

Limited mitigation is possible using “best practice” mitigation measures during construction. It is not possible to propose specific mitigation measures based on existing known detail of

construction activities

None deemed necessary other than the use of “best

practice” mitigation measures during construction.

Operational Impacts (other than Water Quality) of Intake and discharge operation

Entrainment of Biota (Pipeline Intakes)

no mitigation Medium Medium Medium Medium n/a

with mitigation Low Low Low Low n/a

Entrainment of Biota (Beach Well Intakes)

no mitigation insignificant insignificant n/a n/a n/a

with mitigation No mitigation deemed necessary n/a n/a n/a

Entrainment of Biota (Borehole intakes along Causeway)

no mitigation n/a n/a insignificant insignificant insignificant

with mitigation n/a n/a No mitigation deemed necessary




Site 2

Site 3 (Small Bay)

Site 3 (Big Bay)

Site 3 (Caisson 3)

Flow Distortion


with mitigation Insufficient detail in project description to define mitigation measures, how appropriate engineering design should negate any potential impacts

Impacts on Sediment Dynamics

no mitigation insignificant Low Low Low insignificant

with mitigation No mitigation required No detailed mitigation can be recommended as insufficient detail in project description, however appropriate engineering design should negate any potential impacts

No mitigation required

Specific Impacts associated with the discharged brine (and associated co-discharged)

Salinity



Temperature

no mitigation Low Low/Medium* Low Low Low


Oxygen (no O2 scavengers)


with mitigation Mitigation not deemed necessary unless oxygen scavenging compounds are added to the feed-water. This will only occur should chlorine be used as a biocide that results in a potential discharge of residual chlorine into the marine environment. Presently the project description specifies that chlorine will not be used in

this role

Oxygen (with O2 scavenger)

no mitigation Medium Medium Medium Medium Medium

with mitigation Very Low Very Low Very Low Very Low Very Low

Oxidising Biocides (NaOCl): Beach well or borehole intakes


with mitigation If biocide dosage is as specified, no mitigation is likely to be required, other than optimal discharge diffuser design n/a n/a

No mitigation likely to be required, other than optimal





Site 2

Site 3 (Small Bay)

Site 3 (Big Bay)

Site 3 (Caisson 3)

Oxidising Biocides (NaOCl): Pipeline intakes

no mitigation Medium Medium Low Low n/a

with mitigation Very Low Very Low Very Low Very Low n/a

Non-oxidising Biocides (DBNPA): Beach well or borehole intakes


with mitigation*** If biocide dosage is as specified, no mitigation is likely to be required, other than optimal discharge diffuser design

n/a n/a No mitigation likely to be

required, other than optimal discharge diffuser design

Non-oxidising Biocides (DBNPA): Pipeline intakes

no mitigation Medium Medium Low Low Low

with mitigation*** Low Low Very Low Very Low Very Low

Co-discharges (beach well or borehole intakes) no mitigation Low Low n/a n/a Low

with mitigation** Very Low Very Low n/a n/a Very Low

Co-discharges (pipeline intakes) no mitigation Low Low//Medium Low Low n/a

with mitigation**** Very Low Very Low Very Low Very Low n/a

Overall Ranking

All criteria 4 5 3 2 1

* Note that the intensity of the thermal impacts at Site 2 may be reduced and assumed to be low, i) given appropriate engineering design mitigation (i.e. assuming that it would be possible to locate the pipeline intake in a manner that would ensure sufficiently lower seawater temperatures at the intake) or ii) if it is assumed that beach well intake water temperatures are lower that the near surface water temperatures assumed in the modelling.

** A proposed mitigation (if necessary and to be applied to the extent required) is de-chlorination by the use of an oxygen scavenger such as Sodium metabisulphate, however as this does potential exacerbate impacts associated with low concentrations of dissolved oxygen in the marine environment, the mitigation preferred is that of carefully monitored and managed dosing to ensure minimal chlorine concentrations in the discharge.

*** The proposed mitigation to reduce DBNPA impacts in the marine environment is to design the brine basin so as to ensure greater and sufficient dilution of the DBNPA residuals in the effluent stream before discharge or to revert to the use of an oxidising biocide (chlorine) in this role.

**** Mitigation in unlikely to be required except for perhaps Site 2, should monitoring studies indicate the accumulation of sediments to the extent that it becomes a concern. The assessment with mitigation assumes that particulates are removed from the flocculant/backwash sludge to the extent required.



9.3 Recommendations

The recommendations from this study include mitigation measures (optional and required) and

monitoring requirements to be able to better assess and manage potential impacts.

9.3.1 Mitigation Measures The recommended mitigation measures are listed below for both the construction and operational

phases of the RO Plant.

9.3.1.1 Construction Impacts Heavy vehicle traffic associated with pipeline and well construction on the beach must be kept to a

minimum, and be restricted to clearly demarcated access routes and construction areas only. All

construction activities in the coastal zone must be managed according to a strictly enforced

Environmental Management Plan. Good house-keeping must form an integral part of any

construction operations on the beach from start-up, including, but not limited to:

• drip trays under all vehicles parked on the beach;

• no vehicle maintenance or refuelling on beach;

• accidental diesel and hydrocarbon spills to be cleaned up accordingly; and

• no concrete mixing on beach.

9.3.1.2 Operational Impacts Intake In the case of intake pipelines, manual cleaning of the intake structure and sump will be necessary as

marine growth, scaling and sediment settlement will occur. Most marine pipelines employ a pigging

system for regular maintenance cleaning, in which a ‘pig’ (bullet-shaped device with bristles) is

introduced into the pipeline to mechanically clean out the structure. For intake systems, the pigging

device is introduced at the intake structure and allowed to travel to the sump, from where it is

retrieved. For the discharge pipeline, it is introduced in the RO plant, and is removed again by divers

on the seaward side. Pigging is optional, but should be considered as it can reduce the costs for

biocides.

The use of subsurface intakes such as beach wells or infiltration galleries or boreholes as an

alternative to pipelines is recommended where feasible and, in the case of the first two, where these

will not cause significant negative impacts to beach topography or freshwater aquifers. While these

may have higher initial construction costs, they are likely to result in long-term operational savings

due to reduced pre-treatment and chemical requirements. They also operate with minimal effects on

local marine life in the form of entrainment and impingement. They may also be able to operate at

times when pipeline intakes need to be shut-down due to unsuitable seawater conditions, or during

port-development activities such as dredging.



In the case of pipeline intakes, there are several alternative design or mitigation measures that can

completely avoid or reduce the impact of impingement. Intake velocities should be kept below ~0.15

m.s-1 to ensure that fish and other organisms can escape the intake current. This can be achieved

through a combination of pumping rates and intake design. Alternatively, the use of velocity caps

(concrete structures placed over the intake leaving a gap between the cap and intake) change the

predominant intake flow from vertical to horizontal, thereby significantly reducing impingement of fish,

which are better able to detect a horizontal change in water velocity. These require little ongoing

maintenance once installed. Further mitigation options involve screens, which are specifically sized

to prevent fish from entering the system while still allowing adequate water flow. Travelling screens

enable fish to be moved out of an intake system, often unharmed (California Coastal Commission

2004). These are generally installed at the landward end of a pipeline intake and built in conjunction

with a fish return system, which routes fish and a portion of the feed-water back to the ocean. Such

systems, however, involve ongoing maintenance and personnel to operate them.

Discharges Possible mitigation measures for brine effluents include alternative discharge options such as

evaporation ponds, dilution with sewage effluents, deep well injection, crystallisation, live-stock

watering and beach well disposal (DWAF 2007). Beach well disposal was considered at Site 1,

however the results of the geotechnical study ascertained that sediment characteristics were

unsuitable and that this discharge option was thus not feasible (Visser et al., 2007). The other

alternative discharge options (e.g. evaporation ponds, crystallisation, etc) are discussed in the Basic

Assessment Report for the proposed RO Plant (PDNA/SRK JV, 2008)

Mitigation measures should also include specific selection of technologies and processes that

minimise or eliminate the need for hazardous chemicals. This will not only reduce the disposal

requirements for such substances, but also lessen the impacts of potential spills or releases from the

RO plant, thereby reducing discharges of hazardous components into the bay. Every effort should

thus be made to select the least environmentally damaging option for feed-water treatment and

cleaning of plant components.

Although the footprint of the biocide plume in general is not substantial, marine organisms are

extremely sensitive to residual chlorine. Considering the relatively close proximity of mariculture

activities and commercial and recreational fisheries, it is important, therefore, to ensure that the

residual chlorine (or other non-oxidising biocide) concentration in the discharged purge water are

specifically reduced to a level below that which may have lethal or sublethal effects on the biota,

particularly the larval stages. Should the exceedance of the recommended guideline be a more

persistent or recurrent event, there could be serious implications for nearby mariculture operation.

Taking the dilution factor of the brine into account, the effects of biocide must be actively minimised

by ensuring that biocide concentrations do not exceed the No Observed Effect Concentration

(NOEC) and/or the relevant water quality target values. This can be achieved with appropriate



design and implementation of mitigation measures, to ensure that the biocide concentrations comply

with the relevant water quality guidelines. The impacts due to chlorine can be mitigated by de-

chlorination of the purge water with, for example sodium metabisulfite, before discharge into the

marine environment. This may though not always succeed in eliminating chlorine toxicity as there is

the possibility that the presence of organic substances in the environment results in the production of

chlorinated residuals that are resistant to sodium metabisulfite de-chlorination (Yonkos et al., 2001).

The use of sodium metabisulfite as a neutralizing agent for chlorine is further associated with oxygen

depletion in the effluent, as the substance is an oxygen scavenger. Given that chlorine presently is

not the biocide of choice for the RO Plant and that it is intended to use only a non-oxidising biocide

(DBNPA – 2.2 Dibromo-3-nitrilopropionamide), it is unlikely that sodium metabisulfite will be required as a

neutralising agent. However should chlorine become the biocide of choice and sodium metabisulfite

be required as a neutralising agent , aeration of the effluent prior to discharge is recommended,

particularly if the discharge is into Small Bay which already experiences a fairly regular oxygen deficit

in bottom waters during the late summer and winter months. A better option would be carefully

monitored dosing to ensure minimal chlorine concentrations in the discharge.

Mitigation measures to ensure low residuals of DBNPA in any discharge to the marine include

appropriate design of the brine basin so as to ensure greater and sufficient dilution of the DBNPA

residuals in the effluent stream before discharge or to revert to the use of an oxidising biocide

(chlorine) in this role. However, once again a better option would be carefully monitored dosing to

ensure minimal DBNPA concentrations in the discharge.

No acute toxic effects of the backwash sludge containing sediment particles and the flocculant Ferric

Chloride are generally expected. However, ferric chloride may cause a discoloration of the receiving

water, and the sludge discharge may lead to increases in turbidity and suspended matter. Impacts

such as reduced primary production or burial of sessile organisms by increased turbidity in the

discharge may thus occur. A process for removal of the particles is thus recommended before

discharge of the sludge. If this is not possible, dilution of the backwash waters by blending with the

brine flows is a potentially viable, but less desirable, alternative. However, should monitoring studies

indicate the accumulation of sediments from the flocculant/backwash sludge to the extent that it

becomes a concern (most likely only to be a concern for the sheltered Site 2), a process for

particulate removal will need to be applied to the extent necessary

The model results are based on assumed conceptual designs that assume active mixing of the dense

brine into the water column during discharge. While conceptually surf-zone discharge or a single port

diffuser has been assumed, the discharge of brine particularly at the deeper discharge locations

needs to be through a diffuser or single port discharge design that results in near-field behaviour

consistent with that assumed in the modelling. A conservative discharge design nominally would be

one where initial dilutions of 50 are predicted by near-field models such as CORMIX (Jirka et al.,

1996) and UOUTPLM (Baumgartner et al., 1971). It is particularly important that the development of



a coherent density flow of brine along the seabed is avoided by ensuring as complete mixing

throughout the full extent of the water column as is possible at the point of discharge.

The optional and required mitigation measures are summarised in Table 9.2 below:

Table 9.2: Optional and required mitigation measures


Construction Impacts: - Limiting and restricting vehicular traffic on the

beach - Good house-keeping - Active rehabilitation above high water mark

Required

Required Required

Use of sub-surface intakes (beach wells/boreholes)

Required (highly recommended)

Pipeline intakes: - adjustment of intake velocities and/or velocity

caps - intake screens - operational cleaning (pigging) of intake pipes

Required

Required Optional

Discharges: - evaporation ponds / crystallisation - beach well disposal - Carefully controlled dosing of biocides based on

feedback from monitoring systems - dechlorination of purge water

- Reduction of residual DBNPA concentration in effluent to be discharged by designing the brine basin so as to ensure greater and sufficient dilution of the DBNPA residuals in the effluent stream before discharge, or to revert to the use of an oxidising biocide (chlorine) in this role.

- aeration of purge water

Not feasible Not feasible

Required

Required only if chlorine is used in systems discharging to the marine environment (presently not indicated) and then only in the case of pipeline intakes as chlorine dosing for beach well and borehole intakes is assumed to be low Required only in the case of pipeline intakes as DBNPA dosing for beach well and borehole intakes is assumed to be low Optional but highly recommended if chlorine is used in systems discharging to the marine environment (presently not indicated) and sodium metabisulfite is used to neutralise chlorine residuals in the effluent stream discharged to the marine environment




- Removal of sludge particles from backwashing

of RO modules - Optimal diffuser design based on acceptable

water quality target values

Optional, to be undertaken to the extent required (i.e. as informed by the monitoring results for the discharge.) Required

9.3.2 Monitoring Recommendations Very little information is available on the intertidal sandy substrate communities in Small Bay and Big

Bay. Should Site 1 be chosen as the preferred site for intake and discharge structures, it is strongly

recommended that a well-designed monitoring plan be developed as part of the RO Plant

environmental requirements. This would involve establishing a baseline of intertidal and shallow

subtidal invertebrate macrofaunal communities before any construction commences, followed by

regular monitoring thereafter to assess recovery of the impacted communities following construction,

as well as responses of the communities to a continuous hypersaline discharge. Similar surveys are

recommended for Site 2 should it be anticipated that construction activities and RO plant discharges

are likely to impact upon the sandy beach fauna.

Independent of which site and design is ultimately chosen, monitoring programmes should be

developed to study the impact of the brine on potentially affected communities, particularly the

subtidal benthic communities. This recommendation is reinforced by the Guidelines for

Environmental Evaluation for Seawater Desalination which has recently been published by the

Department of Water Affairs and Forestry (DWAF 2007), in which it is stated that it is essential that

the effects of the discharge of brine into any water body, on the fish and benthos life in that water

body, be monitored according to a monitoring programme performed at 6-monthly intervals over a

period of approximately 4 years. Depending on initial results, reduced monitoring (i.e. annually) may

be acceptable. This monitoring will include measurement of the main water quality parameters such

as temperature, salinity and dissolved oxygen as a minimum. It is further recommended that every

effort be made to publish the results in a peer-reviewed journal.

The waste brine often contains low amounts of heavy metals, which tend to enrich in suspended

material and finally in sediments. It is recommended that after commissioning of the RO plant that

the effluent be monitored regularly for heavy metals until a profile of the discharge in terms of heavy

metal concentrations is determined. These heavy metal concentrations in the brine effluent would

then need to be assessed based on existing guidelines (DWAF 1995, ANZECC 2000). A summary of

these guidelines is provided in Table 8.1.

Similarly, to ensure complete confidence in the controls of the dosing regime and that the consequent

residual biocides in the discharge are being managed to concentrations that together with possible



synergistic effects other co-discharges will not have significant environmental impacts, it will be

necessary to undertake toxicity testing of the discharge for a full range of operational scenarios (i.e.

shock-dosing, etc). Such sampling and toxicity testing need only be undertaken for the duration and

extent necessary to determine an effluent profile under all operational scenarios.

The mariculture industry is extremely sensitive to tainting or even the perception of tainting. On the

basis of present information the discharges will contain negligible or no tainting substances.

Finally this assessment is based on numerical modelling results. Typical brine and thermal footprints

need to be confirmed by sampling with a conductivity-temperature-depth (CTD) probe after an initial

period of operation of the discharge both to confirm the performance of the discharge system and the

numerical model predictions. This should be done for a suitably representative range of

“conservative” environmental conditions, i.e. conditions for which dispersion of the effluent is likely to

be the most limited. It is envisaged that two to three field surveys of one to two days duration would

be adequate to confirm the performance of the discharge system and the accuracy of model

predictions. It is likely, in any case, that most of these measurements would be needed to be

included in the monitoring programmes developed to study the impact of the brine on potentially

affected communities, particularly the subtidal benthic communities.

All of the above recommendation, except for the monitoring to confirm the model predictions, are

deemed to be essential. However the performance of the outfall according to design will need to be

assessed and the same data as collected for confirming the performance of the discharge system

could be used to assess the modelling results.



10 REFERENCES

Al Malek, S.A. and A.M.O. Mohamed (2005). Environmental impact assessment of off shore oil spill

on desalination plant. Desalination, 185, 9-30.

Anchor Environmental Consultants (2004). Saldanha Bay Water Quality Monitoring Programme:

Benthic Macrofaunal Monitoring - Monitoring Report: 2004. Anchor Environmental

Consultants CC Report, July 2004, 27pp+5pp App.

Anderson, R.J., Levitt, G.J., Keats, D.W. and R. H. Simons (1993). The role of herbivores in the

collapse of the Gracilaria resource at Saldanha Bay, South Africa. Hydrobiologia, 260-261.

Anderson, R.J., Monteiro, P.M.S., and G.J. Levitt (1996). The effect of localised eutrophication on

competition between Ulva lactuca (Ulvaceae, Chlorophyta) and a commercial resource of

Gracilaria verrucosa (Gracilariaceae, Rhodophyta) Hydrobiologia, 326-327, 291-296.

Anderson R.J., Smit A.J. and G. J. Levitt (1999). Upwelling and fish-factory waste as nitrogen

sources for suspended cultivation of Gracilaria gracilis in Saldanha Bay, South Africa.

Hydrobiologia, 398/399, 455-462.

ANZECC (2000). Australian and New Zealand guidelines for fresh and marine water quality. Volume

2, Aquatic ecosystems. National water quality management strategy; no.4. Australian and

New Zealand Environment and Conservation Council, Agriculture and Resource

Management Council of Australia and New Zealand ,Canberra, Australia. ISBN 0 642 19562

5 (www.deh.gov.au/water/quality/nwqms/introduction/).

Atkinson, L., Hutchings, K., Clark, B., Turpie, J., Steffani, N., Robinson, T. and A. Duffell-Canham

(2006). State of the Bay 2006: Saldanha Bay and Langebaan Lagoon. Technical Report.

Prepared for Saldanha Bay Water Quality Trust, 93 pp. + Appendices.

Bamber, R.N. (1990) Power station thermal effluents and marine crustacea. J. Therm. Biol., 15(1), 91-96.

Bamber, R.N. (1995) The influence of rising background temperature on the effects of marine thermal

effluents. J. Therm. Biol., 20(1/2), 105-110.

Bamber, R. N. and R. M. H. Seaby (2004) The effects of power station entrainment passage on three

species of marine planktonic crustacean, Acartia tonsa (Copepoda), Crangon crangon

(Decapoda) and Homarus gammarus (Decapoda). Mar. Environ. Res., 57, 281 –294.

Barnabe, G. (1989) Aquaculture. Volume 1. New York, Ellis Horwood. 528 pp.

Baumgartner, D J, Trent, D S and Byram, K V (1971). User’s guide and documentation for outfall

plume model. USEPA, Pacific Northwest Laboratory, Corvallis, Oregon, Working Paper

No.80.



Bayne, B.L. (1965). Growth and delay of metamorphosis of larvae of Mytilus edulis L. Ophelia, 2,1-

47.

Bolsch, C.J. and G.M. Hallegraeff (1993). Chemical and physical treatment options to kill toxic

dinoflagellate cysts in ship’s ballast water. J. Mar. Env. Eng. 1,23-29.

Bornman, T.G.and N.T.W. Klages (2004). Upgrade of Kenton on Sea / Bushmansrivermouth Bulk

Supply. Appendix H: Specialist Report - Effects of brine discharge from a reverse osmosis

plant on the salinity and temperature of the Bushmans River Estuary, IECM Report No. C84,

Institute for Environmental and Coastal Management, Environmental Consulting and

Research, University of Port Elizabeth, 10pp.

Branch, G.M. and C.L. Griffiths (1988). The Benguela ecosystem part V: the coastal zone. Oceanogr.

and Mar. Biol. Ann. Rev. 26, 395-486.

Branch, G.M., Griffiths, C.L., Branch, M.L., and L.E. Beckley (1994). Two Oceans. David Philip

Publishers, Cape Town, South Africa, 360pp.

Brown, A. C. & A. McLachlan, 1990. Ecology of sandy shores, pp. 1-328 Amsterdam, Elsevier.

Brüchert, V., B. Barker Jǿrgensen, K., Neumann, D., Riechmann, M., Schlösser and H. Schulz

(2003). Regulation of bacterial sulfate reduction and hydrogen sulfide fluxes in the central

Namibian coastal upwelling zone. Geochim. Cosmochim. Acta, 67(23), 4505-4518.

California Coastal Commission (2004). Sea water desalination and the California coastal Act. March

2004. (Source: http://www.coastal.ca.gov.)

California Desalination Task Force (2003a) Issue paper on beach wells. Draft Document.

(http://www.owue.water.ca.gov/recycle/desal/desal.cfm)

California Water Desalination Task Force (2003b). Issue paper on concentrate management

associated with desalination facilities. Draft Document. August 2003.

(http://www.owue.water.ca.gov/recycle/desal/desal.cfm)

Campbell, R.L. and A.T. Jones (2005). Appropriate disposal of effluent from coastal desalination

facilities. Desalination, 182, 365-372.

Carter, R. (2008) Environmental Impact Assessment: Phase 2 Expansion of the Saldanha Iron Ore

Export Terminal Ecological Effects of Increased Ballast Water Discharges, Lwandle Report,

(draft)

Capuzzo, J.M., Davidson, J.A., Lawrence, S.A and M. Libni (1977). The differential effects of free and

combined chlorine on juvenile marine fish. Estuar. Coastal Mar. Sci. 5, 733-741.

Christie, N.D. (1976). A numerical analysis of the distribution of a shallow sublittoral sand macrofauna

along a transect at Lambert’s Bay, South Africa. Trans. Roy. Soc. S. Afr. 42, 149-172.

Christie, N.D., and A.G.S. Moldan (1977). Effects of fish factory effluents on the benthic macrofauna

of Saldanha Bay. Marine Pollution Bulletin, 8, 41-45.



Clark, B.M. (1997). Variation in surf-zone fish community structure across a wave-exposure gradient.

Estuarine, Coastal and Shelf Science, 44, 659-674.

Clark, J.V. (1992). Physiological responses of adult Penaeus semisulcatus (de Haan) to changes in

salinity. Comp. Biochem. Physiol. A. 101(1), 117-119.

Cook, P.A. (1978). A prediction of some possible effects of thermal pollution on marine organisms on

the west coast of South Africa, with particular reference to the rock lobster, Jasus lalandii.

Trans. Roy. Soc. S. Afr. 43(2), 107-118.

Cooper, J. (1995). Specialist study on the impact of underwater blasting on marine invertebrates,

especially African Penguins, in Saldanha Bay. CSIR Report EMAS-C 94095D. 33pp.

CSIR (1976) Oceanographic investigation for waste disposal of greater Saldanha. CSIR Report

C/SEA 7611, 15pp + 48 pp App.

CSIR (1995) The Saldanha Bay temperature regime. CSIR Report EMAS-C 95009, 12pp.

Day, J.H. (1959). The biology of Langebaan Lagoon: a study of the effect of shelter from wave action.

Trans R Soc S Afr., 35, 475-547.

Day, J.H. (1974). A guide to marine life on South African shores, Vol. A. A. Balkema, Cape Town

Day, J.H. (1981) Summaries of current knowledge of 43 estuaries in southern Africa. In:

Estuarine ecology with particular reference to southern Africa. Day, J.H. (Ed.) A.A.

Balkema, Cape Town, pp251-329.

Diaz, R.L. and R. Rosenberg (1995). Marine benthic hypoxia: a review of its ecological effects and

the behavioural responses of benthic macrofauna. Oceanogr. Mar. Biol. Ann. Rev. 33, 245-

303.

Dugan J.E., Hubbard, D.M., McCrary, M.D. and M.O. Pierson (2003). The response of macrofauna

communities and shorebirds to macrophyte wrack subsidies on exposed sandy beaches of

southern California. Estuarine, Coastal and Shelf Science, 58S, 133-148.

Department of Water Affairs And Forestry (DWAF) (1995). South African water quality guidelines for

coastal marine waters. Volume 1. Natural Environment. Volume 2. Recreation. Volume 3.

Industrial use. Volume 4. Mariculture. Pretoria.

Department of Water Affairs And Forestry (DWAF) (2000). Water use authorisation process for

individual applications. Edition 1: Final Draft for implementation and use. Revision 3,

December 2000, Pretoria. ISBN 0-620-26602-3 (http://www-dwaf.pwv.gov.za/Documents/)

Department of Water Affairs And Forestry (DWAF) (2002). National Water Quality Management

Framework. Water Quality Management Sub-series No. MS 7, Draft 2. Pretoria.



Department of Water Affairs And Forestry (DWAF) (2003a). Aide Mémoire for the preparation of a

water quality management report to support the application for licences for sewage treatment

works in terms of the requirements of the National Water Act, 1998 (Act 36 of 1998).

Management Series. Operational Guideline No. U2.1. First Edition, February 2003. Pretoria.

Department of Water Affairs And Forestry (DWAF) (2003b). Towards a strategy for a waste discharge

charge system. Water Quality Management Series. Sub-series No. MS 11, First Edition, 30

September 2003. Pretoria (www-dwaf.pwv.gov.za/Documents/).

Department of Water Affairs And Forestry (DWAF) (2004a). Operational policy for the disposal of

land-derived water containing waste to the marine environment of South Africa, Water

Quality Management Series Sub-Series No. MS 13.2, Edition 1. Pretoria

Department of Water Affairs And Forestry (DWAF) (2004b). Operational Policy for the disposal of

land-derived water containing waste to the marine environment of South Africa: Guidance on

Implementation, Water Quality Management Series Sub-Series No. MS 13.3, Edition 1.

Pretoria

Department of Water Affairs And Forestry (DWAF) (2004c). Operational policy for the disposal of

land-derived water containing waste to the marine environment of South Africa: Appendices.

Water Quality Management Series Sub-Series No. MS 13.4, Edition 1. Pretoria.

Department of Water Affairs and Forestry (DWAF) (2007). Guidelines for the Evaluation of possible

Enviornmental Impatcs during the Development of Seawater Desalination Processes. August

2007, Project No. 337/2006, Pretoria

Diaz, R.L. and R. Rosenberg (1995). Marine benthic hypoxia: a review of its ecological effects and

the behavioural responses of benthic macrofauna. Oceanogr. Mar. Biol. Ann. Rev. 33, 245-

303.

Dugan J.E., Hubbard, D.M., McCrary, M.D. and M.O. Pierson (2003). The response of macrofauna

communities and shorebirds to macrophyte wrack subsidies on exposed sandy beaches of

southern California. Estuarine, Coastal and Shelf Science, 58S, 133-148.

Einav, R., Harussi, K. and D. Perry (2002). The footprint of the desalination processes on the

environment. Desalination, 152, 141-154.

Ellingsen, K.E. (2002). Soft-sediment benthic biodiversity on the continental shelf in relation to

environmental variability. Mar Ecol Prog Ser., 232, 15-27.

Emanuel, B.P., Bustamante, R.H., Branch, G.M., Eekhout, S., and F.J. Odendaal (1992). A

zoogeographic and functional approach to the selection of marine reserves on the west coast

of South Africa. South African Journal of marine Science 12, 341-354.

Engledow, H.R. and J.I. Bolton (1992). Environmental tolerances in culture and agar content of

Gracilaria verrucosa (Hudson) Papenfus (Rhodophyta, Gigartinales) from Saldanha Bay. S.

Afr. J. Bot. 58(4), 263-267.



US EPA (1994) Reregistration Eligibility Decision (RED) - 2,2-dibromo-3-nitrilopropionamide

(DBNPA) – List C (Case 3056), EPA 738-R-94-026, Office of Pesticide PProgrammes,

Washington DC, 179pp.

Eppley, R.W., Renger E.H., and P.M. Williams (1976). Chlorine reactions with seawater constituents

and inhibition of photosynthesis of natural marine phytoplankton. Estuarine and Coastal

Marine Science, 7, 291-301.

European Union Risk Assessment Report (2004). Edetic Acid (EDTA). CAS No: 60-00-4. EINECS

No: 200-449-4. Risk Assessment Vol. 49. Final Report, 2004.

http://ecb.jrc.it/documents/Existing-

Chemicals/RISK_ASSESSMENT/REPORT/edtareport061.pdf

Field, J.G. and C.L. Griffiths (1991). Littoral and sublittoral ecosystems of southern Africa. In: A.C.

Mathieson and P.H. Nienhuis (eds). Ecosystems of the World 24. Intertidal and Littoral

Ecosystems, Elsevier Science Publishers, Amsterdam.

Flach, E. and L. Thomsen (1998). Do physical and chemical factors structure the macrobenthic

community at a continental slope in the NE Atlantic? Hydrobiologia, 375/376, 265-285.

Greenwood, C. and J. Taunton-Clarke (1992) An "atlas" of mean monthly and yearly average sea

surface temperatures around the southern African coast. Sea Fisheries Research Institute

Internal Report, 124, 112pp.

Greenwood, C. and J. Taunton-Clarke (1994) Time series of monthly mean sea temperatures around

the South African coast : an addendum to No. 124. Sea Fisheries Research Institute Internal

Report, 124A, 36pp.

Griffiths, C.L., Hockey, P.A.R., Van Erkom Schurink, C. and P.J. Le Roux (1992). Marine invasive

aliens on South African shores: implications for community structure and trophic functioning.

S Afr J Mar Sci, 12, 713–722.

Grindley, J.R. (1977). The zooplankton of Langebaan lagoon and Saldanha Bay. Transactions of the

Royal Society of South Africa, 42, 341-370.

Gutiérrez-Galindo E.A. (1981). Effect de lÉDTA sur láccumulation et lélimination du Mercure par la

moule Mytilus edulis. Chemosphere, 10, 971-976.

Hanekom, N. and P. Nel (2002). Invasion of sandflats in Langebaan Lagoon, South Africa, by the

alien mussel Mytilus galloprovincialis: size, composition and decline of the population. African

Zoology, 37, 197-208.

Hashim, A. and M. Hajjaj (2005). Impact of desalination plants fluid effluents on the integrity of

seawater, with the Arabian Gulf in perspective. Desalination, 182, 373-393.

HERA (2003). Human & Environmental Risk Assessment on ingredients of European household

cleaning products. Sodium Tripolyphosphate (STPP) CAS: 7758-29-4. Draft June 2003.

(http://www.heraproject.com).



Heugens, E.H.W., Hendriks, A.J, Dekker, T., Van Straalen N.M and W. Admiraal (2001). A review of

the effects of multiple stressors on aquatic organisms and analysis of uncertainty factors for

use in risk assessment. Crit. Rev. Toxicol., 31(3):247-284.

Hoepner, T. and S. Lattemann (2002). Chemical impacts from seawater desalination plants - a case

study of the northern Red Sea. Desalination, 152, 133-140.

Hutchings, K. and S.J. Lamberth (2002). Bycatch in the gillnet and beach-seine fisheries in the

Western Cape, South Africa, with implications for management. South African Journal of

Marine Science, 24, 227-241.

Jackson, L.F. and S. McGibbon (1991). Human activities and factors affecting the distribution of

macrobenthic fauna in Saldanha Bay. S Afr J aquat Sci 17:89-102.

Jaramillo, E., McLachlan A. and J. Dugan (1995). Total sample area and estimates of species

richness in exposed sandy beaches. Mar. Ecol. Prog. Ser., 119, 311-314.

Jirka, G.H., R.L. Donecker and S.W. Hinton (1996) User’s Manual for Cormix: A hydrodynamic mixing

model and decision support system for pollutant discharges into surface waters. DeFrees

Hydraulics Laboratory, School of Civil and Environmental Engineering, Cornell University,

Ithaca, New York.

Jury, M.R. and L. Guastella ( 1987) Coastal wind and dispersion conditions at Koeberg: analysis of

data with application to mesoscale forecasting, S. Afr. J. Sci., 83, 435-440.

King, M.G. (1977). Cultivation of the Pacific oyster, Crassostrea gigas, in a non-tidal hypersaline

pond. Aquaculture, 11, 123-136.

Klaine, S. J., Cobb, G. P., Dickerson, R. L., Dixon, K. R., Kendall, R. J., Smith, E. E., and Solomon,

K. (1996). An ecological risk assessment for the use of the biocide,

Dibromonitrilopropionamide (DBNPA), in industrial cooling systems. Environmental

Toxicology and Chemistry 15, 21-30.

Kruger, N., Branch, G.M., Griffiths, C.L. and J.G. Field (2005). Changes in the epibenthos of

Saldanha Bay, South Africa, between the 1960s and 2001: an analysis based on dredge

samples. African Journal of Marine Science, 27, 471-477.

Krug, M. (1999) Circulation through the mouth of the Langebaan Lagoon and implications, MSc.

Thesis, University of Cape Town, 133pp.

Kungvankij, P. et al. (1986a). Shrimp hatchery design, operation and management. NACA Training

Manual Series No 1. Network of Aquaculture Centres in Asia, Regional Lead Centre in the

Philippines. 88 pp.

Kunvankij, P. et al. (1986b.) Shrimp culture: pond design, operation and management. NACA

Training Manual Series No 2. Network of Aquaculture Centres in Asia, Regional Lead Centre

in the Philippines. 68pp.



Latorre, M. (2005). Environmental impact of brine disposal on Posidonia seagrasses. Desalination

182:517-524.

Lellis, W.A. and J.A. Russell (1990). Effect of temperature on survival, growth and feed intake of

postlarval spiny lobster. Aquaculture, 90(1), 1-9.

Lesser, G.P., J.A. Roelvink, J.A.T.M. van Kesterand G.S. Stelling (2004). Development and

validation of a three-dimensional morphodynamic model. Coastal Engineering, 51, 883-915.

Luger, S., R.C. van Ballegooyen and R. Carter (1997). Possible use of seawater cooling by Saldanha

Steel: Three dimensional modelling of the thermal plume and assessment of ecological

impacts. CSIR Report ENV/S-C97058, 66p + figures.

Luger, S., Monteiro, P.M.S., van Ballegooyen, R.C., Schoonees, J.S. and J. Moes (1999). Medium

term expansion of the Port of Saldanha into Small Bay: A modelling-based predictive study of

hydrodynamic, water quality, sediment transport and wave resonance considerations of four

alternative port layouts. Volume 1. Text, Volume 2. CSIR Report. ENV-S-C 99050 A, B.

Stellenbosch

Luksiene, D., Sandstrom, O., Lounasheimo, L. and J. Andersson (2000). The effects of thermal

effluent exposure on the gametogenesis of female fish. J. Fish Biology, 56(1), 37-50.

MCM (Marine and Coastal Management). (2007). Recommendations for Management towards

Sustainable Beach-Seine and Gillnet (Smallnet / Driftnet) Fisheries in 2006/2007. August

2007.

MCM (Marine and Coastal Management). (2006). Commercial gillnet (drift/set net) permit conditions:

July 2006 - 31 December 2006. Section B: Permit conditions: Limited commercial smallnets:

Gillnet/Drift/Set Fishery: Area 12 (Langebaan - West Coast National Park).

Mabrook, B. (1994). Environmental impact of waste brine disposal of desalination plants, Red Sea,

Egypt. Desalination, 97, 453-465.

MacPherson, E. and A. Gordoa (1992). Trends in the demersal fish community off Namibia from

1983 to 1990. S. Afr. J. mar. Sci., 12, 635-649.

Martinez-Arroyo, A., Abundes, S., Gonzalez, M.E. and I. Rosas (2000). On the influence of hot-water

discharges on phytoplankton communities from a coastal zone of the Gulf of Mexico. Water,

Air, and Soil Pollution, 119(1-4), 209-230.

Maurer, D., Keck, R.T., Tinsmann, J.C. and W.A. Leathem (1981a). Vertical migration and mortality

of benthos in dredged material: Part I - Mollusca. Marine Environmental Research, 4, 299-

319.

Maurer, D., Keck, R.T., Tinsmann, J.C. and W.A. Leathem (1981b). Vertical migration and mortality

of benthos in dredged material: Part II - Crustacea. Marine Environmental Research, 5, 301-

317.



Maurer, D., Keck, R.T., Tinsmann, J.C. and W.A. Leathem (1982). Vertical migration and mortality of

benthos in dredged material: Part III - Polychaeta. Marine Environmental Research 6:49-68.

McClurg, T.P. (1974). Studies on some environmental requirements of Penaeus indicus M. EDW.

Twenty-ninth Steering Committee Meeting on Marine Disposal of Effluents. CSIR, NIWR,

79pp.

McLachlan, A., Jaramillo, E.., Donn T.E.. and F. Wessels (1993). Sandy beach macrofauna

communities and their control by the physical environment: a geographical comparison. J.

coastal Res. Spec. Issue 15, 27-38.

McLusky, D. S. (1981). The estuarine ecosystem. Blackie & Son, Glasgow, 150pp.

McQuaid, C. D. and G. M. Branch (1984). Influence of sea temperature, substratum and wave

exposure on rocky intertidal communities: an analysis of faunal and floral biomass. Marine

Ecology Progress Series. 19, 145-151.

McQuaid, C. D., Branch, G. M. and A. A. Crowe (1985). Biotic and abiotic influences on rocky

intertidal biomass and richness in the southern Benguela region. South African Journal of

Zoology, 20, 115-122.

Miri, R. and A. Chouikhi (2005). Ecotoxicological marine impacts from seawater desalination plants.

Desalination, 182, 403-410.

Moldan, A.G.S. (1978). A study of the effects of dredging on the benthic macrofauna in Saldanha

Bay. South African Journal of Science ,74, 106-108.

Montiero, P.M.S. and G.B. Brundrit (1990) Interannual chlorophyll variability in South Africa’s

Saldanha Bay system, 1974-1979. S. Afr. J. mar. Sci., 9, 281-287.

Monteiro, P.M.S. and J.L. Largier (1999). Thermal stratification in Saldanha Bay (South Africa) and

subtidal, density-driven exchange with the coastal waters of the Benguela Upwelling System.

Estuarine, Coastal and Shelf Science, 49, 877-890.

Monteiro, P.M.S. and A. Kemp (2004).Water quality management plan for Saldanha Bay: Phase III:

Setting of critical loads using a modelling based ecosystem approach. CSIR Report ENV-S-

C 2004-083, 30pp. Stellenbosch.

Monteiro, P.M.S., Anderson, R.J. and S. Woodbourne (1997). δ15N as a tool to demonstrate the

contribution of fish waste-derived nitrogen to an Ulva bloom in Saldanha Bay, South Africa. S

Afr.J. mar. Sci. 18, 1-9.

Nel, P., Pulfrich, A. and A.J. Penney (2003). Impacts of beach mining operations on sandy beach

macrofaunal communities on the beaches of Geelwal Karoo, Report by Pisces Environmental

Services (Pty) Ltd prepared for Trans Hex Operations (Pty) Ltd, Cape Town.



Nel, P., Pulfrich, A. and A.J. Penney (2004). Impacts of beach mining operations on sandy beach

macrofaunal communities on the beaches of Geelwal Karoo. 2004 Beach Survey, Report by

Pisces Environmental Services (Pty) Ltd prepared for Trans Hex Operations (Pty) Ltd, Cape

Town.

Newell, R.C., Seiderer, L.J. and D.R. Hitchcock (1998). The impact of dredging works in coastal

waters: a review of the sensitivity to disturbance and subsequent recovery of biological

resources on the seabed. Oceanography and Marine Biology: an Annual Review, 36, 127-

178.

OECD (Organisation for Economic Co-operation and Development) (1997). Screening Information

Data Set: (SIDS) Initial Assessment Profile: Sodium Dodecyl Sulfate.

http://www.chem.unep.ch/irptc/sids/OECDSIDS/151213.htm

Parsons, T.R., Takahashi, M. and B. Hargrave (1977). Biological Oceanographic Processes.

Pergamon Press, New York: 332pp.

PDNA/SRK Joint Venture, (2007). Environmental Impact Assessment (Basic Assessment) Process:

Proposed reverse osmosis plant, iron-ore handling facility, Port of Saldanha – Background

information Document, SRK Project 364470/42G, 4pp.

PDNA/SRK Joint Venture, (2008). Basic Environmental Assessment (DEAT Application No

12/12/20/958) for The proposed Reverse Osmosis Desalination Plant at the Transnet Iron

Ore Handling Facility, Saldanha Bay, Western Cape, (Transnet Contract No. H500107-

CPS006).

Pitcher, G.C. and D. Calder (1998). Shellfish mariculture in the Benguela system: Phytoplankton and

the availability of food for commercial mussel farms in Saldanha Bay, South Africa. J.

Shellfish Res., 17, 15-24.

Pitcher, G.C. and D. Calder (2000). Harmful algal blooms of the southern Benguela Current: a review

and appraisal of monitoring from 1989 to 1997. S. Afr. J. Mar. Sci., 22, 255-284.

Pitcher, G.C., Calder, D., Davis G. and G. Nelson (1994). Harmful algal blooms and mussel farming

in Saldanha Bay. Proc. Aquacult. Assoc. sth. Afr. 5, 87-93.

Probyn, T, Pitcher, G, Pienaar, R. and R. Nuzzi (2001). Brown tides and mariculture in Saldanha Bay,

South Africa. Marine Pollution Bulletin, 42, 405-408.

Puttick, G.M. (1980). Energy budgets of Curlew Sandpipers at Langebaan Lagoon, South Africa.

Estuarine coastal marine Science. 11:207-215.

Rainbow, P.S., and W.H. Black (2002). Effects of changes in salinity and osmolality on the rate of

uptake of zinc by three crabs of different ecologies. Marine Ecology Progress Series, 244,

205-217.



Raventos, N., Macpherson, E. and A. Garcıa-Rubies (2006). Effect of brine discharge from a

desalination plant on macrobenthic communities in the NW Mediterranean. Marine

Environmental Research, 62, 1-14.

Roast, S.D., Rainbow, P.S., Smith, B.D., Nimmo, M. and M.B. Jones (2002). Trace metal uptake by

the Chinese mitten crab Eriocheir sinensis: the role of osmoregulation. Marine Environmental

Research, 53, 453-464.

Robinson, T.B. and C.L.. Griffiths (2002). Invasion of Langebaan Lagoon, South Africa, by Mytilus

galloprovincialis-effects on natural communities. African Zoology, 37, 151-158.

Robinson, T.B., Branch, G.M., Griffiths, C.L., Govender, A. and P.A.R. Hockey (2007b). Changes in

South African rocky intertidal invertebrate community structure associated with the invasion

of the mussel Mytilus galloprovincialis. Mar Ecol Prog Ser., 340, 163-171.

Robinson, T.B., Griffiths, C.L., Branch, G.M. and A. Govender (2007a). The invasion and subsequent

die-off of Mytilus galloprovincialis in Langebaan Lagoon, South Africa: effects on natural

communities. Marine Biology in press.

Rosenthal, G. and S. Grant (1989) Simplified tidal prediction for the South African coastline. S. A. J.

Sci., 85, 104-107.

Sandstrom, O., I. Abrahamsson, J. Andersson and M. Vetemaa (1997) Temperature effects on

spawning and egg development in Eurasian perch. J. of Fish Biology; 51(5), 1015-1024.

Schils, T., Clerck, O.D., Leliaert, F., Bolton, J.J. and E. Coppejans (2001). The change in macroalgal

assemblages through the Saldanha Bay/Langebaan Lagoon ecosystem (South Africa).

Botanica Marina 44:295-305.

Schüttrumpf, H., A. Kortenhaus, K. Peters and P. Fröhle (2006). Expert judgement of uncertainties in

coastal structure design. Coastal Engineering, 5, 5267-5278.

Seckel, G. R., Beaudry, F.H., 1973. The radiation from sun and sky over the north Pacific Ocean,

Trans. Amer, Geophys. Union, 54, 1114.

Shannon, L.V. and G.M. Stander (1977). Physical and chemical characteristics of water in Saldanha

Bay and Langebaan Lagoon. Transaction of the royal Society of South Africa 42(3 & 4):411-

461.

Shannon, L.V., R.J.M. Crawford, D.E. Pollock, L. Hutchings, A.J. Boyd, J. Taunton-Clarke, A.

Badenhorst, R. Melville-Smith, C.J. Augustyn, K.L. Cochrane, I. Hampton, G. Nelson, D.W.

Japp and R.J. Tarr (1992) The 1980s - a decade of change in the Benguela System. S. Afr.

J. mar. Sci., 12, 271-296.

Siebert, T. and G.M. Branch (2007). Influences of biological interactions on community structure

within seagrass beds and sandprawn-dominated sandflats Journal of Experimental Marine

Biology and Ecology, 340, 11-24.

Simons, R.H. (1977). The algal flora of Saldanha Bay. Trans. Roy. Soc. S. Afr. 42, 461-482.



Smith, G.G., G. Diedericks, C. Soltau and M. Rossouw (2007). Environmental Impact Assessment:

Phase 2 Expansion of the Saldanha Iron Ore Export Terminal – Shoreline Stability Specialist

Study, CSIR REPORT: CSIR/NRE/ECO/ER/2007/0132C, 91pp + 15 pp App. (Draft)

Snelgrove, P.V.R. and C.A. Butman (1994). Animal-sediment relationships revisited: cause versus

effect. Oceanogr Mar Biol Annu Rev., 32, 111-177.

Soares, A.G. (2003). Sandy beach morphodynamics and macrobenthic communities in temperate,

subtropical and tropical regions - a macroecological approach. PhD Thesis, University of Port

Elizabeth. 72 pp.

Sorvari, J. and M. Sillanpää (1996). Influence of metal complex formation on heavy metal and free

EDTA and DTPA acute toxicity determined by Daphnia magna. Chemoshere 33(6), 119-127.

Sotero-Santos, B., Rocha, O. and J. Povinelli (2007). Toxicity of ferric chloride sludge to aquatic

organisms. Chemosphere, 68(4), 628-636.

Soulsby, R.L., L. Hamm, G. Klopman, D. Myrhaug, R.R. Simons. and G.P. Thomas (1993). Wave-

current interaction within and outside the bottom boundary layer. Coastal Engineering, 21,

41-69.

Spolander, B. (1996) Entrainment in Saldanha Bay. MSc thesis, Oceanography Department,

University of Cape Town, 67pp.

Stegenga, H., Bolton, J.J. & Anderson, R.J. (1997). Seaweeds of the South African west coast. pp.

[2] + 1-655. Cape Town: Bolus Herbarium, University of Cape Town.

Stenton-Dozey, J.M.E. and A.C. Brown (1994). Exposure of the sandy-beach bivalve Donax serra

Röding to a heated and chlorinated effluent II. Effects of chlorine on burrowing and survival.

Journal of Shellfish Research, 13, 451-454.

Stenton-Dozey, J.M.E., Jackson, L.F. and A.J. Busby (1999). Impact of mussel culture on

macrobenthic community structure in Saldanha Bay, South Africa. Marine Pollution Bulletin

39, 357-366.

Stenton-Dozey, J., Probyn, T. and A. Busby (2001). Impact of mussel (Mytilus galloprovincialis)

raftculture on benthic macrofauna, in situ oxygen uptake, and nutrient fluxes in Saldanha

Bay, South Africa. Can J Fish Aquat Sci., 58, 1021-1031.

Sutherland, J., D.J.R. Walstra, T.I. Chester, L.C. van Rijn and H.N. Southgate (2004). Evaluation of

coastal area modelling systems at an estuary mouth. Coastal Engineering, 51, 119-142.

Swartz, C.D., J.A. du Plessis, A.J. Burger and D. Offringa (2004) A desalination guide for South

African municipal engineers, Water SA Vol. 32 No. 5 (Special edn. WISA 2006), 641-647.

Taljaard, S. (2006). The development of a common set of water and sediment quality guidelines for

the coastal zone of the BCLME (Project BEHP/LBMP/03/04) Report submitted to UNOPS as

part of the Benguela Current Large Marine Ecosystem Programme. CSIR Report

CSIR/NRE/ECO/ER/2006/0011/C. Stellenbosch, South Africa (www.bclme.org).



Taljaard, S. and P.M.S. Monteiro (2002). Saldanha Bay marine water quality management plan.

Phase I: Situation Assessment. Report to the Saldanha Bay Water Quality Forum Trust.

CSIR Report ENV-S-C 2002-115/1. Stellenbosch.

Taljaard, S., Monteiro, P.M.S. and W.A.M. Botes (2006). A structured ecosystem-scale approach to

marine water quality management. Water SA, 32, 535-542.

Taunton-Clarke, J. (1990) Environmental events within the South-East Atlantic (1906-1985) identified

by analysis of sea surface temperature and wind data. S. Afr. J. Sci., 86, (7-10), 470-472.

Taunton-Clarke, J. and L.V. Shannon (1988) Annual and interannual variability in the South-East

Atlantic during the 20th century. S. Afr. J. mar. Sci., 6, 97-106.

Thiyagarajan, V., Harder, T. and P.-Y. Qian (2003). Combined effects of temperature and salinity on

larval development and attachment of the subtidal barnacle Balanus trigonus Darwin. J. Exp.

Mar. Biol. Ecol., 287:223-236.

UNITED STATES ENVIRONMENTAL PROTECTION AGENCY (US-EPA) (1985). Initial mixing

characteristics of municipal ocean discharges. Environmental Research Laboratory,

Naragansett, Report No EPA/SW/MT-86/012(a & b).

Van Ballegooyen, R.C. and S.A. Luger (1999). Assessment of a potential thermal discharge in a

coastal embayment using a three dimensional hydrodynamic model. In: Fifth International

Conference on Coastal and Port Engineering in Developing Countries: Proceedings of the

COPEDEC V, Cape Town, April 1999, (ed. G.P. Mocke), Volume 3: 2018-2027.

Van Ballegooyen, R.C., Steffani, C.N. and P.D. Morant (2004). Kudu Power Plant Environmental

Impact Assessment: Assessment of a proposed cooling water discharge into the marine

environment. CSIR Report ENV-S-C 2004-086, 62pp+ 22 pp. Appendix.

Van Ballegooyen, R.C., Pulfrich, A., Penney, A. and N. Steffani (2005). Kudu Power Plant

Environmental Impact Assessment: Assessment of a proposed cooling water discharge into

the marine environment at a location near Uubvley. CSIR Report ENV-S-C 2005-054, 141pp.

Van der Lingen, C.D. (1994). Aspects of the early life history of Galjoen Dichistius capensis. S. Afr. J.

mar. Sci., 14, 37-45.

Van Erkom Schurink, C. and C.L. Griffiths (1993). Factors affecting relative rates of growth in four

South African mussel species. Aquaculture, 109, 257-273.

Van Tonder, A (1994). The development and testing of a system for the measurement of directional

wave spectra in the nearshore coastal zone. Phd dissertation, University of Stellenbosch.

South Africa, December 1994.

Villacastin-Herroro, C.A., Crawford, R.J.M, and J.G. Field (1992). Statistical correlations between

purse-seine catches of chub-mackerel off South Africa and select environmental parameters.

S. Afr. J. mar. Sci., 12, 157-165.



Visser, D, M. Goes and P. Rosewarne (2007) Environmental Impacts Assessment: Port of Saldanha

proposed reverse osmosis water desalination plant – Groundwater resources impact

assessment, SRK Report, Draft Report Version 2, 106pp.

Vries, M.B., Delvigne, G.A.L. and R.A.H Thabet (1997). Relocation of desalination plant ’s outfall in

the U.A.E in order to minimise environmental damage. Madrid: I.D.A.World Congress on

Desalination and Water Reuse.

Weeks, S.J., A.J. Boyd and P.M.S. Monteiro (1990) Water budgets in the Saldanha Bay and

Langebaan Lagoon systems under differing hydrological conditions - implications for the

residence times of pollutants. Marine Eutrophication Project: Report 1: The currents and

circulation in Saldanha Bay: Post 1975., Oceanography Department, University of Cape

Town, 66pp.

Weeks, S.J., A.J. Boyd, P.M.S. Monteiro and G.B. Brundrit (1991a) The currents and circulation in

Saldanha Bay after 1975 deduced from historical measurements of drogues S. Afr. J. mar.

Sci., 11, 525-535.

Weeks, S.J., A.J. Boyd, G. Nelson and R.M. Cooper (1991b) A note on the wind-driven replacement

flow of the bottom layer in Saldanha Bay, South Africa: Implications for pollution. S. Afr. J.

mar. Sci. 11, 579-583.

Wetland Consulting Services (2007). Dust suppression: An assessment of water availability with

notes on costs of supply, Wetland Consulting Services Pty (Ltd) Report 244/2006, 14pp +

24 pp App.

White, G.C. (1986). Handbook of chlorination. 2nd edition, New York: Van Nostrand Reinhold

Company Inc.

WL | Delft Hydraulics (2005). Delft3D-FLOW User Manual Version 3.12. WL|Delft Hydraulics, Delft,

The Netherlands.

WORLD BANK (1998). Pollution Prevention and Abatement Handbook. Thermal Power: Guidelines

for New Plants. WORLD BANK GROUP, p413-426.

Wu, R.S.S. (2002). Hypoxia: from molecular responses to ecosystem responses. Marine Pollution

Bulletin, 45, 35-45.

Yonkos, L.T., Fisher, D.J., Burton, D.T., Whitekettle, W.K. and J.C. Petrille (2001). Effectiveness of

the sulphur(IV) compound, sodium bisulfite, in reducing chlorine, chlorine dioxide, and

chlorite toxicity to Daphnia magna in well water and pond water. Environmental Toxicology

and Chemistry, 20(3), 530-536.

Zoutendyk, P. (1989). Oxygen consumption by the Cape rock lobster, Jasus lalandii. S.Afr. J. mar.

Sci., 8, 219-230.





EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay

Appendix A:

Possible implications of policy, legislation and approval and licensing

procedures for the proposed RO Plant water discharge





EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay Page A.1

The implications of the policy, legislation and approval and licensing procedures described in the

above policy, legislation and associated documents for the proposed RO Plant water discharge into

Saldanha Bay, can be summarised as follows:

• In addition to the requirements of the National Water Act of 1998 and the policies and approval

and licensing procedures of DWAF, the Water Quality Management Plan (and associated

scientific studies) developed for the Saldanha Bay Water Quality Trust, together with

representations from the DEAT, will play a significant role in it a decision on whether or not the

proposed purge water discharge will be acceptable.

• The DWAF water quality management policy comprises a hierarchy of decision making which

contains elements of the Receiving Water Quality Objectives approach, as well as the

precautionary principle of environmental protection. This require first that there is the prevention

of waste production and pollution wherever possible (Pollution prevention principle), secondly

that there is the minimisation of pollution and waste at source (Pollution minimisation principle).

Only when these first two principles have been satisfied to the greatest extent possible, is

responsible disposal considered and then by applying the precautionary approach.

• An integrated assessment approach will need to be followed that is based on the principles of

Integrated Environmental Management, taking cognisance of concepts such as Strategic

Environmental Assessment, and Environmental Impact Assessment (EIA) and assessment of

the project from “cradle to grave” based on principles of strategic adaptive management, best

practice, consistent performance, flexibility in approach and continuous improvement.

• Alternative options of managing wastewater must be investigated as disposal to the marine

environment is not considered the ‘default’ option in coastal areas. (This is also a requirement

of the Basic Assessment and/or EIA process where the no-go option has to be explicitly

considered).

• Transparent and adequate consultation with all potential stakeholders and interested and

affected parties both prior to and after gaining approval for the disposal of effluents in Saldanha

Bay (i.e. from application through to report back on monitoring results). There will need to be

agreement upon the assumed environmental objectives for the Saldanha Bay and its surrounds

by all relevant stakeholders. In addition, after implementation of the RO Plant discharge, there

will need to be adequate and relevant ongoing monitoring to the extent appropriate. The results

of any such monitoring program will need to be communication to all stakeholders on an

ongoing basis.

• The approval and licensing procedure and the Basic Assessment/EIA process are separate

processes, each with their own specific requirements. Note that an Basic Assessment/EIA

authorisation cannot replace a water use licence application, since the former does not

necessarily address all the requirements of the National Water Act



• It is a legal obligation for the project proponent to provide a full description of the proposed

method of effluent disposal proposed and a full characterisation of the effluent(s) in terms of:

− Flow rates (average, maximum, minimum and diurnal/seasonal variations) for

present and future scenarios

− All relevant constituents (average, maximum, minimum and diurnal/seasonal

variations) for present and future scenarios.

The specification should include all future, planned expansions in order to avoid going through

the whole process of applying for a new licence and/or the requirements for a new EIA at the

time of the future expansion.

• Discharge of land-derived wastewater to any area declared a Marine Protected Area under the

Marine Living Resources Act 18 of 1998 is prohibited. The discharge will have to be located

appropriately such that there is no discharge into such an area and similarly that there are no

adverse environmental impacts in such areas.

• Site-specific environmental quality objectives for the marine environment must take into account

the South African Water Quality Guidelines for coastal marine waters (RSA DWAF, 1995a) or

any future updates thereof.

• Exemption from compliance to wastewater standards and/or/guidelines will only be considered

in exceptional circumstances provided that the receiving water body remains fit for use in

accordance with the Receiving Water Quality Objective approach.

• A licence application will require that the potential impacts on the receiving environment are

adequately investigated in both the near and far field, taking into account other anthropogenic

activities and waste inputs so as to address possible synergistic and/or cumulative effects. A

precautionary approach must be followed in the assessment and design of any marine disposal

system in which the temporal and spatial coverage and accuracy of physical and chemical

oceanographic data do not adequately describe site-specific conditions or where the potential

impacts associated with the discharge are uncertain. Where it cannot adequately be

demonstrated that the potential environmental impacts associated with the discharge of limited

or no significance, the precautionary approach requires that one does not proceed with the

implementation of the proposed discharge system until such time that sufficient information

exists that indicates that all potential environmental impacts are found to be acceptable.

• The details on Management Systems and Pollution Prevention Methods will be critically

assessed, whereby the applicant’s ability to effectively manage the proposed wastewater

disposal facility, will be critically assessed in the licence application procedure.

• The Polluter Pays principle requires that the project proponent and its partners pay for all

environmental costs incurred for rehabilitation of environmental damage, and the costs of

preventive measures to reduce or prevent such damage, necessary studies and on-going

monitoring programs associated with the proposed discharge, should they be required. In

addition there is a possibility that in future there is likely to be a waste discharge charge



associated with all significant discharges to the marine environment. This proposed waste

discharge charge system was planned to be implemented from 2006 onwards (RSA DWAF,

2003b), however there exists no clarity on the progress to date in implementing the proposed

waste discharge charge system. This needs to be factored in when assessing the economic

viability of a project.

• Any authority or industry responsible for the operation and management of a marine disposal

system will be subject to the implementation of an appropriate monitoring programme.

• Any authority or industry responsible for the operation and management of a marine disposal

system will be required to provide the DWAF with a regular evaluation of the performance of the

marine disposal system. Where performance evaluations indicate non-compliance with the

predetermined specifications (including the environmental quality objectives), the responsible

authority or industry will be required to propose mitigating actions to ensure compliance (e.g.

rehabilitation or alternative treatment options). The responsible authority and the industry

operating the wastewater disposal system will be required to implement such actions at their

own cost upon approval of the DWAF.

• Discharge licences are subject to review every 5 years. For a discharge such as the one

proposed there is little that is possible in terms of continual improvement of the discharge,

consequently a high degree of certainty is required around the magnitude of the potential

impacts associated with the discharge to avoid the licensing problems after the review process.

In addition, Saldanha Bay being a highly sensitive, multi-user region (e.g. mariculture activities,

close proximity to marine protected areas, etc) in terms of marine quality, will require that there

is a high degree of certainty around the magnitude of potential environmental impacts of the

proposed discharge before approval is given to proceed with the proposed discharge.

• Where a land-based activity, which requires a licence under section 21 of the NWA, falls within

a commercial harbour area, the National Ports Authority, as the landowner, is responsible to

ensure that such an activity meets the requirements of the relevant laws (DEAT, 2007).

While the above principles/requirements are focussed on larger discharge systems, many are likely to

be relevant to the proposed RO Plant discharge.






Appendix B:

Hydrodynamic and Water Quality Model Set-up and Calibration





EIA for the proposed Reverse Osmosis Desalination Plant in Saldanha Bay Page B.1

B.1 Introduction

The currents in Saldanha Bay are forced by wind, waves, large-scale currents, tides, Coriolis effects

(due to the rotation of the earth), inertial effects and by baroclinic effects induced by changes in water

temperature. To be able to simulate the transport and fate of a dense RO Plant brine discharge within

Saldanha Bay requires that the three-dimensional processes typical of the hydrodynamics of

Saldanha Bay as well as the three-dimensional behaviour of the discharge plume be accurately

simulated. To meet this requirement the Delft3D computer modelling suite has been set up for

Saldanha Bay and applied to simulate the waves, and three-dimensional currents with derived

turbulence quantities.

For the hydrodynamic simulations two modules are used, namely Delft3D-WAVE and Delft3D-FLOW.

These modules are used in the so-called “online” mode which implies that the two modules operate

interactively. The Delft3D-WAVE module uses a time series of waves and calculates a wave field

every 2 hours and it uses the exact water levels from the Delft3D-FLOW module during these

calculations. Every two hours Delft3D-WAVE passes the relevant wave information to Delft3D-FLOW.

This information, together with wind and atmospheric data are used in DELFT3D-FLOW to simulate

the hydrodynamics of the bay that include processes such as upwelling, wave- and wind-driven flows

and turbulent mixing within the water column.

Prior to this RO Plant modelling study, the model was set-up and calibrated to investigate potential

environmental impacts associated with the proposed Phase 2 expansion of the iron-ore export

facilities. The locations of the instrumentation providing the calibration data are given in Figure B.1.

Two modelling simulation periods were considered:

• The first modelling period extends from 20 October 2006 to 15 December 2006. This period

coincides with a measurement campaign to ensure accurate calibration of the model, the focus

of these calibrations being on the wave and wind-driven flows in Big Bay.

• The second modelling period extends from 1 July 1999 to 1 July 2000. For this period, the

objective of the modelling was to simulate the currents that may occur for the situation with and

without the harbour alterations. During this period measurements were made of the water

column stratification in Small Bay that also have been used to calibrate the model. These

calibrations focussed on ensuring accurate simulation of the upwelling dynamics over the

adjacent shelf and their influence on flows and water column mixing processes within the bay,

the main focus being on Small Bay. These dynamics are important for the accurate simulation

of dense discharges i.e. the RO Plant discharges.

In the section that follows, the set-up of the model and its calibration for the Phase 2 expansion of the

iron-ore export facilities study, are described in detail. However, it has been necessary to modify the

computational grid for DELFT3D-FLOW used in the Phase 2 studies (Figure B.2a and b) for use in this study in order to resolve the plume dynamics in the immediate vicinity of the proposed RO Plant

discharge locations.



Waverider

Figure B.1: Position of the instrumentation providing the data used to calibrate the model.



Two modifications of the computational grid have been made. In the first modified grid (Figure B.3),

the spatial resolution of the grid was dramatically increased in the vicinity of the initial proposed discharge sites (Site 1, Site 2 and discharges into Small and Big Bay from Site 3). When a further

discharge site was identified, that was expected to have the least impact on the marine environment

of all options considered, it was necessary again to modify the computational grid (Figure B.4) to

ensure adequate resolution of the plume dynamics at the new site (i.e. discharge at Caisson 3 from

Site 3).

Such increases in the spatial resolution of the computational grid are expected to result in more

accurate simulation of the hydrodynamics at these locations where the spatial resolution has been

increased. Given that these changes occurred only in the computational grid of the DELFT3D-FLOW hydrodynamics model and furthermore are localised, the original calibration exercise

undertaken for the Phase 2 expansion of the iron-ore export facilities study is deemed also to be

adequate for this study (i.e. the modelling of discharges from the RO Plant). Consequently this

calibration exercise is reported in some detail here.

B.2 Model Set-ups

B.2.1 Computational grids Computational grids were set-up for both the wave modelling and the three dimensional

hydrodynamic modelling. The computational grid used in the wave modelling (Figure B.5) was more

extensive than that used in the hydrodynamic model (Figures B.3 and B.4) and extended offshore

into water depths exceeding 200 m, both to ensure that i) wave conditions were available over the full

extent of the hydrodynamic model and ii) that the wave conditions at the offshore boundary are

applied in water depths where refraction has not yet occurred. The lateral boundaries are also

located far from the region of interest to prevent inaccuracies in boundary conditions affecting the calculations in the area of interest. Where the grid used by Delft3D-WAVE overlaps that of Delft3D-

FLOW, the computational grids from the two modules are exactly the same, i.e. there is no loss of

information due to interpolation between the different modules because the Delft3D-FLOW grid

coincides with the Delft3D-WAVE grid.

The hydrodynamic grids have 112 x 124 cells in the horizontal and are designed so that the grid size

is as large as 3.5 km x 4.0 km in the offshore region, but is refined to a size of approximately

20 m x 25 m in the immediate vicinity of the proposed discharges into Saldanha Bay. In the vertical,

10 sigma layers were used with the thickness of the layers from surface to seabed being 10 %, 15 %,

15 %, 15 %, 10 %, 10 %, 10 %, 8 %, 5% and 2 % of the local water depth. The thinnest layers are located near the seabed to resolve the dense brine plume behaviour near the seabed as it will either

be discharged at these depths or rapidly sink to these depths in the immediate vicinity of the

discharge. The XY-coordinate system used in the simulations is based on the Universal Transverse

Mercator coordinate system (UTM 33 South) and the WGS84 spheroid.



Figure B.2a: The computational grid used in the hydrodynamic simulations for which a detailed calibration has been undertaken (i.e. Phase 2 studies for the iron-ore expansion project – Smith et al., 2007).



Figure B.2b: A zoomed in view of the computational grid used in the hydrodynamic simulations for which a detailed calibration has been undertaken (i.e. Phase 2 studies for the iron-ore expansion project – Smith et al., 2007).



Figure B.3: The computational grid used for the initial RO Plant discharge simulations.



Figure B.4: The computational grid modified to increase the spatial resolution in the vicinity of the Caisson 3 discharge location.



Figure B.5: The computational grid used in the wave simulations.



B.2.2 Bathymetry

The model bathymetry was obtained from the following data sources:

• S A Navy hydrographic chart of Langebaan Lagoon (SAN 2052);

• S A Navy hydrographic chart of the Entrance to Saldanha Bay (SAN 1011);

• S A Navy hydrographic chart of the Saldanha Bay Harbour (SAN 1012).

Higher resolution surveys were added in the following regions (see Figure B.6):

• From Lynch Point to the Reclamation Dam and offshore area to the bay entrance – Nov/Dec

2006 CSIR survey;

• Lynch Point to Langebaan Point – Surveys conducted in 2003 and 2005 by CSIR for PRDW;

• Region south-west of the Iron Ore Jetty, extending from Reclamation Dam to the seaward end

of the jetty – surveyed in late 2005 survey by Marine GeoSolutions.

The detailed bathymetry in the region of interest is given in Figure B.1.

Figure B.6: The higher resolution bathymetry used in additional to the South African Navy chart bathymetry data.



B.2.3 Wave Simulations

The wave refraction predictions were obtained by using the numerical model SWAN (Simulating

Waves Nearshore). The wave refraction model SWAN was executed within the Delft3D-WAVE

environment which provides a convenient interface for pre- and post-processing of the results. The

Delft3D-WAVE (SWAN) model is based on the discrete spectral action balance equation and is fully

spectral in all directions and frequency, implying that short-crested random wave fields propagating

simultaneously from widely different sources can be accommodated, e.g. a swell with superimposed

wind sea.

The Delft3D-WAVE model can account for refractive propagation due to currents and depth. The

model also represents explicitly the processes of wave generation by wind, dissipation by white-

capping, bottom friction and depth-limited wave breaking and non-linear wave-wave interactions

(quadruplets and triads) explicitly with state-of-the-art formulations. Wave blocking by currents is also

explicitly represented in the model. Diffraction is not modelled in SWAN, but diffraction effects can be

simulated by applying directional spreading of the waves. Wave reflection can be modelled by

specifying the location and reflection coefficient of a linear obstacle, e.g. a breakwater. In the present

application, the processes simulated were wave refraction, shoaling, bottom friction and breaking.

Local wind-wave generation within Saldanha Bay was not modelled since wind-waves were assumed

largely to be included in the boundary specifications.

In the application of SWAN, the first step is to synthesize an offshore wave climate. Once the

offshore wave climate is determined, this climate can be simulated in the model and the resulting

wave shear stresses and wave energy dissipation are used by the Delft3D-FLOW module.

Offshore wave climate To execute Delft3D-WAVE, the model requires an offshore wave climate. For the modelling period of

20 October 2006 to 15 December 2006 used to calibrate the model wave parameters, the offshore

wave conditions for the SWAN refraction model were obtained from measurements at the Slangkop

waverider buoy (Figure B.7). The three-hourly measurements from 01 October 2006 to 31 December

2006 comprise a total of 736 data points. For the period from 01 July 1999 to 30 June 2000, hindcast

data obtained from OCEANOR were used. For this period, the hindcast wave time series consists of

1464 wave conditions at six-hourly intervals.

For the modelling, Delft3D-WAVE and Delft3D-FLOW were executed in “online” mode with a wave

calculation taking place every two hours: This implies that every two hours the water levels calculated

by Delft3D-FLOW are passed to Delft3D-WAVE which calculates the appropriate wave field and then

passes the wave shear stresses and rate of wave energy dissipation, as well as other parameters to

Delft3D-FLOW. Because the wave calculations take place at two hourly intervals, should a wave

calculation be performed at a time that does not coincide with a time in the wave time series, the

wave conditions for the modelling are obtained from the measured or hindcast data via linear



interpolation of the significant wave height, peak wave period and mean wave direction. The

measured wave time series from Slangkop are presented in Figures B.8a and B.8b for the 20

October 2006 to 15 December 2006 period while the OCEANOR hindcast wave data is presented in

Figures B.9a to B.9b for the period 01 July 1999 to 30 June 2000. The latter data was used to

generate the one year hydrodynamic database used in both the RO Plant and the Phase 2 expansion

of iron-ore export facilities specialist studies.

Figure B.7: Location of the Slangkop waverider buoy relative to the waverider buoy within

Saldanha Bay.



Figure B.8: The Slangkop wave data from 19 October 2006 to 15 December 2006.



Figure B.9a: OCEANOR wave data from 30 June 1999 to 1 January 2000.



Figure B.9b: OCEANOR wave data from 1 January 2000 to 1 July 2000.



SWAN vs 71079 (off the Iron Ore Jetty)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

14-Nov 19-Nov 24-Nov 29-Nov 4-Dec 9-Dec

Date

Hm

o

SWAN - Madsen=0.05 Seapac 71079

SWAN vs 71079 (off the Iron Ore Jetty)

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0

18.0

20.0

14-Nov 19-Nov 24-Nov 29-Nov 4-Dec 9-Dec

Date

Tp


Estimated accuracy

Seapac

Hmo: 5 % Direction: ± 5°

Wave Model Calibration Besides the bathymetry, calculation domain and offshore wave conditions, the Delft3D-WAVE model

also has certain parameters that may be set by the user. During the 2006 measurement campaign

wave data was collected at location A with a SeaPac and at location B (the same SeaPac instrument

as used at location A but deployed at later date and also an S4 deployemnt) as shown in Figure B.1.

The wave measurements at locations A and B, together with other shallow water measurements in

Big Bay were used to calibrate the Delft3D-WAVE model (Smith et al., 2007). The comparisons

between the model simulations and these measured wave data are presented in Figures B.10 and

B.11.

Figure B.10: Comparison between Delft3D-WAVE simulations and measurements at location A (Seapac 71079). The estimated accuracy of the SEAPAC measurements indicated in the plots are based on the work of Schüttrumpf et. al. (2006).



Figure B.11: Comparison between Delft3D-WAVE simulations and measurements at location B (Seapac 71079 and S4). The estimated accuracy of the SEAPAC measurements are based on the work of Schüttrumpf et. al. (2006).

Examination of the Figures B.10 and B.11 indicates that the Delft3D-WAVE wave height and period

parameters generally compare well with the measurements. Some minor discrepancies occur that

may be the result of local wind energy not being included in these calibration simulations. In Figure

B.10, the small differences in the Delft3D-WAVE and SeaPac directions (taking into account the

SWAN vs S4

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0

18.0

20.0

14-Nov 19-Nov 24-Nov 29-Nov 4-Dec 9-Dec 14-Dec

Date

Tp

SWAN - Madsen=0.05 S4 Seapac

Big Bay Location

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5


Date

Hm

o

SWAN - Madsen=0.05 S4 Seapac 71079

Big Bay Location

220.0

225.0

230.0

235.0

240.0

245.0

250.0

255.0

260.0


Date

Mea

n D

ir (d

eg)


S4

Seapac

Hmo: 5 % Direction: ± 5°

Estimated accuracy



measurement accuracy of the SeaPac) confirms that the model represents the wave direction within

measurement accuracy at this location (i.e. in the middle of Big Bay).

The SEAPAC measurements and simulations at the jetty location A, both indicate that the waves

were small, with significant wave heights (Hmo) less than 0.5 m for most of the time (Figure B.10).

Considering that the SeaPac was placed in a water depth of 12 m, it is expected that the reliability of

the measured wave direction will be poor, due to the limited sensitivity of the SEAPAC in measuring

small waves from this depth. This is due to the fact that the wave directional information is being

derived from the small horizontal orbital current velocity components measured by the SEAPAC by applying a transfer function, based on the Linear Wave Theory (Van Tonder, 1994). Therefore, the

direction data are not considered for as being adequate for calibration purposes and consequently

are not presented in Figure B.10. (It should be noted that the primary function of this instrument

placed at the Jetty was to measure long-waves as part of a separate study.)

Although the computational grid in the present study is different from the ones used in the study upon

which these calibrations are based ( Smith et al., 2007) , the model calibration is considered to

remain valid as the same parameter settings have been used in both studies. These are summarised

in Table B.1. Further, for the simulation undertaken for the 1999-2000 period, it was ensured that the wave module completed the same number of iterations for both RO Plant simulations (this study) and

the Phase 2 simulations both with and without the changes to the harbour (Smith et al., 2007). The

wave module has been “on-line” with the flow module (i.e. in a coupled mode) to ensure that the

wave simulations use the correct water levels at all times.

Table B.1 Wave model parameters

Parameter Value Directional sector 0° - 360°

No. directions in directional space 36 Lowest frequency 0.03 Hz Highest frequency 0.5 Hz No. frequency bins 24

Spectrum Jonswap Spectral peak enhancement factor 2.0

Width energy distribution 18.2 - 1.66xTp for Tp < 9.3 s 1.09xTp - 7.6 for Tp > 9.3 s

Depth induced breaking Battjes and Janssen Alfa 1

Gamma 0.73 Bottom friction Madsen

Friction coefficient 0.05 Non-linear triad interactions Activated

Wind growth No Whitecapping Activated Quadruplets No

Frequency shift Activated Reflection No



B2.4 Hydrodynamic simulations and Water Quality Simulations Delft3D-FLOW (Lesser et al., 2004) is a three-dimensional hydrodynamic model that includes

formulations and equations that take into account the following processes:

• tidal forcing;

• wind forcing;

• wave forcing;

• baroclinic currents and vertical mixing induced by changes in water temperature resulting from

both advection of warmer/cooler water into the bay as well as local air-sea interactions;

• the effect of the earth’s rotation (Coriolis force).

The system of equations in Delft3D-FLOW comprise the horizontal momentum equations, the

continuity equation, the equation of state and the advection-diffusion equation for heat, salt and other

conservative tracers which are solved using the Alternating Direct Implicit scheme. Vertical

turbulence is modelled using the k-ε turbulence closure model. The computation grid is an irregularly-

spaced, orthogonal, curvilinear grid in the horizontal and a sigma-coordinate grid in the vertical. The

equations and their numerical implementation are described in detail in WL|Delft Hydraulics (2005)

and a very clear exposition is also presented by Lesser et al. (2004).

Processes included in the model The processes modelled in the Saldanha Bay-Langebaan Lagoon system include the following:

• Tides: The Saldanha Bay-Langebaan Lagoon system has a strong marine character with its

waters originating in the adjacent Benguela Upwelling System. Tidal forcing is strong at depth,

in the vicinity of the mouth of Saldanha Bay and with increasing proximity to Langebaan

Lagoon. Tides play an important role in the forcing of currents in this system and tides have

thus been included in the modelling via water level variations applied to the model boundary.

• Wind-driven flows and mixing: In general, wind-forcing is the dominant physical forcing

mechanism determining the surface layer current speed and direction in both Small and Big

Bay as well as mixing processes in the water column. Wind times series are specified for the

whole model domain. Wind setup and Coriolis tilt effects on the water levels at the model

boundaries also have been taken into account in the simulations.

• Wave-driven flows, wave stirring and generation of turbulence: Saldanha Bay is exposed

to swell waves from the Atlantic Ocean as well as locally-generated wind-waves. It is expected

that wave-driven currents inside Saldanha Bay will be confined to the narrow surf-zone in Big

Bay and are unlikely to play a significant role in the large-scale circulation of Saldanha Bay.

However these flows are expected to play a significant role in the dispersion of effluents at

Site 1. Waves also play a significant role in the fate of fine particles via the enhancement of

bed shear stresses (i.e. stirring up of sediments) at the seabed. Explicit simulations of the

transport and fate of sediments (i.e. backwash sediments) have not been undertaken here.



• Water column structure, Temperature and Salinity: Both the temperature and salinity of the

water column has been modelled. The pycnocline dynamics and baroclinic currents play an

important role in the hydrodynamic and water quality functioning of Saldanha Bay. Measured

salinities in Saldanha Bay lie in a limited range from 34.8 to 35.0 psu in Big Bay (Lufer et al.,

1997) and thus do not play a major role in the baroclinic dynamics and water column density

structure in the bay. Temperature effects thus almost exclusively determine the density

dynamics of the water column both within the bay and over the adjacent shelf. However, given

that the proposed brine discharge plume behaviour is highly dependant on salinity, it has been

necessary to simulate both temperature and salinity in this study.

The water column structure in Saldanha Bay is seasonal, varying from a strongly thermally

stratified water column for the most of the year (August to May) to well-mixed conditions during

the mid-winter months (June to July). The water column structure is determined by the

opposing forces of buoyancy inputs (due to atmospheric heat fluxes into the surface waters

and the input of cold bottom waters into the Bay due to upwelling forcing from the adjacent

shelf) that enhance water column stratification and mixing of the water column by primarily

winds but also waves (i.e. wind- wave- and tidally driven currents and turbulence resulting in

mixing of the water column). These processes thus control the pycnocline dynamics and

vertical mixing of the water column which, together with advective fluxes by wind- wave- and

tidally-driven currents, ultimately determine the behaviour of biogeochemical parameters and

discharges into the Bay. Thus it has been necessary to explicitly include all of these processes

in the model (e.g simulation of air-sea fluxes , wave generated turbulence, etc)..

Model input times series The input times series to the model include sea level (predicted tides and wind-driven sea level

changes) and water column structure at the open boundaries of the model, wind and a range of

atmospheric variables. The wave information is obtained from the coupled wave model.

Air-sea fluxes A heat flux model is incorporated to account for air-sea interactions. The incoming solar and

atmospheric radiation, air temperature and relative humidity is prescribed, while the terms related to

heat loss (evaporation, back radiation and convective heat flux) are computed by the model.

The times series of incoming solar radiation (Figure B.12) is estimated from theoretical clear sky

radiation (Seckel and Beaudry, 1973), reduced by a factor of 0.685 (assumed transmissivity of the

atmosphere in this region) and further modified (reduced) by measured cloudiness (Figure B.13)

based on the observed mean cloudiness at the nearby Cape Columbine lighthouse. The relative

humidity values (Figure B.14) used are those measured at Cape Columbine daily at 08:00B, 14:00B

and 20:00B while the air temperature time series (Figure B.15) used are those measured at Saldanha

Bay Port Control.



Figure B.12: Theoretical clear sky radiation for the period simulated.

Figure B.13: Measured cloudiness converted to percentage cloud cover from cloudiness estimates in octals made daily at the Cape Columbine lighthouse at 08:00, 14:00 and 20:00 B).



Figure B.14: Air temperatures measured at Saldanha Port Control for the period simulated.

Figure B.15: Relative humidity (%) measured daily at 08:00B, 14:00B and 20:00 B) at the Cape Columbine lighthouse.



WInd The wind used in the simulations was based on hourly wind data measured at Sladanha Bay Port

Control in Small Bay. Since the sensor is located on a hill 50 m above mean sea level (MSL), the

measured wind speeds were reduced to represent the wind speed 10 m above MSL which is required

by the numerical model. Therefore, a correction factor of 0.9 was applied to the measured wind

speeds. Time series of wind speed and direction for the 1999-2000 model simulation period are given

in Figures B.16a and b. The wind speed and direction for the 2006 calibration period are plotted in

Figure B.17.

Sea Level at the Open Boundaries The sea level specified at the open boundaries of the model comprises predicted tides as well as

wind-driven water level changes over the shelf.

The open boundaries are located along the three offshore edges of the model, that is the Southern,

Western and Northern boundaries. At these open boundaries, a water level time-series is specified

which is based on the predicted tide with a modification to compensate for time-varying wind setup

and Coriolis tilt effects. The 8 largest amplitude tidal constituents along the West Coast (Rosenthal

and Grant, 1989) used to predict the tide are listed in Table B.2. There is only a small tidal lag along

the west coast of southern Africa, so differences in the tidal phases between the northern, western

and southern open boundaries were ignored. The water levels at eth open boundary are specified at

2 minute intervals.

Table B.2: Tidal constituents for Saldanha Bay (Rosenthal and Grant, 1989)

Tidal constituent

Amplitude

(m)

Phase

(degrees)

M2 0.489 90.6

S2 0.213 112.0

N2 0.132 78.8

K2 0.070 112.2

K1 0.056 134.8

MU2 0.021 64.4

O1 0.015 260.1

P1 0.014 131.0

The tide sea levels specified in the model for the full one year simulation period (1999 to 2000) is

shown in upper panels of Figures B.16a and b. The predicted sea levels for the 2006 calibration

period are plotted in Figure B.17.



Figure B.16a: Water levels and wind time series input for the modelling from 30 June 1999 to 1 January 2000.



Figure B.16b: Water levels and wind time series input for the modelling from 1 January 2000 to 1 July 2000.



Figure B.17: Water levels and wind time series input for the modelling from 19 October 2006 to 15 December 2008.



Water column Stratification at the Open Boundaries

The model includes baroclinic effects (currents and vertical mixing) induced by changes in water

temperature resulting from both advection of warmer/cooler water into the bay as well as local

insolation effects. The inputs for the heat flux model have already been discussed (see start of this

section).

The advection of warmer/cooler water into the bay is controlled by the temperature profile specified

on the open boundaries. Based on available measurements temperature profiles for the 1999 to 2000

period (Table B.3) were developed for input into the model.

Table B.3: Seasonal temperature profiles for 1999-2000

The temperature profiles used at the model boundaries in the monthly simulations undertaken for the

RO Plant and Phase 2 expansion of iron-ore export facilities specialist studies were obtained from

these measurements via linear interpolation and are listed in Table B.4.

Table B.4 Temperature profiles applied on model boundaries for 1999-2000

Temperature (oC) Depth (m) May

1999 June199

9 July 1999

Aug 1999

Sept 1999

Oct 1999

Nov 1999

0 16 16 16 16 16 16 16 15 13 13.5 14 14 14 14 14 30 12 12 12 13 12.7 12.3 12 50 11 10.5 10 12 11 10 9 120 9 9 9 10 9.7 9.3 9 200 9 9 9 9 9 9 9 280 8 8 8 8 8 8 8

Temperature (oC) Depth (m) Dec

1999 Jan 2000

Feb 2000

Mar 2000

Apr 2000

May 2000

Jun 2000

0 16 16 16 16 16 16 16 15 13.7 13.3 13 13 13 13 13.5 30 11.7 11.3 11 11.3 11.7 12 12 50 9 9 9 9.6 10.3 11 10.5 120 9 9 9 9 9 9 9 200 9 9 9 9 9 9 9 280 8 8 8 8 8 8 8

Temperature (oC) Depth (m) May 1999 Aug 1999 Nov 1999 Feb 2000 May 2000 0 16 16 16 16 16

15 13 14 14 13 13 30 12 13 12 11 12 50 11 12 9 9 11 120 9 10 9 9 9 200 9 9 9 9 9 280 8 8 8 8 8



Based on the temperature measurements taken from 23 November 2006 until 15 December 2000

and the summer profiles listed in Table B.3 and B.4, a stationary temperature profile as listed in Table

B.5 was adopted for use in the 2006 simulations used to calibrate the wave and hydrodynamic

models.

Table B.5: Temperature profile applied on model boundaries for 2006

Depth (m)

Temperature (°C)

0 16 15 14 30 12 50 9

120 9 200 9 280 8

In addition to this information, the model also requires certain parameter specifications. The

additional input parameters for the hydrodynamic model which were determined from the 2006

simulation period are listed in Table B.6. Thess have been used in the RO Plant modelling study.

Table B.6: Hydrodynamic model parameters

Parameter Value Wind drag coefficient (Cd) 1.1 x 10-3 + 6.5x10-3 x Vwind

Horizontal eddy viscosity Variable (Grid dependant)

Background vertical eddy viscosity 0.0005 m2.s-1

Horizontal eddy diffusivity Variable (Grid dependant)

Background vertical eddy diffusivity 0.000001 m2.s-1

Bed friction (White-Colebrook coefficient) 0.05 m

Correction for sigma coordinates On

Horizontal Forester filter On

Vertical Forester filter Off

Time step 1 minute

The wave forcing is obtained from the rate of wave energy dissipation computed by the Delft3D-

WAVE model. The enhanced bed stresses due to the wave effects are incorporated in the model

using the friction formulation of Fredsøe (Soulsby et al. 1993).



Calibration of the Hydrodynamic Model

The hydrodynamic model was calibrated by comparing the model results to currents and water temperatures measured by the CSIR at the 3 locations (marked A to C in Figure B.1) for the period

20 October 2006 and 15 December 2006. The measurements were obtained during two

deployments. At location A currents were measured at 1.5 metres from the seabed for the period

20 October 2006 to 22 November 2006. Similarly, at location B currents were measured at 1.5

metres from the seabed for the period from 23 November 2006 until 15 December 2006. During this

latter period temperature was measured at a range of depths through the water column at position C.

A comparison of the modelled and predicted currents in the vicinity of the jetty (position A) is

presented in Figures B.18 and B.19. In these figures the U-component indicates the Easterly-component of the crrent while the V-component is the Northerly-component of the current.

Scatterplots of the northerly components against the easterly components reveal the distribution of

current magnitudes and directions as shown in Figure B.20. The scatterplots show that Delft3D-

FLOW tends to predict smaller variations in velocity than those measured. The model, nevertheless

still represents the measurements very well at this particular location.

The level of calibration of the model can further be analysed by considering the performance

statistics advocated by Sutherland et al. (2004). According to Sutherland et al. (2004), the

performance of a model compared to measurements can be assessed by the adjusted relative mean absolute error (ARMAE) where the adjustment indicates incorporation of the influence of

observational errors (OE). They also provide a classification system based on the adjusted relative

mean absolute error to assess the performance of a model compared to measurements. For current

measurements, Sutherland et al. (2004) recommend a value for the observational errors of 0.05 m.s-1

which is much larger than the average error (2% of average speed) in the SeaPac instrument which

is approximately 0.001 m.s-1. Using the value of 0.05 m.s-1 as observational error gives an excellent

calibration according to the ARMAE values listed by Sutherland et al. (2004) while an observational

error of 0.001 m.s-1 yields a reasonable calibration. This indicates that the calibration of the

hydrodynamics is sufficient in the vicinity of the jetty in Big Bay (i.e. at locations proposed for RO

Plant discharges into Big Bay).

The comparison between the bottom currents near the centre of the bay (position B in Figure B.1) is

presented in Figures B.21 and B.22. Here the model also represents the measurements fairly well

despite the fact that in this part of the bay there exist large circulation eddies driven by the interaction

between the tidal flow and wind driven currents. The scatterplots in Figure B.23 of the northerly

current components against the easterly current components also show that the measured current

distributions are complex and that there is greater variability in the measured currents than produced

by the model. However, the measured outliers in Figure B.23 along Ueast of approximately -0.16 m.s-1

are attributed to a single strong wind event on 10 December 2006. The performance statistics outlined in the previous paragraph indicates a reasonable to good calibration for this position.



Figure B.18: Measured and modelled bottom currents at position A (see Figure B.1) from 22 October 2006 to 5 November 2006

Figure B.19: Measured and modelled bottom currents at position A (see Figure B.1) from 5 November 2006 to 22 November 2006.



-0.2 -0.1 0 0.1 0.2Ueast (m/s)

-0.2

-0.1

0

0.1

0.2

Uno

rth (

m/s

)

-0.2 -0.1 0 0.1 0.2Ueast (m/s)

-0.2

-0.1

0

0.1

0.2

Uno

rth (

m/s

)

Measurements Model

Figure B.20: Scatterplots of the measured and modelled bottom currents at position A (see Figure B.1).

Figure B.21: Measured and modelled bottom currents at position B (see Figure B.1).



Figure B.22: Measured and modelled bottom currents at position B (see Figure B.1) from 3 December 2006 to 15 December 2006.

Figure B.23: Scatterplots of the measured and modelled bottom currents at position B (see Figure B.1).

-0.2 -0.1 0 0.1 0.2Ueast (m/s)

-0.2

-0.1

0

0.1

0.2

Uno

rth (

m/s

)

-0.2 -0.1 0 0.1 0.2Ueast (m/s)

-0.2

-0.1

0

0.1

0.2

Uno

rth (

m/s

)

Measurements Model



Comparisons between the measured near the centre of the bay (position C in Figure B.1) and

modelled surface and bottom temperatures at the same location are presented in Figures B.24 and

B.25. The model predictions show the correct trends although the exact measured temperatures are

at times not reproduced. Additional information on the calibration is presented in the regression plots

presented in Figure B.26. These indicate that the surface temperatures are well correlated while the

model tends to indicate warmer bottom temperatures than do the measurements. The reason for this

is that on occasion the water column stratification is not sufficiently maintained in the model. The

performance statistics of Sutherland et al. (2004) may also be applied to temperatures. Using a

minimum observation error of 0.1 ºC (Seamon temperature gage), the calibration of temperatures at

this location in the model may be classified as good.

In general, it may be concluded that the model can reliably simulate the overall tidal, wind-driven,

wave-driven circulation and water column mixing processes. Differences between the modelled and

measured data, in general, may be ascribed to penetration into the bay of large–scale circulation

features over the adjacent shelf and further offshore (i.e. features generated outside the model

domain), as well as localised effects such as spatial variation of winds over the bay and rip currents

not accounted for in the model forcing. Limited spatial resolution in the computational grids also may

result in differences between the model results and measurements. This is expected to be less of an

issue for the RO Plant simulations as the resolution of computational grids used in this study

significantly exceeds that of the computational grids used in the Phase 2 modelling study and to

calibrate the model.

Figure 6.24: Comparison between measured and modelled temperatures at position C (see Figure B.1) from 23 November 2006 to 3 December 2006.



Figure B.25: Comparison between measured and modelled temperatures at position C (see Figure B.1) from 3 December 2006 to 15 December 2006.

Figure B.26: Regression plots of the measured and modelled surface and bottom temperatures at position C (see Figure B.1).

5 10 15 20Tmodel (oC)

5

10

15

20

T mea

s (o C

)

5 10 15 20Tmodel (oC)

5

10

15

20

T mea

s (o C

)

Surface Bottom



Figure B.27: Measured and modelled temperatures at North Buoy (see Figure 6.2) from 1 July1999 to 1 October 1999

Figure B.28: Measured and modelled temperatures at North Buoy (see Figure 6.2) from 1 October 1999 to 1 January 2000.



Figure B.29: Measured and modelled temperatures at North Buoy (see Figure B.1) from 1 January 2000 to 1 April 2000.

There exists additional data in the form of temperatures measurements that were taken during 1999-

2000 at North Buoy in Small Bay (indicated on Figure B.1) that can be used to further calibrate or

verify the model. The measured temperatures near the surface and near the bottom are compared

to the modelled temperatures in Figures B.27, B.28 and B.29. As indicated in these figures, the

modelled results compare very well to the measurements. Considering the position of the

measurement location (North Buoy), these results indicate that the intrusion and retreat of cold water

associated with summer upwelling and downwelling as well as the tidal fluxes into the bay are

predicted correctly.

In summary, the overall accuracy of the model. is considered satisfactory for the purpose of simulating the transport and fate of the brine discharge plumes.


