Water Journal March 2008

AWA AUSTRALIAN WATER ASSOCIATION ABN 78 096 035 773

entiey WaterGEMS V8 XM Edition continues with Haestad Methods' tradition of pioneering research and innovation, advancing the water modelling technology standard once again.

B

ELIMINATE PLATFORM RESTRICTIONS Being anchored to a single platform is costly and risky.

WaterGEMS V8 XM Edition is the only solution that lets everyone work within their favorite platform, sharing a single model file.

SHATTER PERFORMANCE BARRIERS Designed to support all-inclusive models, WaterGEMS V8 XM Edition gets a new power boost to build, run, edit, and map large water distribution models with ease, at recordbreaking speeds.

EASIER THAN EVER! The legendary ease of use of Haestad Methods solutions is magnified with WaterGEMS V8 XM Edition. Enjoy a fresh new stand-alone interface, an improved ArcGIS integration, and an all-new MicroStation platform with dozens of new features and ground-breaking utilities:

~~ ~-: ~~~ ""

~~~ <Ill

• Criticality analysis

• Stand-alone GIS demand allocation

• Network trace queries

• Network segmentation analysis

• Stand-alone Terrain Extraction

• Pressure-dependant demands

• New isolation valve element

• Network navigator

• Variable speed pump batteries

• Centralised water use management

• New hydrant element

• Unit demand engineering libraries

• Dynamic SOL-based queries

• Auxiliary fire flow results

Learn more about the award-winning WaterGEMS V8 XM Edition at:

www.bentley.com/AWA Contact us today for more information: Toll-free (Austra lia): 1800 245 005 • Toll-free (New Zealand): 0800 444 046 e-mail: anz.marketing@bentley.com

ENVIRO 08: Visit booth 165 to pick up your FREE Water Modelling Guide CD! May 5-7, 2008 • Melbourne Exhibition & Convention Centre

~ AWA CONTACT DETAILS â&#x20AC;¢ 'Promoting the sustainable : ~ management ofwater'

POSTAL ADDRESS PO Box 388, ARTARMON NSW 1570

Journal of the Australian Water Association

EMAIL info@awa.asn .au WEBSITE www.awa.asn.au PRESIDENT

ISSN 0310-0367

David Barnes - president@awa.asn.au

CHIEF EXECUTIVE Tom Mollenkopf - cbeer@awa .asn.au

CHIEF OPERATIONS OFFICER Ian Jarman - ijarman@awa.asn.au

EVENTS Wayne Castle - 61 2 9495 9921 wcastle@awa.asn.a u

MEMBERSHIP INFORMATION AND ENQUIRIES Michael Seller - 61 2 6581 3483 mseller@awa.asn.au

MEMBERSHIP RENEWALS AND CHANGES Membership Team - 1300 361 426 info@awa.asn.au

MEDIA AND MARKETING Edie Nyers - enyers@awa.asn.au

SCIENTIFIC AND TECHNICAL INFORMATION Diane Wiesner PhD - 61 2 9495 9906 dwiesner@awa.asn.au

WATER EDUCATION NETWORK Corinne Cheeseman - 61 2 9495 9907 ccheeseman@awa.asn.au

NATIONAL SPECIALIST NETWORKS Laura Evanson - 61 2 9495 9917 levanson@awa.asn.au

AWA BRANCHES: AUSTRALIAN CAPITAL TERRITORY and NEW SOUTH WALES Tanya Webeck - 6 1 2 9495 9908 actbranch@awa.asn.au, nswbranch@awa.asn.au NORTHERN TER RITORY Hayley Galbraith - 61 2 9495 99 19 ntbranch@awa.asn.au SOUTH AUSTRALIA Sarah Carey - 61 8 8267 1783 sabra nch@awa.asn.au QUEENSLAND Kathy Bourbon - 61 7 3397 5644 qldbranch@awa.asn.au TASMANIA & VICTORIA BRANCH Rachel-ann Martin - 61 3 9235 14 16 tasbranch@awa.asn.au vicbranch@awa.asn.au WESTERN AUSTRALIA Cath Miller - 0416 289 075 wa branch@awa.asn.au INTERNATIONAL WATER ASSOCIATION, AUST. (IWAA) iwabranch@awa.asn.au

DISCLAIMER Australian Water Association assumes no responsibility for opinion or statements of facts expressed by contri butors or advertisers.

COPYRIGHT AWA Water Journal is subject to copyright and may not be reproduced in any format without written permission of AWA. To seek permission to reproduce Water Journal material email your request to: enyers@awo. asn.au

Volume 35 No 2 Morch 2008

AWA WATER JOURNAL MISSION STATEMENT 'To provide a print ;ournal that interests and informs on water matters, Australian and international, covering technological, environmental, economic and social aspects, and to provide a repository of useful refereed papers. ' PUBLISH DATES Water Journal is published eight times per year: February, March, May, June, Augusl, September, November and December EDITORIAL BOARD Chairman: FR Bishop BN Anderson, TAnderson, CDiaper, GFinlayson, AGibson, GA Holder, BLabza, MMunlisov, CParler, DPower, FRoddick EDITORIAL SUBMISSIONS Water Journal inviles editorial submissions for: Technical Papers and topical articles, Opinion, News, New Products and Business Information.

Acceptance of editorial submissions is subject to editorial board discretion.

Email your submissions to one of the following three categories: 1. TECHNICAL PAPERS AND FEATURES Bob Swinton, Technical Editor, Water Journal: bswinton@bigpond.net.au AND journal@awa.asn.au Papers of 3000-4000 words {allowing for graphics); or topical stories of up to 2,000 words. relating to all areas of the water cycle and water business.

Submissions are tabled at monthly editorial board meetings and where appropriate are assigned to referees. Referee comments will be forwarded to the principal author for further action. See box on page 4 for more details.

2. OPINION, INDUSTRY NEWS, PROFESSIONAL DEVELOPMENT Edie Nyers, enyers@awa.asn.au Arlicles of 1000 words or less 3. WATER BUSINESS Brian Raul!, National Sales & Advertising Manager, Hallmark Editions brian.rault@halledit.com.au Water Business updates readers on new products and associated business news within the water sector.

ADVERTISING Brian Raul!, National Sales & Advertising Manager, Hallmark Editions Tel: 61 3 8534 5014 (direct), 61 3 8534 5000 (switch), brian.rault@halledit.com.au Advertisements are included as an information service to readers and are reviewed before publication to ensure relevance to the water environment and objectives of AWA. PURCHASING WATER JOURNAL Single issues available @ $12.50 plus postage and handling; email dwiesner@awa.asn.au BACK ISSUES Water Journal back issues are available to AWA members at www.awa.asn.au PUBLISHER Hallmark Editions, PO BOX 84, HAMPTON, VICTORIA 3188 Tel: 61 3 8534 5000 Fax: 61 3 9530 8911 Email: hallmark.editions@halledit.com.au Journal of the Australian Water Association

water

MARCH 2008 ,_

Journal of the Australian Water Association

Sydney Desalination Plant: Pre-treatment Pilot Testing - see page 90

South Gippsland Water - Saline Wastewater Project - see page 100

Queensland's Western Corridor Recycled

Water Project - see page 12 1

OPINION AND INDUSTRY NEWS OPINION Ozwater- Melbourne 2009, Brisbane 2010, Adelaide 2011 Giving Water a Voice WaterAid My Point of View CROSSCURRENT International, National Issues and Policy, States, People in the News AWA NEWS NATIONAL SPECIALIST NETWORKS YOUNG WATER PROFESSIONALS NATIONAL EVENTS CALENDAR AWA MEMBERSHIP NEWS New Members SPECIAL REPORT· CLIMATE CHANGE Australia's Climate Change Policies Operating in a Carbon Constrained World - The Challenges and Opportunities Accounting for Carbon - Report of the AWA Conference Observations from Presentations at the AWA Accounting for Carbon Conference CONFERENCE REPORT

DBarnes, President, AWA TMollenkopf, Chief Executive, AWA ALaidlaw SPosseh

4 5 6 8 12 22

34 38

39 40

Senator Penny Wong Ross Young EA (Bob) Swinton Diane Wiesner

41

44 46 52 54

TECHNICAL FEATURES !I ·,I indicates the paper has been refereed) EMISSION MINIMISATION [ii Fugitive Greenhouse Gas Emissions from Wastewater Systems Quantifying fugitive emissions is an area of uncertainty for the industry AUSTRALIA'S DESALINATION PLANTS Environmental Monitoring Program for Sydney's Desalination Plant Testing for impacts of intake and concentrate discharge [i] Sydney Desalination Plant: Pre-treatment Pilot Testing Deciding on the most appropriate and economic pre-treatment process Desalination In Australia: A Review Aquick review of what's going on around the continent SALINE OUTFALLS South Gippsland Water - Saline Wastewater Project Trenchless technology reduces environmental impact MEMBRANE TECHNOLOGY [I] Pre-treatments for Seawater Reverse Osmosis Conventional coagulation/sedimentation/filtration or low pressure membranes 2 MARCH 2008

Water

Journal of the Australian Water Association

J Foley, PLant, PDonlon

62

RKidson, STrousdale, J Wood, GAllen

74

CLPort, SJ Roddy

90

GCrisp, EA (Bob) Swinton

94

EFoley

100

BBolto, TTran, MHoang

105

Volume 35 No 2 March 2008

Asian experts share SA' s groundwater expertise - see page 125

Odour Exposure Criteria and Odour Modelling in WA - see page 144

Environmentally Sustainable Odour Control for Merrimac WWTP - see page 149

Research on Advanced Membrane Technologies Australia at the forefront of novel membrane development [ii Optimising Mixing in Full-Scale MBRs: CFO Modelling and Validation CFD will prove to be a valuable design and optimisation tool FURTHER OPTIONS FOR WATER RESOURCES [ii Effectiveness of Domestic Rainwater Tanks Most domestic rainwater tanks are not connected to household plumbing Queensland's Western Corridor Recycled Waler Project: Alliances in Action Ano-blame culture predominated the alliance ASIA PACIFIC ISSUES Asian Experts Share SA's Groundwater Expertise The Challenges in Asia lil Stormwaler in Thailand: APolled History Wetlands as a key management approach to excess overland flow

SGray

111

MWDBrannock, YWang, GLeslie

114

RECollins

116

CBerry

121

J McKay Compiled by EA {Bob) Swinton

125 127

KHarriden

129

ODOUR MANAGEMENT [i] Predicting Hydrogen Sulfide Formation in Sewers: A New Model Simulating biological, chemical and physical transformations KSharma, DWde Haas, S Corrie, KO'Halloron, J Keller, ZYuan [ii Odour Control by Chemical Dosing: A Case Study Oxygen probably the most cost-effective if dosing is optimised DWde Haas, KSharma, S Corrie, KO'Halloron, J Keller, ZYuan [i] Odour Exposure Criteria and Odour Modelling in Western Australia Odour criteria should reflect a distribution of odour concentrations IWallis, KCadee Environmentally Sustainable Odour Control for the Merrimac WWTP Upgrade Four bioscrubbers followed by four activated carbon filters represent minimum whole-of-life cost GFinke, POliver, MThomas, I Evanson COMMUNITY CONSULTATION [ii Attitudes lo the Natural Resources Management Levy in Adelaide "The problem with water is not water but society. Lundqvist et al. (200 I) 11

132 138 144 149

ZWu, J McKay, EHemphill

154

ZSlavnic

157

OPERATION Step Screens - Design, Installation and Operational Factors Apragmatic reviewof a relatively new mechanism

WATER BUSINESS NEW PRODUCTS AND BUSINESS INFORMATION · SPECIAL FEATURE: UV DISINFECTION ADVERTISERS' INDEX I

160 176

OUR COVER As urban development relentlessly reduces buffer zones, the management ofodour becomes increasingly important. The paper, page 149, describes the technology installed at the Merrimac WWTP on the Gold Coast. Photo courtesy of the Men ·imac WWTP Waterfuture Alliance.

Journal of the Australian Water Association

Water

MARCH 2008 3

OZWATER - MELBOURNE 2009, BRISBANE 2010, ADELAIDE 2011 management ream lead by Wayne Castle. It is planned that there will be enhanced continuity in technical areas such as programme planning and sponsorship to assist local organising committees. The technical and management support will be more appropriate with the annual cycle and should reduce the burden of reinvention for each Ozwater.

I am delighted to report that the highly regarded Ozwater Conference and Exhib ition is to be held annually.

AWA was formed as the AWWA in June 1962 in September 1964, with the first Federal Convention held in Canberra. At this first event there were 64 delegates and one of the themes was water reuse. Since that rime the event has been held biennially and has grown and developed substantially.

In 1993 at the 14th Federal Convention a fo rmal exhibition was included for the first time. The most recent Ozwater was held in Sydney in March 2007 and attracted over 1100 delegates, over 200 trade displays, 17 sponsors and an impressive array of speakers.

The biennial format has served AWA well, however, in the non Ozwater years it has left a void in the national programme. This has become more noticeable in recent years

water

Editorial Submissions Technical Papers Water journal welcomes the submission of papers equivalent to 3,000-4,000 words (allowing for graphics) relating to all areas of the water cycle and water business to be published in the journal. Topical stories of up to 2,000 words may also be accepted.

All submissions of papers intended for the main body of the journal should be emailed to the Technical Editor, bswinton@bigpond.net.au and copy, fo r back up, to journal@awa.asn.au.

Shorter news items should be emailed to enyers@awa.asn.au. A submitted paper will be tabled at a monthly Journal Committee meeting where, if appropriate, it will be assigned to referees. Their comments will be passed back ro the principal author. If accepted and after any comments have been dealt with, the final paper can be emailed with the text in MS Word bur with high resolution graphics (300 dpi riff, jpg or eps files) as separate files.

Authors should be mindful that Water journal is published in a 3 column 'magazine' fo rmat rather than the full-page format of Word documents. Graphics should be set up so that they will still be clearly legible when reduced to C\Yo-column size (about 12cm wide). Tables and figures need to be numbered with the appropriate reference in the text e.g. see Figure 1, nor just placed in the text with a (see below) reference as they may end up anywhere on the page when typeset. See page 1 for more derails on rhis and other editorial submissions.

4 MARCH 2008

Water

Dr David Barnes, AWA President

with the unparalleled interest and activity in all aspects of water. In response to feedback received from members, sponsors, exhibitors and delegates it has been concluded that an annual Ozwater Conference and Exhibition is in the best interests of the water sector.

As the key national event it will continue to bring together AWA members and other industry players. The broad goals remai n unchanged; to bring the water industry together to meet, transfer knowledge, network, do business and ensure that they are abreast of current and emerging issues.

Other AWA activities such as specialist and regional conferences, Enviro, IWA meetings, seminars, master classes etc will continue with the usual vigour. Clearly any additional revenue generated from the expanded Ozwater program will be reinvested in member services.

This is an important step for AWA and further reinforces the dynamic nature of our industry and individual careers. I look forward to the ongoing involvement of members and to successful meeting in Melbourne in March 2008 then subsequent years in Brisbane, Adelaide, Sydney and fo r many years to come.

OZWATER - The story to date...

• June 1962 Inaugural meeting of Australian Water and Wastewater Association - initial membership 364

A consequence of this initiative is that the 12 year cycle that takes Ozwater around Australia will be accelerated. The plan is to hold the Conference and Exhibition in the even years in the major east coast centres (Brisbane 2010, Sydney 20 12, Melbourne 2014). T his will ensure that adequate exhibition and conference space is available and facili rates attendance fo r local and international delegates. In the odd years (starting from 2011 in Adelaide) it will progress around other centres. The location will be selected based upon local support, relevance of conference themes, available facilities and the need to provide a national service. This model will offer more regular major events in local cities and the opportunities fo r centres that have not previously hosted Ozwater to participate.

For example Darwin, with its new convention centre and proximity to Asia, could provide an original theme and context for Ozwater.

• September 1964 First Federal Convention held in Canberra delegates

This change has been possible because of the expanding skill, expertise and experience of AWA' s in-house events

• March 2009 Ozwater Melbourne

Journal of the Australian Water Association

• 1966 second convention delegates

64

203

• 1976 IAWPR international convention held in Sydney - no AWWA convention held • 1977 Federal Convention held in Canberra which was then to be held in odd numbered years • 1993 Formal Exhibition held for the fi rst time in association with the 14th Convention • 1999 17th Convention and Exhibition held in Adelaide and AWWA becomes AWA • 2007 Largest ever Ozwater event • 2008 AWA decide to hold Ozwater an nually

David Barnes

AUSTRALIA'S CLIMATE CHANGE POLICIES Senator Penny Wong Pen ny Wong was elected as a Labor Senator for South Australia in November 2001 and began her term in July 2002. In December 2007 she was appointed to the Federal Cabinet in the new Rudd Labor Governme nt as the M inister for Climate Change and Water.

Before entering parliament, Pen ny Wong was a barrister and solicitor in Adelaide and a lso worked as a n adviser to the Carr Government in New South W a les.

The followi ng is an edited extract from Senator Wong ' s speech to an Australian Industry Group Lunc heon in Melbourne on 6 February 2008: Over many years, scientists have gathered a body of evidence which makes the case that climate change is real and is being caused by human activity. For some ti me now, that evidence has been irrefutable. It is no embellishment to say that cl imate change is the challenge o f our generatio n. Now it's up to us. Future generations will look back o n us all and ask what we d id.

T he Government's climate change policy is built o n three pillars: (i) reduci ng Australia's greenhouse gas em issions; (ii) adapting to climate change that we can' t avoid; and (iii) help ing to shape a global solution.

T he fi rst pillar - reducing Australia's greenhouse gas red uctio ns - is marked by our existing commitment to a target of reducing em issions by 60 per cent of 200 0 levels by 2050. T he Government is also com mitted to setting a medium term target.

The Government will d eliver measures to reduce emissions at least cost, and with

I a~ united ~ Utilities

already u nderway, so we must p repare ourselves for the inevitable changes alread y built into the climate system. T his will involve far reaching impacts on o u r economy, h uman amenity and o u r enviro nment.

Senator Penny Wong

greatest poten tial to drive n ew growth, create jobs and develop new ind ustries.

That's why at the heart of our efforts to red uce emissions will be a system of emissions trading.

T he second pillar is adapting to climate change that we cannot avoid. C limate ch ange resulting fro m human influences is

The third p illar is helping to sh ap e a global solution to rhis global problem. Ratifying the Kyoto Protocol has meant that fo r the fi rst ri me Australia is a fu ll negotiati ng partner in all key international forums.

This will be a hard and long road. Any posr-20 12 approach needs to secure wid espread agreement of countries with d iverse interests and entrench ed posit ions.

Nevertheless, we are commi tted to working through these multilateral negotiations toward an effective global agreement. In working toward that agreement, we understand that the developed world has to lead. Our policies to make substantial cuts to domestic emissio ns underline t hat leadersh ip.

Acciona Agua, a world leader in seawater desalination, with more than 70 major facilities globally, and United Utilities, which provides safe clean water supplies to more than 20 million people globally, including in four Australian states, have formed a joint venture in Australia .

The joint venture will be a leading contender for the large scale desalination plants currently under consideration nation-wide.

Through Acciona Energy, which in Victoria is currently constructing the largest wind farm in the Southern Hemisphere, the joint venture will also be able to offer green electricity.

Further information: Mark Griscti +61 8 8408 6500

Reducing Greenhouse Emissions The best way co drive reductions is co use market based mechanisms. I t is no t enough co simply sec targets co reduce emissions, and hope for che best. Nor should we be just imposing action on those industries and companies that are carbon intensive.

In putting emissions trading ac the heart of our efforcs co reduce greenhouse, we will place a limit - or a "cap" - on the emissions we will allow co be produced. Permits would chen be issued up co the level of the cap and each year firms would surrender co the Government a number of permits equal co their emissio ns. This will pcoduce a market for permits, which will be actively traded and will attract a price. le is chis price - "the cost of carbon" - chat will change the way chat decisions are made throughout che economy. Companies chat can easily reduce emissions will do so co avoid this cost, thereby freeing up permits for chose companies who have fewer opportunities co reduce their em issions.

A system co enable the trading of emissions will help promote private secto r innovation and it will help address the market failure chat has contributed so profoundly co climate change.

entities co trade permits, thereby putting a price on carbon. There are several reasons for chis approach - one of che most important being international consistency.

• The quancicacive limits on emissions chat define its effectiveness and environmental contribution will be designed co place Australia on a low emission path in a way chat best manages the economic impacts of transition, while assuring our ongoi ng economic prosperity.

• The scheme will have maximal coverage of greenhouse gases and sectors, co the extent chat chis is p ractical. There is wide agreement chat over 70 per cent of our national emissions can be practically covered by emissions trading and we will proceed cowards scheme design o n this basis.

• The system will be designed co enab le in ternational linkages, while ensuring it suits Australian economic conditions. The racificacion of che Kyoto Protocol opens the door co a range of carbon trading opportunities for Australian businesses, and links us co che mulcibillion dollar trading market chat already exists internationally.

Design Principles for Emissions Trading

• The d esign will address che competitive challenges facing emission-intensive trade exposed industries in Australia. The introduction of a carbo n price ahead of effective international action can lead co perverse incentives for such industries co relocate or source production offshore. There is no poin t in imposing a carbon price domestically which results in emissions and production transferring internationally for no environmental gain.

• It will be a 'cap and trade' scheme. That is, it will sec an overall environmental cap by issuing a sec number of permits, and allow

• The scheme will also address che impact on strongly affected industries.

The introduction of em issions trading will constitute che most significant economic and structural reform undertaken in Australia since che trade liberalisation of che 1980s. It will create a major new financial market aimed at achieving an environmental obligation.

• Measures will be developed co assist households - particularly low income households - co adjust co the impact of carbon p rices.

Of cou rse, it stands co reason chat in doi ng chis, we will be looking co facilitate changes in ch e behaviour of firms and households.

This is a highly complex piece of policy development, with important implications for our economy and society. Therefore, the Government will cake a careful and deliberate approach co fi nalising the scheme design, drawing on many sources of advice co achieve the best quality policy outcomes and minimising implementation risks.

Specialist environmental solutions Reliable and timely environmental services tailored to your business requirements

ConsultinQ I Monitoring and Technical Services I Analytical

I GIS

Professor Garnauc is co examine che impacts of climate change on the Australian economy. H e will recommend medium co long-term policies co improve the prospects for sustainable prosperity. The Australian Treasury will also make an important contribution co emissions trading design. In particular, it has a team modelling the effects of different emission reduction trajectories o n growth, employment, income and prices across the economy.

Complementary Policies for Emissions Reduction • Under a working group established through the Council of Australian Governments, co work co-operatively with the states and territories co implement a national 20 per cent Renewable Energy Target. The design work will be finalised by September. Legislation will be introduced early next year.

• The Government's Clean Energy Plan will promote che development of renewable energy technologies and clean coal.

• The government will lead on emissions reductions. In May lase year, it announced a number of specific actions chat it would cake in office co reduce i cs own emissions.

We need a coherent approach chat reduces greenhouse emissions, chat deals with che climate change chat is already happening, and that seeks a global solution. This challenge is going to require the efforcs of all of us.

42

MARCH 2008

Water

Journal of the Australian Water Association

OPERATING IN A CARBON CONSTRAINED WORLD - THE CHALLENGES AND OPPORTUNITIES

By Ross Young, Executive D irector, Water Services Association Australia Climate change and its impact on water resources is one of the greatest challenges of our time. T he urban water industry has been intimately involved for sometime now in measuring and reporting publicly on its greenhouse gas emissions . The industry has every reason to be concerned about climate change and to be taking a leadership role in the implementation of both adaptation and mitigation strategies for the following reasons: • The water industry is the fi rst industry to be impacted by climate change as surface runoff that fills our water storages is the

first casualty of changing and more variable rain fall patterns

higher levels of treatment being required which increases energy consumption

• The industry is a large user of energy, particularly electricity wh ich is used fo r the pump ing and treatmen t of both water and wastewater

• Water industry customers and the com munity expect water to b e managed in a sustainable manner.

There are four categories of emissions which can be contributed to the urban water industry:

• T he new non-traditional sources of water being developed for our cities such as desalination and recycled water are far more energy intensive than traditional water supply systems

1. Carbon dioxide - primarily due to energy consumption (e.g. pumping, treatment, recycling, desalination etc)

• More stringent environmen tal standards for discharges of wastewater results in

2. Meth ane - by-product o f wastewater transport and treatment

CAPS. Your total air solutions provider for water projects.

For 25 years CAPS Australia has been supplying a wide range of oil -free, low and high pressure air blowers for conveying, flotation cells, water treatment, leach tank aeration, bacterial oxidation, and other water projects. Our expertise in delivering reliable, quality and cost-effective air blowers is unsurpassed. You'll also be impressed with our can-do attitude and capability to manufacture customised solutions.

Whether you need to buy or rent air blowers, CAPS makes it easy.

With branches in every major city and many regional areas, service and support is only a phone call away. Contact us on 1300 858 763 or email feedbackccucapsaust.com.au

CD11/JUEST

AIRMAN..

ICDtl.Ell.POt-/R S't'STE/v\S

I DGARDNER DENVER

JIOBUSCHI

(fjj)1ngersoll Rand lndusttM/TkhnologJti

www.capsaustralia.com.au

44 MARCH 2008

Water

Journal of the Australian Water Association

@ t:111'S

Australia

Pfyltd Compressed Air and Power Solutions

3. Nitrous Oxide - by-product of wastewater transport and treatment 4. Non-Methane volatile organic compounds - wastewater transport and treatment.

Both adaptation and mitigation strategies are required to meet the challenges posed by climate change. The adaptation strategies being implemented involve: • Diversification of water sources to include rainfall independent sources such as desali nation and recycled water • Water conservation measures which reduce water industry greenhouse gas emissions and household emissions particularly where the measure target reducing hot water such as shower heads and water efficient washing machines.

The urban water industry is very fortunate compared to other industries in that it has the ability to introduce mitigation measures and many of these measures are being actively implemented now. Mitigation measures include: • Setting carbon emission targets and timelines for becoming carbon neutral • Producing renewable energy by installation of mini hydros and collection of biogas at wastewater treatment plants • Purchasing renewable energy • Sequestration through tree planting • Improving energy efficiency of operations such as installation of variable speed pumps • Project option evaluations which include a price on carbon.

The most significant challenge for the ind ustry will be dealing with the issue of fugit ive emissions wh ich include nitrous-oxide and methane largely attributable to wastewater systems and receiving water environments. The reason they are a challenge is that small quantities have large impacts. For instance, one tonne of ni trous-oxide is equivalent to 310 tonnes of carbon dioxide and methane has an equivalent of 21 tonnes of carbon dioxide.

WSM is working wirh rhe Australian Greenho use Office to undertake work which will assist in overcoming the knowledge gaps char currently exist in th is area.

WSM recently commissioned the CSIRO to undertake a thorough audit of the energy/water relationship in the industry and this report will be completed by May 2008.

WSM is soon to release an Occasional Paper, "Energy and G reenhouse Mitigation Strategies" which outlines rhe measures being undertaken in the industry to minimise carbon footprints.

As energy prices invariably increase and with the introduction of a carbon trading scheme, I believe that chis wi ll open up many opportunities for urban water industry to develop innovative ways of creating energy from waste and water including biosolids. It could well be that in decades to come our wastewater treatment systems and water distribution systems are valued equally for their ability to generate green energy as what they are valued for now, in providing vital water and wastewater services. T h is is where the great opportunity arises and we must seize it for the benefit of the industry and the planer.

ACCOUNTING FOR CARBON REPORT OF THE AWA CONFERENCE, FEBRUARY 2008

By E A (Bob) Swinton G lobal warming, and the imminence of carbon permits and trading, are top ics high on the agenda, and a high-level audience of a hundred delegates took advan tage of a program which ranged fro m rhe co mplexity of the legalities, present and future, to down-to-earth methods to reduce global warming impacts by energy effi ciency in the vario us sectors of ou r industry.

David Barnes, President of AWA, reviewed the situation in his opening ad d ress, and pointed out that rhe likely effects of atmospheric carbon dioxide on the Earth's temperature were predicted by a famo us physicist in 1896 (not l 996). Arrhenius, who was awarded a Nobel Prize for a completely different aspect of physicalchemisrry, can be regarded as the 'Father of C limate Change' .

So now, over a 100 years later, rhe heat is on to red uce the emission of such gases.

How does this affect the water industry?

We currently account fo r only 1 to 2% o f the total CO2 emitted by Australia bur as ocher sectors reduce their emissions fas ter than we do, that relative percentage will rise and we will face the full fo rce of carbon acco unting. More ominously, as mentio ned by o ther speakers, and summarised in the paper by Foley and Lant (th is issue) wastewater treatment to reduce ni trogen in effluents to low levels, using the BNR techniques of nitrification-denitrification, results in increasi ng quan tities of nitrous oxide, N 20 , being d ischarged to the atmosphere. Since th is gas has been assessed as over 300 times more damaging than CO2, even small quantities are goi ng to become items of concern, add to which methane, CH4, formed in anaerobic processes, is some 25-40 times more poten t than CO 2.

As an industry we can measure, report, and economise on the electrical energy used fo r pumping and aeration but he questioned whether such simplistic accoun ting covered all our carbon emissions, q uoting such matters as official personal travel. David concluded chat we will undergo p ressure to reduce our emissio ns, and that will beco me harder as aquatic environmental discharge limits are made increasingly stringent.

The C D containing the PowerPoint presentations and Professor Garnaur's Interim Report is a source of valuable info rmat ion. It is available from the AWA bookshop, bookshop@awa.asn.au.

46

MARCH 2008

Water

Pt,blicslionby 28 Feb

Figure 1. Ti me lines for mandatory reporti ng .

The Legalities An introd uction to rhe new area of carbon accounti ng, emissions trading and rhe legalities which will ensue was given by Ro Coroneos who had helped frame the NSW Greenhouse Gas Abatement Scheme, the first in the world. She quoted from the Interim Repo rt by Prof Garnau t (a copy of which is included in the C D of the Proceedings) and attempted to clarify the jargon which is build ing up around the topic. The concep t of carbon being rhe 'new currency' was initiated by Simmel, in 1978, and as prices per tonne of CO 2 equivalent are established in a trading market, it will come to do minate operations and project plan ning as much as the mighty dollar itself. T he implications for corporations are set to increase in complexity.

The devil is in the derail, and lots of details were provided by Brendan Bateman, of C layton U tz, Tony Hill of Blake Dawson, and Peter Brisbane, Director of the Green house and Energy Reporting Taskforce of the Department o f C limate Change ("AGO").

The Government has committed Australia to reduce its total emissio ns to help safeguard the planer. What actual targets and dates are not yet decided, to some extent dependent on the final Garnaut report, b ut in the long term, percentages such as 60% reduction by 20 50 are being bandied around. T here has been actio n already in some Scates, but there is no doub t that Australia as a whole will be

Journal of the Australian Water Association

co mmitted, and that extends to every person and every activity, even essential services such as ou r ow n.

The National Greenhouse and Energy Reporting System (NGER) Act, 2007 will ensure a robust and transparen t data base, with a si ngle reporting po int to assist Commonwealth, Seate and Territory programs and activities. It will house the d ata which will u nderp in the Australian Emissions Trading Scheme. The latter will be designed by the end of 2008, based o n eco nomic modelling by T reasury, the Garnaut Review, and co nsultation with industry, publ ic and all levels of government.

So will this affect you r corporatio n? It assuredly will, but exactly when depends on how large are you r emissions. NGER will be applied in three tranches, as depicted in Figure 1. The major authorities will obviously be involved from D ay 1.

Reporting will be mandatory. We will be included in Division D, along with electricity, gas and waste, subsection 28.

Peter's PowerPoint p resentation outlined the defi nitions, the details, the regulations, the dates. H ow Permits, Cap and Trade and the carbon market will operate is yet to be fully fl eshed out, but by 2009 , its implications will be biting home, both in obligatory reduction of emissions and u nfortunately, in the inevitable costs.

This, o f course, co mes as no surprise, and all the major autho rities and some of rhe smaller ones have already started to reduce their consumption of electrical energy by

Greenhouse Gas Emissions (Annual Net CO 2 e tonnes per property) Utility

2000-01

2001-02 2002-03 2003-04 2004-05 2005-06

Sydney Water

223

241

241

221

226

208

Water Corp (P)

n.a.

264

240

248

240

242

Yarra ValleyW.

n.a.

41

38

39

40

23

South East W

77

78

74

74

65

55

SA W ater {Adel )

n.a.

643

925

581

573

532

Brisbane Water

n.a.

488

504

497

494

Gold Coast W.

n.a.

424

406

459

..

452

,.,....

.. ..,.

¡ .,. 425.,co ., ,., ., 328,, ,., I

Figure 2. Results of current actions by some o f our major authorities. efficiency drives, are investigating other means ro reduce CO 2 emissions, and learning how to adapt to a changing climate, nor just the weather, bur the eco nomy.

emissions, some of the results being summarised in Figure 2.

His concl usion was char as an energyintensive industry we have a moral and social obligation co act on climate change

issues, and we have been pro-active in mitigation and adaptation. However 'fugi tive emissions' were the 'X Facror'. Yet carbon neutrality was likely to become the norm, and we must plan for carbon trading and a carbon constrained world. In question time he re-iterated David's concern chat the environmental regulators are not yet considering the energy demands of tighter discharge limits for nitrogen co the aquatic environment nor the consequent discharge of some of ir co the atmosphere, not as elemental N, but as the damaging N 20.

Kerry Schott, MD of Sydney Water, backed up by Daniel Cooper, Energy Manager and Phil Woods, Climate Change Manager, summarised Sydney Water's strategy in face of risi ng demand. Ir was noted char production of renewable energy is a significant contribution, from use of biogas, installation of mini-hydros instead of pressure reduction valves (including one at the bottom of the drop-shaft at North Head STP), and co-generation sets where possible. Estimates fo r such 'green energy' coral 6 GWh/a by 20 10, about 12% of current demand.

Karen Rouse and Tim Kelly outlined SA Water's similar strategy in co nsiderable

Russell Beatty, MWH, outlined their software "CO 2 Map". Ir consricu res a carbon foo tprint decision support system, facil itating rhe examination of costs and benefits of different abatement strategies.

He claimed it covered the emissions from just about everything, including use of office paper (and presumably, toiler paper as well).

Current Initiatives Ross Young (WSM) summarised rhe actions of the major authorities. An Occasional Paper summarising the workshop held last September will soon be published and the derailed study by CSIRO of energy use by the water industry should be completed in May. He said it was ironic that the fi rst casualty of global warming was reduced run-off, and the water industry will have ro respond by using more energy fo r recycling and desalination. Our major cities are blessed with most of their water gravitating from high level dams so that the energy for supply averages a mere 0.2/kWh/ kL, compared ro 4.3 for seawater desalination. Recycling would use 2.8 for purification but fo r acceptance as a potable supply would involve pumping the recovered water back up co the dams so the coral energy involved would be lirrle different.

He reviewed initiatives being taken since 2002 by major authori ties co reduce

........ o_jo;n(

betu.i.en

AWA

=a:

00

â&#x20AC;˘- --=~

Portner

,,.

.......

Media Portner

Gold Sponsors

G~

_, WATER

~

oimâ&#x20AC;˘ PB -=--- =-- --

Dinner Sponsor

ongin

detail. Their pumping costs are very high, due to reliance on lifting water from the Murray over the ranges, but a hydro at the Hope Valley Terminal will recover 6 GWhr/a. In a complex discussion on the ins-and-outs of carbon accounting, Tim noted that this initiative would not be credited since it is 'just' an internal trade between sites. This was but one example of how the rules and regulations are being developed too rapidly to offer the right incentives for carbon reduction. However, as one speaker said, we have no time to hasten slowly.

Action is by no means limited to the big players. Paul Hackney, NSW Water Sol utions, a government-owned business, described, as a case study, the initiatives of Eurobodalla Shire. The permanent population is 37,000 over 3400 km 2, rising to 140,000 in summer.

Water su pply and wastewater treatment account for 50% of their energy use. Their plans are to focus on efficiency of the pumping stations, control the bubble aerator at a dam, install mini-hydros and use their coastal zone advantage to investigate wind farms and solar collectors.

However, these latter have a pay-back

period of several decades at current electricity prices so are impracticable unless government grants can be obtained.

However, one advantage of a country authority is that carbon offsets through forestry are eminently feasible.

Peta Maddy gave an inspiring presentation on the plans and achievements of another medium sized authority. Western Water, on the outskirts of Melbourne, was a predominantly rural authority but now hosts one of the most rapidly developing suburban zones in Australia. It services some 50,000 properties on 3000 km 2 . I t purchases its water from the bulk supplier, Melbourne Water, because in the past seven years their reservoirs have declined to less than 7% capacity.

Their emissions in 2004/5 were 30,000 tonnes CO 2e, which is taken as their baseline, but by 2006/7 these had been reduced to 24,000. In 2007 the Board set key objectives, 25% reduction by 2008/9: 50% by 2012/13: 75% by 2014/15, and 100% by 2017, i.e. carbon neutrality. T hey obtained $1 OK for an opportunity study from the Victorian Energy and Greenhouse Review and $15K from G reenhouse Strategy Template. Most significantly, they

NOBLE Water· Your water treatment solution providing complete turn key projects for: •

Sewage treatment plants using the latest technology to provide re-use in small local applications or large townships.

•

Potable water treatment plants for surface, bore and seawater,

•

Industrial wastewater treatment and recycling systems that comply

built on site or in a transportable design. with effluent discharge or higher quality requirements for re-use applications.

•

Domestic water treatment, reverse osmosis, softeners and filtration.

In addition, components and small packages that adapt to existing infrastructure, enhancing performance.

Whichever your need; Noblewater Is your solution provider.

Noblewater Industries Pty Ltd

T: 07 3888 5222 Web: www.noblewater.com.au

delivering clean water 48 MARCH 2008

water

Journal of the Australian Water Association

poached a young enthusiastic champion from Melbourne Water, whose down-toearth approach galvanised the staff.

Her challenge was to persuade the sceptical staff that it wasn't 'Greenwash'.

The first actions were an independently facilitated workshop involving all staff from CEO down, and this identified 25 opportunities, equivalent to 6400 tonnes.

Ownership of ideas was established and enthusiasm maintained by monthly meetings and publications to the community. However, to achieve their goal in face of a rapidly expanding customer base a further 15,000 tonnes will be needed. Customers themselves were asked whether they would contribute financially and a survey indicated that $30 per year would be acceptable. Lest it be thought that this was entirely idealistic, the CEO insisted that each project should survive a business case, with multi-criteria assessment, independent assessment by an energy consultant and a 3-year payback.

Immediate actions were that the main office go carbon neutral immediately, their efficiencies enabling purchase of GreenPower, the new Class A recycling

---

plant to be d esigned carbon neutral, biogas to be captured, their top 11 pumps to b e tuned to top effi ciency and the savings to be applied to the 66 small pumps by purch asing Green Power.

What

:

,,¡

tvm this look like in 2017? ~

.

For the fu ture, all future p rojects to be p lanned as carbo n neutral, action to quantify all diffuse emission sources, and abatement opportunities to be investigated as a last resort to close the gap.

H er earnest hope was that if climate change could be halted, maybe their main dam could revere to its previous levels, as in Figure 3 .

Kate McA.uliffe of C ity West Water summarised their approach towards nee zero emissions, and Kein Gan ofYarra Valley Water discussed the vexed questio n of whether replacement of showerhead s, performed initially for water savi ng, could q ualify as a carbon offset. Sabri Mrayed reviewed the work at UNSW on comparing theoretical with actual energy d emand for various types of desalination. The hope for the fu tu re lies in membrane distillation which could use low-grade waste heat; however, the production of suitable robust membranes is some years off.

Fugitive Emissions

Lake Merrimu

Figure 3. Effect of drought on Lake Merrimu. Hope for the future. favo ur N 2 O and NO productio n. USEPA and WERF are also active, aiming to optimise wastewater and solids operations (OWSO).

Ross Young announced that WSM have signed an MOU with WERF so that results

will be made available to Australia, and last year WSM commissioned the University of Q ueensland to perform a li terature su rvey of data and recom mend where fu rther research was justified. Jeff Foley and Paul Lant presented their findings, a s ummary

We are nor alone on the G lobe! Jay Witherspoon, o f CH2MHill , reviewed che actions bei ng taken by California in p reparation for climate change regulation , way ah ead of the rest of USA. The Californian Global Warming Act, 2006, enforces emissions reporting, sets statewide limits (cu rrently I 990 levels) and markerbased co mpliance mechanisms.

T h e STPs have formed a Statewide Response Group to develop a consistent ap proach, and coordinate with government, with currently over 45 members in Cali fornia, Washingto n, Oregon, with possible members from Colorado and Nevada. Actions taken will be 'ban ked' against the fu ture permits. H e said that with regard to fugitive emissions (CH 4, N 2 O, SOx, organ ics) there is no doubt that some of USEPA's default protocols are flawed, with significam data gaps and estimates based on simplistic chemistry, nor reflective of the actual biological processes.

J ay warned char early estimates o f N 2 O emission s were likely to be wrong, and chat emissions from STPs were probably less than 3% of the total. Mose of it is probably emitted from the storage lagoons rather than the actual BNR plane. T he U niversity of Columbia, together with C H 2MHill, have set up a research group to test the hypothesis char denirrificacion under carbon-limited and low DO conditions

ojoint venture belWMn

AWA

00

-â&#x20AC;˘ -i::a:

~~

PoMar

1M Thi ........

Oif"W'\er Spons0<

--PB

of which is published in chis issue, highlighting the data gaps. Paul Rasmussen of United Water, SA, agreed that much of rhe AGO theory was suspect.

He cited rhe comprehensive work being done in UK and France by Veolia. The best approach is to get down and d irty and measure rhe emissions in rhe field. H e is using odour collection hoods and a C hinese rwo-column SCF with FID , ECD and TCD detectors in series. (Since the instrument had never previously been exported, he had fun with translating rhe Manual). He maintained that if we work fast enough we could position Australia as a source of expertise fo r the wider global market, in Asia, even USA.

When all else fails, reduce your emission balance by purchasing offsets and green energy!

David Freudenberger, of G reening Australia, said that in the long term, even when all sources of green energy are fully deployed, there will still be a h uge role for carbon captu re and sequestration , and photosynthesis by forests was legitimate.

Some 20-30% of global elevated CO2 was already due to clearing of tropical forests (and Australian scrub) so why nor reverse the process? It was a low-risk tech nology, I F done well. But buyer beware! For quality, the forest sinks must be real, additional (i.e. beyond business as usual), measurable and legitimate. Australian native vegetation offered such criteria, being resilient and permanent, offering long-term conversion from bare paddocks to a dynamic forest or plantation. It is both a big challenge and an opportunity. Sequestering just 10% pa of Australians' emissions would require 3 tonnes per year per person to be locked into forest, at a cost of about $75 per person, on some 250,000-500,000 ha per year.

Tony Hill, of Blake Dawson, had cautioned that there were plenty of pseudo schemes already being promoted in the market place, and he advised people to buy only accredited offsets, as there was a risk of the Trade Practices Act and ACCC homing in on some of the wilder claims.

Greening Australia aims to raise $100M by December 2008, to plant 30,000 ha of diverse native species, in selected areas, where the landscape is degraded, in southern WA, south-west Victoria, and west of Bundaberg, where there would be nil environmental damage to water supplies.

Offsets

Acromet the better way fori,all your sewage meeds

Georgina Kelly, NSW DPI, reinforced this argument, noting that use of biosolids or other organics will not only increase the growth rate of plantations on degraded soils, such as coal mine overburden, bur will increase the signifi cant amount of carbon locked permanently into rhe soil structure, and this is the subject of current research.

Just as significant is the replacement of synthetic fertilisers, rhe production of which embodies huge amounts of energy.

Degradation of nitrate in the soil releases significant amounts ofN2 O, some 70% of Australia's total.

The financial feasibility of off-setting, by forestation or plantation, is dependent on rhe marker price of carbon: at $IO/tonne CO 2e, carbon trading would be a mere sideline of such tree-farming, bur if it rose to over $30/tonne, it would change the relative value of timber vs carbon and encourage rhe practice enormously.

She made the interesting point that plantation timber, harvested for structural use and other wood products, would, many years later, probably be dumped in landfill, where rhe embodied carbon would remain sequestered.

Save Water and Reduce Filter Maintenance MULTI CYCLONE

Pressure Sewer Systems Over 100,000 Installed Systems

• Centrifugal water filtration • No filter media to clean or replace • No moving parts

MultiFree Sewage Pump • Run Dry Safe

• 3m 3/hr to 30m 3/hr • Easily manifolded for higher flow rates

Let nature do the work!

Oxynaut is the ideal solution for your personal sewage treatment and even for retrofitting multichamber septic tanks.

Ultra-fine aeration with oxygen provides for optimal cleaning.

Biological Cleaning Performance of up to 99% with continuing de-nitrification

Acromet (Aust) Pty Ltd VIC: Ph: (03) 9544 7333 Fax: (03) 9543 6706 NSW: Ph: (02) 9647 2432 Fax: (02) 9647 2599

Visit us at: www.acromet.com.au

50 MARCH 2008

Water

Journal of the Australian Water Association

MultiCyclone is a pre-filtration device that is capable of saving water and reducing filter maintenance.

The MultiCyclone works on the basis of centrifugal water filtration.

Incoming water enters its 16 hydro cyclones tangentially, generating a strong centrifugal effect.

Incoming sediment is spun out to the hydro cyclone's wall and then spirals down to the sediment sump, while the cleansed water spirals upwards.

Only 15 litres of water is discharged to clean the MultiCyclone.

NSW (Head Office)

OLD

VIC/TAS

02 9898 8686

07 3299 9900

03 9764 1211

SA / NT

WA

08 8244 6000 08 9273 1900

On the subject of biofuels, the jury is still out. Although the theory ch at cap turing solar energy by photosynthesis of either oil plants or biomass, then using chem as fuel, is si mplistically carbon neutral; however, the transportation and processing necessary co harvest a widely d ispersed sou rce would be a very consid erable impose, and may even lead co a negative ou tcome. This does not apply if the biofuel is solely the waste product fro m a food crop already co ncentrated at a processing plant.

Green Energy, Efficient Energy Peter Haenke is from O rigin Energy, a leading supplier of Green Energy. T he company is active in assist ing its customers in carbon reduction schemes and carbon efficiency programs. T ogether with its subsidiary, Sun Retail, it constitutes 36% of che current coral marker.

Apart from supplying natural gas rather than coal-fired power, it is investing in renewab le energy systems. le has a $30 M investment in che SLIVER enhanced photovoltaic project, and is leading the consortium co deliver the Adelaide Solar City Project, and is a participant in the Central Victorian Solar Cities project.

David Brockway, C h ief of Energy Technology, CSIRO, rou nded off the conference with a wide-ranging review of present and fu ture systems for supply of electricity with reduced carbon emissions . The cleaner coal technologies, which are critically important for the intermediate term are pulverised fuel boiler with carbon capture and sequestration, I GCC (gasification o f the coal, separating the CO2 , then co a gas turbine), oxyfirin g of pulverised coal, by recycling the flue gas but adding oxygen. All these require sequestration. T hese are not sufficient co secure the longterm, since they can only achieve efficiencies of 35-40%, and they will yield an enormous and growing amount of CO 2 co sequester. Is there any hope of a 65% efficiency) Yes, provided research is successful.

Combined cycle gas turbine, even pulverised coal runn ing a diesel engi ne, and u ltimately, the carbo n fuel cell.

For sequestration , we m ust capture th e CO 2, comp ress it co liquid fo rm, then inject into subterranean strata, e.g. worked out coal seams, depleted oil/gas reservoi rs, deep saline aquifers. All of chis will cost at least twice current prices.

Wi nd farms suffer not only from capital cost and residents' objectio ns, but the curse of incermiccency. This is most serious at high w ind speeds, when a gust will require sudden shut-down co prevent damage.

An expensive balancing option is a super-capacitance battery.

Globally, wind power generates 60 GW and is a mature tech nology.

Geothermal is promising, but the 5 km drill hole is the greatest di fficul ty, and distance of sires with suitable 'hoc rocks' from urban cen tres will add co cost.

Solar again suffers from the distance factor. Solar thermal is being d eveloped with Australia build ing the world's b iggest in Queensland and also in Newcastle. The steam is to be fed to a coal fired station, the challenge being to march the variable solar output with the preferred steady state of a boiler. Photo-volcaics are improving steadily, the ult imate visio n being co replace silicon by a printed polymer. If solar halves its cost and coal doubles its cost, it has a d istinct future.

A factor which muse not be forgo tten for all ventures in che near future is chat the rapid expansion o f C h ina's eco nomy is straining the capacity of the world's engineering, resulting in higher capital coses.

With regard to nuclear power in Australia, he noted that we do not have enough trai ned engi neers to source, never mind build, operate and even regulate a nuclear plane.

Conclusion As with all AWA Specialise Conferences, chis one provided lots of relevant information and in chis case, was particularly thought provoking.

Journal of the Australian Water Association

water

MARCH 2008 51

OBSERVATIONS FROM PRESENTATIONS AT THE AWA ACCOUNTING FOR CARBON CONFERENCE February, 2008

By Diane Wiesner In the lase week of February 2008, Professor Ross Garnaur handed down his interim report and the news was bad.

Coincidentally, AWA's "Conference Accounting for Carbo n in the Water Industry" fell shortly after.

All the science char Professor Garnaur had read pointed co increased rares of global warming and the pace of change would have co be stepped up. His goal was co have Australia lead by example and set rougher standards co pressure other countries to fo llow suit. So: why does the water industry need to be concerned about climate driven, energy related issues more than projected declines in rainfall and changes in its distribution?

First, the water industry, as a high user of energy for driving treatment plants, operating pumping stations and maintaining operations, will be, and is already, looking co achieve efficiencies to current use. Second, the likely structure of the new emissions trading regime will have impacts on the industry. The Government's preferred model is one char does not allow downstream emitters co earn credi ts for reducing their energy use. Instead, the system will be focused on energy suppliers as was clear from presentations by Ro Coroneos, Brendan Bateman, Tony Hill and Peter Brisbane (from the C limate Change Office).

This will mean char higher energy prices are passed on to owners/operacors of water assets without the benefit of earning credits directly for reducing demand. The price signal will drive change bur the

,Jww Hytran WATER HAMMER SIMULATION SOFTWARE 52 MARCH 2008

Water

system will not provide a positive incentive for action.

This is d iffe rent co the current NSW greenhouse gas abatement scheme, for example, where an owner/operacor can earn (and then sell) credits from upgrading their facilities. The NSW scheme will be superseded by the national scheme when it commences. There will be no grandfathering of entitlements.

Third, research has shown that some of the most readily available saving opportunities fo r the industry lie in strategies such as the incorporation of energy recovery systems into desalination plant design, increased capture of biogas from STPs, trapping gas release from sludge lagoons and a more incense focus on identifying industryspecific offsets like chose offered by biosolids.

And the Views of Conference Participants?

Many conference participants seemed co be of the view that accelerating the start up dare for the new emissions trading scheme for Australia from 20 11 to 201 0 would be extremely d ifficult. With September 2009 as the deadline for submission of corporate and facility emissions data, questions were asked about the likelihood that the data could be rigorously analysed and an appropriate carbon price set by rhe 20 10 start up date.

There was also concern about what appears to be no easy path for abatement claims and offset credit approval at the Commonwealth

level promised for the future. It was apparent char applicants seeking approval of abatement schemes were finding a clearer, easier path through state based schemes. However, these lacked common criteria and process pathways and would become obsolete from 2010.

Fugitive emissions particularly chose from nitrous oxide were identified as an unexpected but potentially serious new category of reportable, which few utilities had factored into the carbon strategies rhus far prepared by them. T he increased regulatory pressures driving the downward push for ever lower levels of nutrients in d ischarges has brought with it the spectre of higher production of N,O gas with 296 rimes the CO 2 emission faccor.

On the positive side, it is evident that the water industry is GH G aware and has been quietly working on how co live in a carbon constrained economy. Many utilities have detailed energy efficiency and carbon strategies. Staff are focused on future impacts. No one is expecting an easy ride.

The Abstracts Handbook and CD of Proceedings from the Accounting for Carbon Conference can be obtained from A WA Bookshop, bookshop@awa.asn.au, at $99 including GST The CD al.so contains a paper by Jeff Foley & Paul Lant on fugitive emissions and a copy of the February 2008 Interim Report on Climate Change Responses

by Ross Garnaut.

Simulate pressu re transients and water hammer in pipes, networks, branches and loops including vapo ur cavitation.

Investigate equipment performance such as air chambers, air release va lves, surge tanks etc. View the transients propagate along pipes in real time and press ure variations at selected locations for conditions such as pump trip and va lve closure.

Draw the pipe system and enter equipment data using intuitive Windows graphics.

Journal of the Australian Water Association

H 2000PtyLtd

Tel: (08) 9364 2211 info@accutech2000.com.au www.accutech2000.com.au

SEWER LINING SYSTEMS

~oat Repair Mortars :rowel or spraying wet or dry

formerly Lafarge Aluminates Australia Pty Ltd

Kerneos Australia Pty Ltd Tel: (02) 9498 5288 Fax: (02) 9498 5311 E-mail : greg.tucker@kerneos.com

FUGITIVE GREENHOUSE GAS EMISSIONS FROM WASTEWATER SYSTEMS J Foley, P Lant, P Donlon Wade Sector Emissions

Abstract

279.4

Current Australian practice for estimating greenhouse gas emissions fro m wastewater treatmen t foc uses mainly o n emissio ns associated with energy use (e.g. electricity and fuel consu mpt ion). Q uantify ing fugi tive emissio ns from wastewater systems is an area of uncertainty for the industry, with less developed and less reliable methodologies. This level of uncertainty in the wastewater industry's "carbon profile" is unacceptable in the emerging business environment of carbon pricing, and managerial comm itmen ts to "carbo n neu trality".

Methane and nitrous oxide em issions have much higher global warming potentials than carbon d ioxide. This paper reviews existing knowled ge o n their p rod uction in wastewater systems and d efines their relative importance compared to emissions from energy usage in wastewater systems and emissions fro m other econo mic sectors.

Fu rther areas of research were identified, to to minimise the level of u ncertainty in the ind ustry.

A draft methodology for fugitive greenhouse gas emissions (GHG) from Australian wastewater systems was proposed for consideration by the Department of C limate C hange (formerly known as the Australian Greenhouse Office (AGO)).

This paper is based on a report by Foley, J. and Lant, P., of t he Advanced Water Management Ce n t r e , Un i vers i ty of Queensland , commissioned by the Water Services Association of Australia.

Wastewater

wale Incineration 0.2%

Hcr,dlng 13.3%

29:)

ICO

Waste water Ha nd ling Emissions

0

- CH, - 1,694 Gg CO 2-e (7 4.8%) • N2 O - 572 Gg CO 2-e (25 .3%)

Figure 1. A ustralia's Greenhouse Gas Emissions 2005.

The Relevance of Methane and Nitrous Oxide Emissions Methane and nitrous oxide are two of the sig nificant greenhouse gases listed under the U nited Nations Framework Convention on C limate Change (UNFCCC). The contribution of methane and nitrous oxid e to total net anth ropogenic radiative forcing is 30% and 10 %, respectively (IPCC, 2007). The combined effect of different gases is calculated using G lobal Warming Potentials (GWPs), as shown in Table I fo r d ifferent time horizons.

It is important to note the global warming potency of methane, and nitrous oxide in particular, and the significant influence of time horizon. The I 00 year horizon fo r UNFCCC accou nting is somewhat arbitrary. A 20 year time horizon may be

more appropriate for water businesses concerned with measuring their carbon foo tprint for strategic development purposes.

Current Estimates of Greenhouse Gas Emissions by the Australian Wastewater Industry A summary of the 2005 Australian National G reenhouse Gas Inventory (NGGI) is shown in Figure I. "Wastewater Handling" emissions in the NGGI are d ue to direct methane and nitrous oxide emissions only. Indirect emissions due to energy consumptio n and other activities are captured under the other sectors.

Quantifying fugitive emissions is an area of uncertainty for the industry.

Table 1. Relative and (Absolute) G lobal W orming Potentials.

Gas

GWPs for UNFCCC

GWPs - IPCC 4th Assessment Report (200n 20 yr Horizon 100 yr Horizon

500 yr Horizon

Carbon Dioxide

1 kgCO2-e/kgCO2

l kgCO2-e/kgCO2 (2.47x l 0- 14 W.yr/m 2 per k9CO2)

l kgCO2-e/kgCO2 (8.69x 10·14 W.yr/m 2 per kgCO2)

1 kgCO2-e/kgCO2 (28.6x10· 14 W.yr/m2 per kgCO 2)

Methane

21 kgCO2-e/kgCH4

72 kg kgCO2-e/kgCH4 (1.78xl 0- 12 W.yr/m2 per kgCH4)

25 kgCO2-e/kgCH4 (2. 17x 10·12 W.yr/ m2 per kgCH4)

7.6 kgCO2-e/kgCH4 (2.1 7x 10- 12 W.yr/m2 per kgCH4)

Nitrous Oxide

310 kgCO2-e/kg N2O

289 kgCO2-e/kg N2O (7.14x 10-12 W.yr/m 2 per k9N2O)

298 k9CO2-e/kg N2O (2.59x10· 11 W.yr/ m2 per k9N2O)

153 kgCO2-e/kg N2O (4.38x 10·11 W.yr/m 2 per kgN2O)

62 MARCH 2008

water

Journal of the Australian Water Association

•

Water pipeline solutions Steel, Ductile Iron, ABS and PVC

•

Water and environmental monitoring and assessment

•

Integrated installation commissioning and rehabilitation services

•

Complete valve, control and automation systems

\

• Industrial, commercial and agricultural irrigation systems •

Water pumps and engineered water storage solutions

'tqca I Flow

/ Control

1300 656 609 www.tycoflowcontrol.com

PRODUCTS

•

SYSTEMS

•

SOLUTIONS

•

SERVICES

The "Wastewater Handling" sub-sector represents 0.4% of national greenhouse gas emissions. This does not include indirect emissio ns due to energy consumption. By current methodologies, methane is the dominant G H G from wastewater systems.

However, the AGO also collects emissions data by economic sector and scope, as shown in Figure 2.

The contribution of the "W ater, sewerage and drainage" sector may seem relatively minor (0.5%) in the overall emissions profile. However, it is equivalent to other prominent economic sectors such as "Railway transport", and "Gas production and equipment", and is larger than many other sectors. It is also important to note that direct and indirect emissions in the "Water, sewerage and drainage" sector are equally significant. Therefore, mitigation efforcs should be pursued with equal vigour in both areas.

Methane and Nitrous Oxide Formation in Wastewater Systems

CO,-e Emissions (Gg) In 2005 140,oo:J ~ -- - 120,(XX)

Biological nitrogen removal (BNR) processes co nvert organic nitrogen and ammonium into nitrogen gas, via nitrification and denitrification. Nitrous oxide can be produced either as a byproduct of aerobic nitrification and/or as an intermediate in anoxic denitrification (Figure 3, developed after (Wrage et al. , 200 1)). Nitrous oxide is highly soluble in water and mass transfer theory shows that it does not easily supersaturate. Furthermore, nitrificatio n and denitrification zones will probably be open to the atmosphere (e.g. aeration tanks) and well-mixed by aeration systems or mixers (i.e. high mass transfer).

Therefore, all nitrous oxide generated in a BNR process will likely be rapidly stripped to the atmosphere.

Existing Literature on Emissions in Wastewater Systems The relevant greenhouse gas emission guidelines for Australia are the 2006 IPCC

64 MARCH 2008 Water

• Indirect Emissions from Ener y Consumptbn

100,(XX) 80,(XX) f:l),(XX)

40,(XX) 20,(XX)

0

Figure 2: Austral ia's Greenhouse Gas Emissions 2005 by Economic Sector.

N20 Ammonia monooxygenase

NH3

Methane is produced in wastewater systems by anaerobic metabolism of organic material by microorganisms. The maximum theoretical production of methane from biodegradable COD can be calculated as 0.25 kg CH4 per kg COD removed. The mass transfer kinetics of methane from the liquid phase to the gas phase are also important and will constrain methane capture particularly in low COD-strength systems. Depending on process conditions and reactor configuration, the dissolved methane concentratio n may be many times greater than its equilibrium value (i.e. supersaturated) (Pauss et al, 1990).

a Direct Emissions

..l..

N20

f? ~

Hydroxytamlne oxidoredoctase

NH20H

!

f

Nitrite

~ oxidoreductase

Nitrate reductase

Nitrite reductase

...!.... ..l.. !.. !..

N03"

NO£

! \!

No2•

L;,,;.

~ ~ ~~hway•

Figure 3. Pathways of Nitrificati on and Denitrification.

Guidelines for National Greenhouse Gas Inventories (IPCC, 2006), and the AGO's Australian Methodology for the Estimation of Greenhouse Gas Emissions and Sinks 2005 (AMEGGES) (AGO, 20066). A summary of key emission factors is published in the

AGO Factors and Methods Workbook 2006 (AGO, 2006a). Members of the "Greenhouse Challenge" program are audited against the AGO Factors and Methods Workbook, even though this document only provides basic and incomplete guidance in the sector of Wastewater H andling. There are also significant inconsistencies between the AGO Factors and Methods Workbook and the Australian Methodology for the

Estimation of Greenhouse Gas Emissions and Sinks 2005: Waste. The existing emission factors for the various wastewater processes in Australia are considered in Table 2 for methane, and Table 3 for nitrous oxide.

The maximum production of methane from COD removal in anaerobic systems is fixed by stoichiometry- 0.25 kgCH 4/ kgCOD removed. The key unknowns are: 1) the proportion of methane that remains dissolved in the liquid phase, and hence is not captured as biogas; and 2) the

Journal of the Australian Water Association

proportion of methane that is oxidised in aerob ic surface layers (e.g. facultative lagoons) . Most studies reviewed from the literature fail to consider these losses, and hence make it very difficult to formulate meaningful gas phase emission factors.

Some studies report dissolved methane losses up to 85% for anaerobic treatment of "weak" domestic wastewater (H ardey and Lane, 2006). The process conditions that influence methane supersaturation and stripping rates are largely unquantified for wastewater systems, and particularly the collection and transfer necwork.Table 2 is a summary of the reviewed literature and recommendations. Full details are available in Foley and Lane (2007).

In contrast, the literature on nitrous oxide emissions was very helpfu l in establishing the range of emission factors for various processes. Table 3 is a summary of the reviewed literature and recommendations.

Full details are available in Foley and Lant (2007).

Nitrous oxide emissions from wastewater systems are enhanced by the followi ng key factors: • Limited carbon substrate, which leads to incomplete denitrification;

BOC Limited, Riverside Corporate Park, 10 Julius Avenue, North Ryde, NSW 2113 Australia contact@boc.com

technical features refereed paper

Table 2. Methane Em ission Factors for Wastewater Systems.

Element

AGO or IPCC

Literature

Comment and Recommendations

0.005 - 0.03 tonne CH 4 per ML flow per year in rising mains. (1 study) 0 - 0.0016 tonneCH 4 per ML flow per year in gravity sewers. (1 study)

The available scientific literature is not sufficient.

Further research is necessary

Wastewater Collection and Transfer Sewers and Rising Mains

No specific guidance

Primary and Secondary Treatment Anaerobic Lagoons

0.20 kgCH4/kgCOD removed

0. 15 - 0.16 kgCH4/kgCOD removed, assuming no dissolved methane losses 0. 19 - 0.21 kgCH4/kgCOD removed, assuming 100% saturation dissolved methane Western Treatment Pla nt, Melbourne - Jan 2006Oct 2007

Losses of dissolved methane in the anaerobic effluent are highly uncertain for low strength wastewaters. The available scientific literature is not sufficient.

Further research is necessary

High-Rate Anaerobic Reactors

0.20 kgCH 4/ kgCOD removed

0.12 kgCH 4/ kgCOD removed, exc. dissolved methane losses - ca ptured biogas only (5 studies; 11 dote points) 0.16 kgCH 4/kgCOD removed, inc. dissolved methane losses (1 study; 4 data points)

Lasses of dissolved methane in the anaerobic effluent are highly uncertain for low strength wastewaters. The available scientific literature is not sufficient.

Further research is necessary

Facultative Lagoons

0.05 kgCH4/kgCOD removed (i.e. anaerobic lagoons < 2m depth)

No specific studies available

Losses of dissolved methane in the anaerobic effluent, and methane oxidation in the surface layer are highly uncertain for low strength wastewaters. The available scientific literature not sufficient. Further research is necessary

Pre-Fermenters Rotating Biological Contactors

No specific guidance

No specific stud ies available

The ava ilable scientific literature is not sufficient Further research should be considered depending on the scale of pre-fermentation and RBC facilities.

Biosolids and Biogas Processing Anaerobic Digestion

0.20 kgCH4/kgCOD removed

0.18 - 0.23 kgC H4/kgCOD removed, regardless Dissolved methane losses from digesters are low, due to the high-COD strength of the liquor. The of dissolved methane losses Operating data from Luggage Poi nt WWTP, Oxley available plant data is sufficient.

WWTP (Brisbane) and Beenyup WWTP (Perth)

Sludge Lagoons

0.20 kgCH4/kgCOD removed

Negligible (1 study)

Methane oxidation in the surface layer of the lagoon is possibly higher than methane production. Further research should be considered

Drying Beds

No specific guidance

No specific studies available

The available scientific literature is not sufficient.

Further research should be considered

Composting

0.01 kgCH 4/ dry kg waste

Negligible (3 studies)

The IPCC default emission factor is an appropriate, conservative estimation.

Further research should be considered

Verm i<omposti ng

No specific guidance

Negligible (2 studies)

Methane em issions are neglig ible from welloperated vermicomposting supporting the IPCC and AGO guidelines.

Incineration

4.85 X 10·5 kgCH4/ dry kg waste

No specific studies available

Well-operated incineration facilities should have negligible methane emissions.

Biagas Combustion

5 kgCH4/ TJ (net calorific basis)

57 - 670 kgCH 4/TJ biogas energy content (2 studies)

The available scientific literature does not conclusively confirm the default emission factors.

Further research on specific combustion facilities should be considered

Terrestrial Receiving Environment Inconclusive and conflicting studies (8 studies)

The literature indicates both positive and negative contributions to methane emissions from landfills contai ning sewage sludges/biosolids.

Further research should be considered

Agricultural Land No specific guidance Application of Biosolids

Li mited and inconclusive studies available

Whilst likely to be negligible the literature is inconclusive. Further research should be considered

Stockpiling

No specific studies available

The avai lable scientific literature is not sufficient.

Further research should be considered in this area

Landfill Disposal of Biosolids

First-order decay model; or 0.12 kgCH 4/kgBODs

No specific guidance

• Low dissolved oxygen, which leads to incomplete nitrification and/or inhibition of the nitrous oxide reductase in denitrification; • Nitrite/ free nitrous acid, which inhibits the nitrous oxide reductase; 66 MARCH 2008

Water

• Low pH and presence of hydrogen sulphide, which inhibits the nitrous oxide reducrase.

These conditions suggest that "advanced" nitrogen removal plants are most susceptible to elevated nitrous oxide emissions. T hese

Journal of the Australian Water Association

plants typically operate close to the limit of substrate availability, with low dissolved oxygen concentrations to promote simultaneous nitrification-denitrification (i.e. via the "nitrite pathway", illustrated in Figure 3). Therefore, regulatory pressure towards ever-lower effluent nitrogen

A global leader in Water Treatments, Ciba (Australia) Pty Limited offers advanced techniques, products and equipment for water treatment applications throughout the industrial and municipal sectors. We have the global reach, technical resources and intense customer focus required to customise a water treatment program that meets your requirements, now and into the future.

Ciba ...

The water treatment experts To find out more about how Ciba can improve the quality of life with Water Treatment solutions, contact us at: 6-8 Donaldson Street, Wyong NSW 2259 Tel: 02 4350 3200 Fax: 02 4353 2136 email: warrick.green@ciba.com www.ciba.com

Value beyond chemistry

technical features

Table 3. Summary of Nitrous Oxide Emission Factors for Wastewater Systems.

AGO or IPCC

Element

Literature

Comment and Recommendations

0.01 kgN20-N/kgN influent (11 studies; 45 doto points) High uncertainty

At low pH, low DO, limiting carbon substrate, significant emissions con be expected. The IPCC approach is likely to be grossly inaccurate. The available scientific literature is not sufficient for the many different types of BNR treatment processes employed in Australia. Further research is necessary in this area

Primary and Secondary Treatment Nitrification and Denitrificotion

0.0032 kgN2O/ person .yr

Aerobic Lagoons Rotating Biological Contoctors Focultative Lagoons

No specific guidance No specific studies available

Further research should be considered in these areas, in concert with the recommendations for general nitrification and denitrification processes.

Biosolids and Biogas Processing Aerobic Digestion

No specific guidance No specific studies available

The available scientific literature is not sufficient. Further research is necessary in this area

Drying Beds

No specific guidance No specific studies available

The available scientific literature is not sufficient. Further research should be considered in this area

Composting

0.0006 kgN 2O/ dry kg waste

0.0007 kgN2O/dry kg sludge. (1 study)

Vermi-composting

No specific guidance

0.02 kgN2O-N/kgN applied. The available scientific literature is not sufficient.

(2 studies-not sewage sludges) Further research should be considered in this area

lnci nerotion

0.00099 kg N2O/kg 0.003 kg N2O/kg dry dry biosolids biosolids

Biagas Combustion

0. l kgN2O/TJ (net calorific basis)

The available scientific literature is not sufficient.

Further research should be considered in this area

The available scientific literature is sufficient (8 studies; 11 doto points).

5.4 - 30.3 kgN 2O/TJ biogos energy content (2 studies)

The available scientific literature does not conclusively confirm the default emission factors suggested by the IPCC or AGO. Further research on specific combustion facilities should be considered

Aquatic Receiving Environment Discharge to Rivers

0.0025 kgN 2O-N/ kgN discharged

0.0003 kgN 2O-N/kgN discharged (4 studies; 12 data points) High uncertainty

The magnitude of emissions is highly dependent on local conditions. For water authorities with large riverine discharges of nitrogen-rich effluent, this represents o significant knowledge gap. Further research on specific river environments is necessary in this area

Discharge to Estuaries

0.0025 kgN20-N /kgN discharged

0.0012 kgN 2O-N/ kgN discharged (5 studies; 15 doto points) High uncertainty

The magnitude of emissions is highly dependent on local conditions, For water authorities with large estuarine discharges of nitrogen-rich effluent, this represents a significant knowledge gap. Further research on specific estuary environments is necessary in this area

Discharge to Ocean (< 200m depth)

No specific guidance 0.002 kgN2O-N/kgN discharged (4 studies; 7 data points - not related to effluent discharges)

The magnitude of emissions is highly dependent on local conditions. For water authorities with large oceanic discharges of nitrogen-rich effluent, this represents a significant knowledge gap. Further research on specific marine environments is necessary in this area

Wetland Polishing

No specific guidance

0.001 kgN20-N/kgN discharged (6 studies; 11 data points)

The available international scientific literature is sufficient.

Landfill Disposal of Biosolids

No specific guidance

Negligible emissions (3 studies)

The available literature on well-operated landfill facilities suggests that nitrous oxide emissions ore negligible. This is consistent with the theory of nitrous oxide formation and supports the position adopted in the IPCC and AGO guidelines.

Agricultural Land 0.01 kgN 2O-N/kgN Application of Biosolids applied

0.0 1 kgN 20-N/kgN applied (7 studies; 11 doto points)

The available international scientific literature is sufficient.

Stockpiling

No specific studies available

The available scientific literature is not sufficient.

Further research should be considered in this area

No specific guidance

standards is driving the design of wastewater treatment plants to operate in exactly those conditions which promote greater nitrous oxide emissions.

The results of the literature review are qualitatively summarised in Figure 4 for the various common wastewater system processes in Australia.

Draft Methodology for Fugitive Greenhouse Gas Emissions Based on our review of the existing IPCC and AGO guidelines, and the international scientific literature, the University of 68 MARCH 2008

Water

Queensland (UQ) has drafted a new methodology for the estimation of fugi tive greenhouse gas emissions from Australian wastewater systems. This methodology has been formu lated with the assistance of the WSAA Project Steering Committee and feedback from the WSAA Water Health Environment and Sustainability (WHES) Committee. It has been presented to the Department of Climate Change for their consideration and inclusion in their soonto-be-released Emissions and Energy Methodologies, under the National

Greenhouse and Energy Reporting Act 2007.

Journal af the Australian Water Association

These new methodologies will replace both the AGO Factors & Methods Workbook 2006 and the Australian Methodology for the

Estimation of Greenhouse Gas Emissions and Sinks 2005. The proposed UQ methodology is attached as a separate document in Appendix B of Poley and Lant (2007).

Illustrated in Figure 5 is an example calculation of fugitive methane and nitrous oxide emissions from a large Australian domestic wastewater treatment plant, treating over 130 ML/d. The plant consists of preliminary treatment, primary

technical features

Collection & Transfer

~ ~~~

~

Primary & Secondary Treatment

":::?~~

~ ®@ ,.

R phoal

Se

n

ocess

Tertiary Treatment

®® ec

Blosollds / Biogas Processing

er

D

-ing

Aquatic Receiving Environment Terrestrial Receiving Environment

Processes likely to produce methane, but not nitrous oX>de

0

O

Processes likely to produce nllrous oxide,

Processes likely to produce nitrous oxide.

but not methane

and methane

Processes unlikely to produce nitrous oxide, or methane

Figure 4. Fug itive Greenhouse Gas Emissions in Wastewater Systems. Wastewater Systems Processes and Likelihood of Methane and Nitrous Oxide Production. sedimentation, biological nutrient removal secondary treatment, tertiary stabilisation and fi ltratio n, anaerobic digestion of primary sludge and thickened waste activated sludge, biogas/energy recovery, and sludge drying pans.

By strict application of the current AGO Factors and Methods Workbook (2006) methodology and the Australian Methodology for the Estimation of Greenhouse Gas Emissions and Sinks: Waste (2005)

(AMEGGES) methodology, this WWTP is estimated to be a methane sink, due to significant underestimation of emissions from anaerobic sludge digestion treatment (i.e. first and second colum ns). This is clearly incorrect, even for I 00% biogas capture.

The third column represents the accounting methodology proposed by the University of Q ueensland. In this estimate, it is assumed that 100% of methane generated from

GHG

175,0:0

Emissions (tC~-e/yr) ............................................................................ ............................ .

anaerobic digestion is captured. T he main difference is a large increase in the sludge treatment emissions, due to application of an appropriate emission factor (i.e. 0.20 kgCH 4/kgCOD removed). Under this methodology, the WWTP is a methane source, which is considered correct.

The fourth column includes a best estimate of methane emissions fro m the collection and transfer network. Using a conservative emission factor of 0.005 tCH 4/ML.yr, the emissions profile of the WWTP is increased by 15%, due to methane produced in the anaerobic rising mains and some parts of gravity sewers.

125,CXX)

75,0CfJ

• • ••• ••••••••• • •• •• • •• ••• •• • • ••• •• • • ••• •••••• •• • • •••• •• • • •• •• • •••• ••••••••••• • ••••• • ••••••H•••• • ••• •• • • • • •••••• • •••••• • •• •o

-25,0:0

-75,0:0

+ - - - - - . - - - -----.-----""T"----...------,.--- - ---t

• CH4

AGOF&M Workbook (2Xl6)

AMEGGES (m5)

UQProposd (axJ7)

0 -38,747

17,595 -52,993

20.366 14,755

UQ Proposd + Collection

"Best Case"Low Emission Fac tors

'Worst Case" High Emission Factors

20,386 19,929

3,671 11,629

127.210 48,949

~-

Figure 5. Estimated Annual Greenhouse Gas Emissions from an example Australian WWTP.

70 MARCH 2008

Water

Journal of the Australian Water Association

The fifth and sixth columns represent "best case" and "worst case" scenarios by estimating emissions using "low range" and "high range" emission facto rs, based on IPCC (2006) guidance and the literature survey. For example, nitrous oxide emissions from wastewater treatment increase by 200% when the emission facto r is changed from 0.0 l to 0.03 kgN 2 0-N per kg N removal.

T his graph highlights the high level of uncertainty in the emission factors, and the consequent high level of sensitivity in the final emissions profile, particularly at the high range.

technical features

Conclusions • Q uantifying fugitive emissions from wastewater systems is an area of uncertainty for the industry, with poorly developed and unreliable methodologies. However, even by current estimates, fugitive methane and nitrous oxide emissions from Australian wastewater systems are significant: - T he co ntribution of the "Water, sewerage and drainage" economic sector to the NGGI is comparable to many other prominent industry sectors (e.g. air transport, railway transport, machinery and equipment) . · The magnitude of direct fugitive emissions and indirect emissions from energy consumption are approximately equal.

T herefore, mitigation efforts should be pursued with equal vigour in both areas.

• T he magnitude of dissolved methane losses in low-strength anaerobic systems (inc. collection and transfer networks) is potentially significant, yet the influencing factors are largely unknown.

• T here is a high level of uncertainty in the nitrous oxide emission factors for "advanced" nitrogen removal processes, especially those operating at low dissolved oxygen, limited substrate availability, low pH , and/or high hydrogen sulphide concentrations. Furthermore, regulatory pressure towards ever-lower effluent nitrogen standards is driving the design of wastewater treatment plants to operate in exactly these conditions which promote greater nitrous oxide emissions.

• The current AGO Factors and Methods Workbook provides incomplete guidance and under-estimates emissions, due to misapplication of emission facto rs for sludge treatment, and omission of key system elements (e.g. nitrous oxide, biosolids disposal, sewers and rising mains) . Therefore, the percentage contribution of "Wastewater Handling" to the NGGI is also probably underestimated.

• There are three major k nowledge gaps for WSAA to address in this field: - Methane formation in collection and transfer systems, and the concentration of dissolved methane in all low-strength anaerobic processes;

New Turbidity and Chlorine Meter from H~NN~ 1ic~l tTurbidity ,. er

Hl93414

\ .1-

Features • • • • • • •

Free or Total Chlorine range - 0.00 to 5.00 mg/L (ppm) Turbidity Calibration - 2. 3 or 4 point Chlorine Calibration - 1 point Turbidity Method - Ratio Nephelometric Method (90") Light Source - Tungsten filament lamp, user replaceable Turbidity Range - 0.00 to 9.99; 10.0 to 99.9; 100 to 1000 NTU auto-ranging LOG memory - 200 records, with time AND location

• Serial Interlace - USB 1.1 and RS232

Tel: +61 3 9769 0666 +61 3 9769 0699 Fax: Email: hannains@hannainst.com.au Web: www.hannainst.com.au

72 MARCH 2008

Water

I

· Nitrous oxide emissions from different types of "advanced" biological nitrogen removal processes; and · Nitrous oxide emissions from effluent d ischarges to specific riverine, estuarine and oceanic environments.

• The emissions profile of the wastewater industry is highly sensitive to the uncertainty in emission factors, as defined by the above three knowledge gaps.

• This profile sensitivity must be considered within the context of the emerging issues of carbon pricing and commitments to "carbon neutrality". It must also be recognised that the baseline for the industry's emissions profile will change as knowledge improves.

T herefore, there is an incentive to address knowledge gaps as early as possible.

The Authors Jeff Foley is a PhD student, Paul Lant (paul.lant@uq.edu.au) is Associate Professor, at the Advanced Water Management Centre, University of Queensland; and Peter Donlon, at the time of this repo rt, was Technical Director at the Water Services Association of Australia.

References AGO. (2006a) AGO Factors and M ethods Workbook. Commonwealth of Australia, Canberra.

AGO. (20066) Australian M ethodology for the Estimation of Greenhouse Gas Emissions and Sinks 2005: Waste. Commonwealth of Australia, Canberra.

Foley, J ., and Lant, P. (2007). Fugitive Greenhouse Gas Emissions from Wastewater Systems. Water Services Association of Australia, Melbourne.

Hartley, K. , and Lant, P. (2006) Eliminating non-renewable CO2 emissions from sewage treatment: An anaerobic migrating bed reactor pilot plant study. Biotechnology and Bioengineering 95 (3), 384-398.

IPCC. (2006). Ch.6 - Wastewater T reatment and Discharge. in H. S. Eggleston, L. Buendia, K. Miwa, T. Ngara, and K. Tanabe, editors.

2006 IPCC Guidelines for National Greenhouse Gas Inventories, Prepared by the National Greenhouse Gas Inventories Programme, vol.5 - Waste.

IGES, Japan.

IPCC. (2007). Summary for Policymakers. In: Climate Change 2007 The

Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. in S. Solomon, 0. Qin, M. Manning, Z. Chen, M. Marquis, K. B. Averyt, M. Tignor, and H. L. Miller, editors, Cambridge University Press, Cambridge, UK and New York, USA. Pauss, A., Andre, G., Perrier, M., and Guiot, S. R. (1990) Liquid-to-Gas Mass Transfer in Anaerobic Processes: Inevitable Transfer Limitations of Methane and Hydrogen in the Biomerhanation Process. Applied and Environmental Microbiology 56(6), 1636-1644.

University of Cape Town, City Council of Johannesburg, and National Institute for Water Research of the CSIR. (1984). Theory, Design and Operation ofNutrient Removal Activated Sludge Processes. Water Research Commission, Pretoria.

Wrage, N. , Velthof, G. L., van Beusichem, M. L., and Oenema, 0.

(2001) Role of nitrifier denitrification in the production of nitrous oxide.

Soil Biology and Biochemistry 33(1213), 1723-1732.

WSAA. (2006). Greencount Methodology and User Guidance Manual. Water Services Association of Australia, Melbourne.

H~NN~

instruments

Journal of the Australian Water Association

Suitable for a multitude of sewerage and drainage applications, Vinidex is proud to offer an advanced piping system, which is environmentally sensitive, cost effective and easy to install. Ranging from 150mm to 900mm, the pipes are complemented by a full range of fittings available in all sizes.

P

13 11 69

I

E info@vinidex.com.au

I

W

www.vi n i dex.com.au

Syst ems & So lutions

tecnn1ca1 teatures

ENVIRONMENTAL MONITORING PROGRAM FOR SYDNEY'S DESALINATION PLANT R Kidson, S Trousdale, J Wood, G Allen Abstract T he Environmental Assessment of Sydney's desalination plant predicted that impacts on the marine environment as a result of the plane's operation were likely co be minimal.

The assessment addressed both the intake of raw seawater, and the discharge of treated seawater concentrate as a by-produce of the reverse osmosis production process.

Approval for che construction of Sydney's desalination plane was granted subject co a number of conditions for environmental performance. One of the conditions was the development of an environmental monitoring program capable of validating these environmental performance predictions.

This paper outlines the environmental monitoring program supporting Sydney's desalination plant - the Marine and Esruarine Mon itoring Program (MEMP).

The MEM P combines a rigorous statistical design with a weight-of-evidence approach for testing the null hypothesis of negligible impact of the desalination plan t's operations on the marine environment around Sydney.

Introduction Like several other major Australian cities, Sydney is constructing a desalination plant as part of a broad approach co ensure longterm water supplies in the face of the current drought and che prospect of longerterm climatic shift. While stakeh olders u nderstand the n eed to secure long-term water supplies, there is also a strong interest in ensuring chat any potential environmental impacts of building and operating a desalinat ion plant are minimised and well managed. Under NSW legislation , planning app roval for the desalination project was granted subject co environmental protection conditions. Boch regulators and che community require evidence char chese conditions have been met.

Testing for impacts of intake and concentrate discharge.

74 MARCH

2008

Water

Botany Boy

Intake and outlet construction areas

Figure 1. Location of Sydney's desalination plant.

Background T he Metropolitan Water Plan (MWP, 2004) is the NSW G overnment's long-term plan for water supply. It addresses population growth, che current drought and potential climate shift as well as the need co provide environmental flows. A key element of the MWP was a diversification of Sydney's water supply, co reduce the city's reliance on rai nfall-dependent supplies. This includes a desalination plane, as well as demand management measures and recycled water schemes (recycling over 70 GL per annum by 2015).

Sydney's Desalination Plant Technical Specifications Sydney's desalination plant is presen dy under construction on the Kurnell peninsula, just south of Sydney's CBD (Figure 1). Kurnell was considered the optimal site after evaluation of alternatives during concept planning. Proximity co the marine rather than estuarine coast was an essential criterion, both in terms of access to consistent water quality for the intake, and also in terms of suitability of the receivi ng water for discharge of the seawater concentrate (the by-product of desalinated water) .

Ocher criteria used co select the locat ion included sufficient appropriately zoned land co enable the plane to be scaled up in the

Journal of the Australian Water Association

fu ture, available power supply and proximity to the water delivery system co allow distri bution of large volumes of water.

The plane will have an initial production capacity of250 ML/day of potable water, scalable co 500 ML/day in future if required. Sydney's unrestricted demand is around 1650 ML/day, therefore a 250 ML/day desalination plane is capable of supplying around 15% of che city's daily demand.

The plane will use two-pass Reverse O smosis (RO) membrane technology after pre-treatment filtration. The desalinated water then undergoes standard treatment with chlorami ne, fluoride, lime and carbon dioxide, co comply with che Australian Drinking Water Guidelines (2004). The resultant potable water will be introduced directly into che existing water distributio n network. A subsurface pipeline constructed beneath Botany Bay will connect the desalinatio n plane co the main local distribution point, located in the suburb of Erskinevi lle.

The plant will p rocess 625 ML/day of seawater via an offshore intake. As a byproduct of the reverse osmosis treatment process at the plant, some 375 ML/day of elevated-salinity seawater will be produced.

In addition co the back-wash water from the pre-treatment fi ltration, there will be some waste produced from the cleaning of the

technical features

RO membranes and this combined byproduct is referred to as seawater concentrate.

This seawater concentrate is released back to the marine environ ment via outlet diffusers. The intake structure and che outlet diffusers are to be located 300-400 m offshore in the Tasman Sea, in a water depth of 20-25 m. T he outlet diffusers will be located 500-1,000 m south of the intake point.

In preparation for the construction of a seawater reverse osmosis desalination plant at Kurnell, Sydney Water undertook pilot plant testing to optimise che design of the seawater pre-treatment process. The Pilot Plane Testing Program was required to optimise a generic conservative design to the most appropriate and economic pretreatment process for seawater off the Kurnell peninsula (Port et al, 2008, chis issue).

The pre-treatment stage of the program commenced in August 2006 and concluded in July 2007. The custody of the plant was then transferred to the desalination plane construction contractor, Blue Water Joint Venture which took control of the site in

76 MARCH 2008

water

August 2007. Construction work commenced in September 2007.

intake pipe prevent large marine animals from accessing the pipe.

T he potential sources of impact of the desalination plant's operations to che marine environment are the intake of seawater and disposal of the seawater concentrate. The strategy to mitigate these potential impacts is twofold:

- Disposal of seawater concentrate: selection of a d ischarge outlet type which maximises dilution and dispersion of the seawater concentrate. D iffuse rs are recognised as a suitable technology to facilitate dilution and dispersion of a discharge scream. Diffusers are also a 'proven' technology - Sydney's Deep Ocean Outfalls have used diffusers for the lase 16 years for the d ischarge of treated sewage effluent. The three D eep Ocean Outfalls are located 2-4 km off the Sydney coastline. Ongoing extensive monitoring of marine sediments and benthic macrofauna have detected no negative effects attributable to che Deep O cean Outfalls, indicating char d iffuse rs are a highly effective method for rapid m ixing of d ischarge screams in a marine setting.

â&#x20AC;˘ Engineering Design Solution:

â&#x20AC;˘ Environmental Solution:

- Intake of seawater: selection of an intake structure which minimises entrainment of marine life into the pipe. The intake structure is fixed to the seabed floor and contains a velocity cap, which reduces the speed at which the seawater flows inco che pipe. The presence of struts around the

- Intake of seawater: selection of a location away from swimming beaches and sensitive marine areas and with strong ambient currents around the intake structure. Off the coast of Kurnell the natural velocity of the water due to the strong currents is greater then the fl ow of seawater into the

Strategies to Mitigate Environmental Impact There are many potential environmental issues associated with building and operating a desalination plane. Einav et. al (2002) provide an excellent review of these issues and mitigation strategies in a generic framework. This paper will d iscuss the potential environmental impact in a specific local context.

Journal of the Australian Water Association

Make sure your lining has our name on it International Protective Coatings supply high performance tank linings tailored to fit your needs

InterlineÂŽ made to measure lining solutions for tanks

4f}r-AKZONOBEL www.tank-coatings.com X , International and Interline

tel: 131 474 www.international-pc.com protectivecoatings@i nternationalpai nt.com

are trademarks of Akzo Nobel. Š Akzo Nobel 2007

technical features

pipe leading to che plane most of the rime.

This reduces rhe potential for entrainment of marine organisms in the intake pipe.

- Disposal of seawater concentrate: selection of receiving waters which are least sensitive to the discharge. In chis case, discharge to che marine environment offshore of Kurnell utilises the strong regional ocean current (the East Australian Current), and highenergy offshore wave action to maximise dilution and dispersion. In contrast, discharge into an estuarine setting does nor harness strong natural mixing conditions to the same exten t. For this reason, estuarine discharge options were considered bur discarded early in the concept planning phase.

These mitigation measures minimise the probability chat an adverse environmental im pact will occur. The role of monitoring is to verify, th rough field evidence, chat this is the case.

The Marine and Estuarine Monitoring Plan (MEMP) In order to verify rhe environmental performance of the desalination plane, Sydney Water Corporation has developed a comprehensive Marine and Estuarine Monitoring Plan (MEMP). The MEMP is one of the statutory conditions for planning approval of the desalination project (Department of Planning, 2006) T he MEMP has the following broad objective:

"To provide a comprehensive and robust program ofbaseline and post-commissioning monitoring of marine and estuarine environmental conditions to confirm quantitatively that there are no adverse effects to the marine and estuarine environment resulting from the operation of the desalination plant".

The focus of th is paper is on the marine components of che MEMP, particularly surrounding the intake and outlet structures. Before the marine components of rhe MEMP are discussed, it is necessary to cover che particular design challenges char che MEMP faced.

Design Challenge 1: Detecting Impacts in 'Open' Systems The classical scientific method specifies rwo seeps for drawing accurate scientific conclusions: 1. Observing (measuring) an impact description; 2. Attributing the correct causal factor(s) to che observed impact - explanation.

O ne of the scientific challenges of a monitoring program in an 'open'

78 MARCH 2008

Water

â&#x20AC;˘ Near Reference

â&#x20AC;˘ Far Reference

(

Figure 2. (Left) Sites for monitoring of the intake structure. (Right) Sites for monitoring of the outlet structure. environment is the inability to control all the variables char may influence the factor being measured. Consider the fo llowing hypothetical scenario:

â&#x20AC;˘ Hypothetical Scenario: Kelp (Ecklonia radiata) distribution offthe coast ofNew South Wales. Professional scuba divers survey a fixed transect of the sea floor off the coast of New South Wales, and accurately map the distribution ofkelp. The divers return in exactly one years' time and repeat the survey over the same area. The results from the fim and second surveys are compared, and indicate that the distribution ofkelp decreased by 30% between the two surveys.

While Step 1 is objective and unambiguous (the kelp d istribution has unequivocally decreased by 30 % - this is purely descriptive), Seep 2 (explanation) can be problematic: what caused the kelp ro decline, and will it recover? Does kelp distribution fluctuate even under natural cond itions? Or was there a specific pollution inciden t off the coast of New South Wales?

O r is a new marine predator responsible?

T he potential explanations for chis observed result are endless and without knowing che 'true' causal factor, it is d ifficu lt to predict whether the kelp distribution will recover. le is also possible char nor one, bur multiple causal factors are acting in combination.

This (typical) hypothetical example illustrates char it is difficulc to categorically identify a cause-effect relationship in an 'open' system like rhe marine environment off rhe coast of New South Wales.

To help overcome this difficulty, fie ld scientists use the BACI (Before-AfterConrrol-Impact) design (Underwood , 1991, 1992, 1994). This design is especially suited to a project that will involve the construction of a new piece of infrastructure, such as Sydney's desalination plane. BACI design is all about sire selection. Before construction of a new structure (or commencement of a new

Journal of the Australian Water Association

activity, such as discharging seawater concen trate) begins, monitoring sires are selected chat are close to the new structure (or activity). These are termed the 'Impact' or 'Proximal ' sites. Additional sires are selected at a distance from che new structure - these are termed che 'Control' or 'Reference' sires. Boch rhe Proximal and the Reference sires are monirored for a period before construction. The sires are then mon itored again after construction. A rwoway statistical comparison is then performed: the 'Before' results are compared against the 'After' results, and the 'Proximal' sires are compared to the 'R eference' sites. If there was a specifi c impact that was exclusively associated with the new structure (or activity) , the BACI comparison is able to detect chis, and differentiate between effects associated with the new structure (or activity), and effects that may be caused by other factors.

The MEMP uses the BACI app roach for the detection of potential impacts of the desalination plant's discharge of seawater concentrate into the marine environment.

Design Challenge 2: Defining Impact Boundaries and Detection Limits In environmental science theory, every human activity has the potential to cause some environmental impact - though chis impact might be infinitesimally small. For chis reason, it is unrealistic to set 'nil environmental impact' as an objective fo r an infrastructure project. Rather, scientists focus on the degree of impact, and whether this impact is significant.

Fortunately, statutory instrumen ts exist under NSW environmental legislation that provide some guiding principles for the defin ition of a significant impact. The desalination project has used the following instruments as references for the design of che MEMP:

\..,

We will be happy to advise you about doing business with the Netherlands, Dutch expertise in Environmental Technology and Dutch government policies.

technical features

• D raft Marine Water Quality Objectives fo r NSW Coastal Waters (EPA 2000) (administered by the NSW Department of Environment and Climate Change (DECC)). These high-level objectives include the protection of:

Table 1. List of parameters analysed in seawater samples. Those parameters only tested for event based sampling shown in bold.

Category

Analyte

Category

Analyte

General

TOC (mgL·l)

Bacteria & Algae

Chlorophyll-a (11gL·1)

DOC

Phoeophytin (11gL·1)

(mgL·1)

- marine ecosystems

BODs (mgL·1)

Total Colilorms (du 100 mL·1)

- aquaculture

UV254 TSS (mgL·1)

Faecal Colilorms (du 100 mL·1)

- recreational opportunities, and

TDS (mgL·1)

- visual amenity.

• Australian and New Zealand Environment Conservation Council (ANZECC) Guidelines (2000) These provide mo re prescriptive methodological detail for the design of monitoring programs.

Anions

In particular, the ANZECC Guidelines recognise a mixing zone fo r point-so urce soluble discharges in to waterways (such as the desalination plant's outlet diffuse rs). A m ixing zone is a spatial area surrounding the discharge point. Within the m ixing zone, limited environment impacts are acceptable; however outside the boundary of the mixing zone, the full managem ent objectives of minimal envi ronmental impact apply. T he environmental outcome is mo re positive if the mixing zone is smaller. (Sydney Water Co rporatio n, 2005) In o rder to define the dimensions of the mixing zone fo r the desalination plant's diffusers, one of the three components of the MEM P is 'M odelling and Oceanographic Processes'. T his component modelled the dispersion of the seawater concentrate after its release from the diffusers.

MEMP Components The central environmental question for the MEMP is: 'What impact, ifany, will the intake ofseawater, and the discharge ofthe seawater concentrate, have on the local sandy bed and rocky reefhabitats?'

T o address this question, three parallel and complementary lines of environmental evidence are being studied in the M EMP: 1. W ater quality characterisation (chemical and toxicity testing):

• of the seawater at the intake • of the seawater concentrate to be released at the outlet 2. M odelling and oceanographic processes, which will study the dispersion of the seawater concentrate after its release through the diffusers 3. Ecological assessments:

80 MARCH 2008

water

E. coli (du 100 mL·1) Organic Pollutants Oil & Grease (mgL·1)

Sulphate (mgL·1) Chloride (mgL·I)

Total cyanide (mgL· 1)

Fluoride (mgL·1)

Total PAH higL·1) THM (11gL" I) MBAS (mgL·1)

Sulphide (11gL·1) Bromide (11gL·1) Cations

Total Potassium (11gL·1)

CTAS (mgL·1) Total Arsenic (11gL· 1) Total Aluminium (11gL·1)

Total Sodium (11gL· 1)

Total Cadmium (11g L· 1)

(11gL·1)

Total Chromium (µgL·1)

Total Calcium (11gL·1) Total Magnesium

Key Concept: Mixing Zones

• reef habitat surveys

Enterococci (du 100 mL-·1)

Total Barium

Total Strontium Other

(11gL·1)

Metals

(11gL·1)

Total Boron (11gL·I)

Total Iron (µgL·1)

Reactive Silica (11gL·1)

Total Manganese (11gL·1)

Alkalinity (Bicarbonate) (mg HCO3 L·1)

Total Molybdenum (11gL·1)

Alkalinity (Carbonate) (mg CO3 L·1)

Total Nickel (µgL·1)

Alkalinity (Total) (mg CoCO3 L· 1) Hardness (Total) (mg CoCO3 L·1) Nutrients

Total Copper (µgL·1)

Total Nitrogen (11gL·1)

Total Lead (11gL· 1) Total Zinc (µgL·I) Total Tin (µgL· 1)

Ammonia (11gL·1)

Filterable Aluminium (µgL·1)

Nitrite Nitrogen (11gL·1)

Filterable Manganese (11gL·1)

Nitrate Nitrogen (11gL·1)

Filterable Iron (µgL·1) Total Mercury (11gL·1)

TKN (11gL·1) Total Phosphorus (11gL·1) Phosphate (11gL·1) Soluble Reactive Phosphorus (11gL·1)

• settlement panels (to monitor non-mobile organisms such as barnacles)

Target Habitat for Monitoring: Rocky Reefs

• fis h surveys

The outlet diffusers for the desalination plant are located on the continental shelf, 300-400 m offshore. The ambient marine environment around the outlet diffusers consists of a large reef shelf with an extensive boulder field (boulders 0.3-2 m diameter) overlaying bedrock. Local relief is 1-3 m, and sand infills have developed in depressions. T his 'rocky reef' habitat supports a diverse ecosystem including kelps, fish, crabs, echinoderms, molluscs and polychaete worms (Plate 1). This subtidal habitat is supported at 20-25 m water depth, which constitutes a further challenge for the MEMP, as manual sampling or surveying at this depth requires scuba divers.

The MEMP will monitor rhe performance of the plant by recording ambient marine conditio ns close to the intake and outlet points of the desalination plant, and comparing these 'Proximal' sites to 'Reference' sites located in similar marine habitat settings, but at a distance from the desalination plant's intake and outlet points (Figure 2). Under these BACI design principles, if the operation of the desalination plant causes negligible effects in the immediate vicinity of the intake and ou tlet points, we would expect the 'Proximal' sites to be statistically indistinguishable from the 'Reference' sites.

Journal of the Australian Water Association

INNOFLOWTECHNOLOG/ES INNOVATIVE WASTEWATER MANAGEMENT

tlvanTexR

Treatment Systems

SUSTAINABLE WASTEWATER MANAGEMENT and WATER RECYCLING SOLUTIONS for COMMUNITIES & COMMERCIAL FACILITIES If you're serious about quality, there is only one choice.

PROVEN FEATURES • Lowest Servicing Required.

• Lowest Power Use.

• Proven track record

(examples below)

- Currumbin Australian Eco Village (110 Lots) - Jack's Point, Queenstown (500 Lots) - Piopio Township (250 Lots) - Centrelink Office Building (Greywater)

• Highest Reliability.

• Specialists in all aspects of wastewater for townships, subdivisions and all domestic applications.

• Highest Quality.

• Grey water and recycle options.

• Full Project Management Services Available.

• Variety of ownership & maintenance models.

• Fully managed turnkey solutions.

• Australasian availability.

Talk to the Proven Experts in the Wastewater Management Field.

INNOFLOWTECHNOLOG/ES INNOVATIVE WASTEWATER MANAGEMENT INNOFLOW AUSTRALIA PTY LTD P.O. Box 263, Ormeau, Queensland 4208, Australia Ph 1800 INNOFLOW, Mob 0432 605 323

INNOFLOW TECHNOLOGIES NZ LTD P.O. BOX 300 572, Albany 0752, Auckland, New Zealand 31 1A Postman Road, Dairy Flat 0794, Auckland, NZ Ph +64 (9) 426 1027, Fax +64 (9) 426 1047

www.innoflow.co.nz • www.innoflow.com.au

technical features

Process

2. Membrane Integrity

1. Filter

4. Reverse

3. Membrane Cleaning

5. Potab ll satlon

Osmosis

: •Sodium hypochlorite

Inputs

! •Sodium hypochlorite !

i • Anti-sealant

1

j • Sulphuric acid ·-------________ :

!

• Sodium blsulfite

'

• Citric acid

1

...

!

,----------·------········,

! · Ferric Chloride ! ! •Polyelectrolyte !

From seawater Intake ~-~-~

I. __ :

,!

•Hydrofluorosillclc acid

:

!

j •Ammonia

l

! ·Lime

!

: •Carbon dioxide

:

·---- i ---...-------·

~:::Soda _______ _!

To potable water distribution (40 - 45%by

volume)

.---------------------······················ By-product Output

!l '

(55 - 60% by volume)

!

•Citrate

•Backwash debris l ____________ _ :

• Bloclde

•Ferric Hydroxide Floes

...........

• Anti-sealant

r· ······· ··· ··········:

• Solids :------+ : ·------···············..,

' - - - - - - ~:

• Supernatant

Neutrali sed components

Landflll

• Sulphuric acid/ Caustic soda

i _•Sodium hypochlorito/Sodium blsulfi•of _,

j

Concentrate treleasad bv diffusers)

'---·-··---·······-····'

Figure 3 . Chemica ls added during treatment a nd their e nd o f process location. sets of samples were comparable, and could thus be pooled together to maximise the total measurement period (in order to capture a fair range of natu ral variability in seawater quality).

Plate 1. Rocky reef sessil e invertebra te fauna on Sydney subtidal reefs (Photograph by Nathan Knoll).

Component 1: Water Quality Characterisation Seawater at the intake The main aim of resting the intake seawater is for public health reasons, as well as to identify the optimal physical and chemical operating conditions and to develop an appropriate chemical dosing regime aimed at minimising any potential toxicological impacts on receiving waters for the full scale plant. For these reasons, the initial water q uality was characterised u sing a p ilot plant. T he pilot plant intake was a horizontally directional-drilled pipe some 300 m offshore. Prior to the pilot plant's commissioning, seawater samples were collected from the intake locatio n between April 200 5 and June 2006 from a boat.

F rom April 2006 until June 2007 water quality for seawater intake was taken from the pilot plant intake itself. A deliberate rime overlap in the two forms of sampling was built into the design to ensure the two

82 MARCH 2008

Water

T o dare, the variability of the seawater samples h as been small.

Seawater concentrate at the outlet diffusers To evaluate the potential of the seawater concen trate to cause an environmental impact, two steps are involved: • C haracterise the co mposition of th e seawater concentrate

• Chloramine

:l

the plant's ou tlet diffusers. Isolating the specific impacts (if any) of the desalination plant from these other potential impacts is a major challenge fo r the statistical design of this monitoring compo nent, as well as the Ecological Assessment component (see subsequent section).

For quality assurance purposes, a number of laboratories were used to analyse fo r key parameters. Table 1 lists the analytes studied. A com bination of routine (once every 18 days) and event-based seawater sam pling was undertaken . 'Even rs' were defined in fo ur categories: • High rainfall (>50 mm in one day); • High wave activity (wave h eight >4 m; wave period >10 s); • Pollution incidents associated with industrial activity in the Botany Bay area (e.g. oil spills, chemical spills); • Algal blooms.

The even t-based sampling explicitly recognises that the full-scale desalination plant will not operate in a 'pristine' marine environment. The close proximity of the desalination plan t to a range of existing industrial operations o n the Kurnell peninsular, Sydney ports areas and the shipping channel means that the desalination plant is not the only human activity potentially contributing to ecological impacts in the marine environment close to

• Undertake toxicity resting of the seawater concen trace.

Composition of the seawater concentrate Table 2 documents the components of the seawater concentrate, and Figure 3 p resents the chemicals added during the treatment and their process end points. It is important to note that these constituents are standard water treatment chemicals routinely used by water supply au thorities to meet high drinking water standards

(Australian Drinking Water Guidelines, 2004), Toxicity testing There are two main approaches to toxicity resting. The first approach is reductionist: the toxic effects of individual chem ical components of a proposed discharge are studied under laboratory co nditio ns. T he advantage of this approach is that it clearly identifies the toxicity of separate chemical agents. T he disadvantage of this approach is that it can be time consuming and costly to characterise the toxicity of a potentially large number of chemicals; this approach also fails to take inro account possible synergistic effects, where the combinatio n of two particular chemicals may be more toxic than the individual chemicals at the same concentration.

Both of these d isadvantages are overcome by the synthesise approach of Whole Effluent Toxicity (WET) resting. WET testing simply rakes samples of the particular streams under study, and determines the toxicity of the sample in its entirety upon target indicator species. In

Table 2. Compone nts of the seawater concentrate to be disc harged from Sydney' s desalination plant.

• Elevoted-solinity seawater • Backwash from pre-treatment filters • Backwash from cleaning of the RO membranes

Journal of the Australian Water Association

Introducing WaterMaster

The perfect balance of power, performance, flexibility and control \

.,... . ¡_ ~

ABB's NEW WaterMaster flowmeters set the standard for water, wastewater, sewage and efflu ent flow measurement and management. Its performance adheres to the most stringent global indust1y standards and is certified to key international approvals.

Whatever and wherever the application, you can specify WaterMaster in complete confidence and with total piece of mjnd.

Proven in the toughest applications, WaterMaster's rugged, robust and buriable sensors eliminate the need of expensive meter chambers.

Thereby, it provides a long, productive and maintenance-free asset life.

See how the benefits stack up ...

WaterMaster Intuitive measurement from ...The Masters of Flow

V : 1300 660 299 : instrumentation@au.abb.com

Power and productivity for a better world TM

jl 1111 11 , ,1,19

technical features

actual operational circumstances, should WET testing return an abnormal result, chis would normally trigger further diagnostic work to identify the constituent responsible for the toxicity. This is termed T oxicity Identifi cation Evaluation (TIE). Therefore, in reality, the synthesise app roach of WET represents the broad-scale surveillance technique, followed-up by the reductionist TIE stage if required.

The MEMP has nominated five target organisms designed to encompass a wide taxonomic range and organism complexity: • Algae (Ecklonia or Hormosfra) • C rusraceae (prawn or amphipod) • Moll usca (mussel or oyster larvae) • Echinodermata (sea urchi n - both fertilis ation and larval development p hases), and • Chordata (fish).

WET testing is conducted at six dilution factors (0, 1:2, 1: 10, 1:20 , 1:40 and 1:80 ). No dilution (Factor 0) corresponds to the worst- case expo sure scenario, immediately after d ischarge from the diffuser. Factor 10 and 20 d ilution are within the near-field m ixing zone, and Factor 40 and 80 are beyond the near-field zone. WET testing will report the fo llowing critical concentrations: • No Observed Effect Concentration (NOEC) • Lowest Observed Effect Concentratio n (LOEC) • 50% Effect Concentration (EC50).

The NOEC is d esigned to detect both lethal and sub-lethal (e.g. developmental d eformi ty) effects. EC50 corresponds to that

POLYMER & PAC MAKE-UP SYSTEMS Integral Control Panel for complete Automated Control

Unique Wetting Cone Assembly

concentration required to achieve a 50% mortality of the species under study.

The WET trial commenced once the pre-treatment configuration had been determined from the pilot plane trials. It is important to note chat organisms involved in the toxicity testing are cultured in the laboratory, and chat this project conforms with the standards of the NSW Animal Ethics Committee.

Component 2: Modelling and Oceanographic Processes This monitoring component involves numerical modelling of the seawater concentrate plume as it disperses into che water column from the diffusers. The subtidal zone in which the diffusers are placed is a high energy, wave-dominated environment. The plume diffusion modelling indicated that after release from the diffusers, full mixing and d ilution of the seawater concentrate was very rapid.

T h is modelling defined a near-field mixing zone of around 50-75 m around the diffusers. Outside of this zone, complete mixing had occurred , and it was not possible to distinguish che seawater concentrate from the background seawater. The small size of the near-field mixing zone means that any potential ecosystem impacts from the discharge are likely to be contained within a very localised area. Once the full-scale plant is operational, field tracer studies will be undertaken to validate the size of the near field zone.

The numerical modelling was complemented by physical modelling in the laboratory. Dye tracer movement through scaled diffuser heads in a flum e was studied in a series of lab experiments intended to optimise the des ign of the diffusers T he flume work is deliberately conservative, as it is performed under static conditions (i. e. background oceanic currents are not replicated in the flum e).

Therefore, the diffusion races measured in the fl ume are che slowest, 'worse-case' scenario - the natural mixing processes in the real field setting (superimposing wave action on o cean currents) are likely to be far more rapid. This means that the mixing zone dimensions in the real field setting are likely to be even smaller than the d imensions indicated by the flum e work. Building th is conservatism into the design is a good example of the application of the precautionary principle.

T he results of the physical modelling supported the conclusions of the numerical modelling - namely that mixing rapidly o ccurred, and the near-field mixing zone was small.

Component

3 : Ecological Habitat Assessments

For the followi ng assessments baseline studies will o ccur for around 1-2 years and post co mmissioning stud ies for up to 3 years.

Rocky reef habitat assessment

- , :i -__ Foot Mounted for forklift access

• Concentrations up to .5% • Turnkey System - ready to go.

• Liquid and Powder Polymer System.

• Stainless Steel / Polypropylene Tank options.

Settlement panels

'

Australia's most comprehensive range of metering pumps Visit us at: www.acromet.com.au

Acromet (Aust) Pty Ltd

VIC: Ph: (03) 9544 7333 Fax: (03) 9543 6706 NSW: Ph: (02) 9647 2432 Fax: (02) 9647 2599

84 MARCH 2008

Water

In this monitoring component, professional scuba divers swim along transects at che 'P roximal' sites and 'Reference' sites, and collect video footage of the transect. T h is video footage is then analysed by expert marine biologists. T he marine biologists are seeking evid ence of change (for example, in the d istribution, diversity and abundance of fish and bottom-dwelling organisms) at the ' Proximal' sites that may be attributable to the o peration of the desalination plan e. The scuba d ivers will collect th is field evidence each year: twice in summer, and twice in win ter (Table 3).

... tht1 btlttt1r way

Journal of the Australian Water Association

Marine organisms can be divided into mobile types (such as fish, which are capable of swimming away from adverse water quality cond itions) and sessile types (such as barnacles), wh ich can not move away from adverse water quality conditions. We would expect sessile organisms in the vicini cy of the desalination plane's d iffusers to be most sensitive to the operatio n of the plant. If any impact occurs, we would expect to see it first in the recruitment of barnacles close to the diffusers.

T: +61 1300 364 946 F: +61 1300 668 763 n:,. ' ,,. Export Awards E: info@ecotech.com .au W: www.ecotech.com.au For all your monitoring applications, large or small.

technical features

T o determine if there are any impacts on the recruitment of sessile organ isms, Sydney Water has deployed arrays of settlement panels in th e vicinity of the plane's diffusers (Table 3). Setclement panels are metal trays that are held in suspension close to the seafloor by anch ored sub-su rface buoys (P late 2 top) . Barnacles reproduce by releasing free-swim ming larvae, which travel some d istance from th eir spawning site, and then seccle o n a suitable rocky subst rate.

The metal secclemenc panels form an artificial substrate for the barnacle larvae to colonise (Place 2 middle and bottom). For the setclemenc panel component, marine biologists are using a BACI approach to measu re any impacts. If the desalination plant's operation is not im pacting b arnacle recrui cmenc, chen the race of setclement o n panels deployed at the 'Proximal' sires should b e no different to che race of secclement on panels deployed far away at the 'Reference' sires.

Table 3. Monitoring program.

Baseline · 2 years (2007/ 08 · 2008/09) Reef ha bitat surveys Water Quality

Settlement panels

Duration Autumn 1 day

Twice

Twice

Twice

Twice

Once

Once

Deployed far 3 months O nce Half day

Water

Twice

Twice

Twice

Twice

1 day

Twice

Twice

Settlement panels

Deployed for 3 months

Once

Once

Toxicity Testing (Final Plant)

Once

Verification Tracer studies ta confi rm extent of the mixing zone under varying environmental conditions

Discharge Fish Survey

Half day

For Sydney's desalination plant, che design for the intake strucrure will use an o pen intake shaft at the seaward end, com plemented by screens lo cated internally within the plane.

86 MARCH 2008

Twice

Water Quality Reef habitat surveys

A poten tial concern with the seawater intake structure is the risk of en trainment of fish due to the water flow into the structure. T h is was a recogn ised design issue for intake structures in some of the early-model desalination planes constructed elsewhere around the world. I ncreased focus o n improving this design aspect h as yielded a n umber of innovative design o ptions in che latest desalination projects which largely address this issue.

mooring (Photograph cou rtesy of M clennans Diving Services). (Middle) Each panel hos two d ownward facing recruitment plates (roughened perspex) that will be sampled (Photog raph courtesy of Clive Holden, Oceanog raphic Field Services). A recruitment plate (bottom) covered by a n abundant and diverse array of sessile a nimals. (Photograph by Nathan Knott)

Twice

Post Commissioning · 3 years (2010 · 201 3)

Fish survey

Plate 2. (top) A n a rray of 3 panels on a

Summer

Performed when deployi ng settlement panels

Toxicity Testing (Pilot Plant) Fi sh Survey

Frequency Winter Spring

Reducing the velocity of water entering the open intake was the key to solving che design challenge of fish entrainment. This is achieved in Sydney's desalination intake st ructure by a wide velocity cap chat mounts the vertical intake shaft (Figure 4). T he expansive surface area of the cap relative to the shaft's cross sectional area effectively reduces water velocity to < 0 .1 ms· 1 over the lip of che cap. T his entry velocity is slower th an most fish swimming speeds, and is negligible compared to the ambient ocean current velocity and local wave action (as confirmed by the Oceanographic scudy, and direct current metering ac the intake location). Raw seawater effectively 'falls' into the intake, rather than being actively pumped; there is no suction effect with this design.

Journal of the Australian Water Association

Twice

Twice

Twice

Twice

The velocity cap consists of a concrete disk su pported by steel b ars. The spacing of che bars would be such chat small fish could gain en try to che area around the end of the p ipeline and cou ld be at risk of entrainment, however the low velocity of water into the intake p ipe means che small fish would not be pulled in.

To validate the effectiveness of chis design in minim ising d istu rbance to local fish assemb lages, video cameras are being deployed in che vicinity of the intake scruccure to record the nu mber and abundance of different fish species (T able 3). T hese data can be compared to che data record ed at Reference sites and again the BACI approach will be used to measure effects. If the desalination plane's intake structure has a neutral effect on local fish communities (the n ull hypothesis), then no d ifference is expected (before and aft er

Sea loYel

Figure 4. Schematic of intake structu re.

Concept design does not necessarily reflect the final desi gn.

·\

www.mcconnelldowell.com \

CIVIL • ELECTRICAL •

MARINE • MECHANICAL • PIPELINES & HOD • TUNNELLING & UNDERGROUND

Head Office: Tally Ho Business Park, 16 Lakeside Drive, Burwood East, Vic 3151 , AUSTRALIA Phone: 03 8805 5200 Fax: 03 8805 5376 Email: macdow_corp@macdow.com.au Sydney 02 8440 6300 • Perth 08 9315 4440 • Brisbane 07 3219 0456 • Adelaide 08 8346 8796

•

DOWELL

technical teatures

commissioning) in the number and abundance of different fish species at sites close to the intake structure, relative to Reference sites located further away.

Natural and Other Disturbances Because of seasonal effects, scientists will need to collect data for these components for several years before we can tell the difference between a desalination-related effect (if any occurs), and other background impacts. Because these subtidal marine habitats are high-energy environments (with powerful wave action and strong ocean currents), the biota that live in these habitats are adapted to continuo us disturbances (e.g. sea swells during major

storm events) that can re-wo rk marine sediments and strip marine plants and animals from the rocky reefs. Sydney beachgoers know that after a big storm, brown kelp is often washed up on Sydney's beaches - this kelp has been stripped from the rocky reefs that we are studying in the MEMP. We can't predict exactly when and how many of these natural disturbance events will occur over the course of a year, but the MEMP studies will measure impacts caused by all sources (both natural events, and human activities). For example, it would be a flawed approach to take fiel d measurements after a big storm, note an absence of kelp on the rocky reefs under study, and conclude from this that the

l\c.unhtk '.(echnology, Pty Ltd B.L. Camtek Technology Pty. Ltd. I~ a leading Australian engineering company committed to the development and implementation of a¢tivated-carbon filters. This ~Id is limitless in its applications.

Some areas covered include large-scale removal of noxious gases, purification of chemicallycontaminated water, air purification for use in reactions, the filtration of gases from reaction vessels and sewage plants.

The unique systems and equipment that have been designed and developed by the company, (and are continuing to be designed), are effective, efficient, and environmentally friendly.

desalination plant's operations were responsible for the absence of the kelp. For this reason , it's important that we are able to distinguish a true impact signal (if one exists) from this background noise.

Weight-of-Evidence Approach The three monitoring components are not designed to be stand-alone: they are designed to synergise with each other. One of the tools used to establish cause-effect relationships (if any are present) in complex 'open' systems is to use multiple lines of evidence - or a 'weight-of-evidence' approach. For example, if a sudden reductio n in the abundance of a crustacean in the vicinity of the outlet di ffusers had been observed (Compo nent 3) the water q uality data fo r that period would be examined (Co mponent 1) ro see if there was any obvious association between the water quality variables measured and the particular crustacean.

Use of the weight-of-evidence approach increases the explanatory power of the MEMP, and the level of confidence in the outcomes measured from the separate components. The weight-of-evidence approach recognises that ecosystem response variables are linked to each other if a change in the biota is associated with a water quality issue, we would expect to see a change in the water quality as well.

Stakeholder Involvement

Two-stage Caustic Scrubber, followed by Activated Carbon Adsorber Capacity Is 12 600 litres/sec This application Is for odour control at sewage treatment plant

The MEMP was not developed in isolation.

Recognising the multi-disciplinary nature of the MEMP , a range of marine scientists from Sydney's universities were engaged to assist in the design of the MEMP, to ensure that maximum statistical rigour can be achieved. A comprehensive consultation process with relevant government agencies (including the NSW Department of Primary Industries and Department of Environment and Climate Change) ensured the involvement and constructive engagement of stakeholder~ at every stage of the development of the MEMP. Finally, formal scientific review of the MEMP was commissioned from national and international academic experts (including CSIRO, U niversity of Me1bourne and the Danish Hydraulic Institute). The outcome of this extensive consultation process is an endorsed MEMP chat meets stakeholder requirements, and has passed critical scientific peer scrutiny.

Conclusions BLC~Ptylld A.B H. 15 0n156152 P O 8ox.c&e. ~ . NSW2Se0. AUSTRAI..IA. Unit 1fl ~ $ltMt, Ha~ NSW 2567, AUSTRALIA Tel +61 (0)246434633

Fax+ 8 1 (0) 2464846n

EmaH: walburgOcamtek.com.su

88

MARCH 2008

Water

www.camfek.com.su

Journal of the Australian Water Association

In light of the need to enhance the city's water supplies, construction of Sydney's first desalination plant is now underway, with the plant d ue to be commissioned in late

"Cecnn1ca1 rea"Cures

2009. In order to meet community expectations and statutory obligations, Sydney Water has designed a comprehensive and in novative marine monitoring program to track the environmental performance of the desalination plant. The marine component of the MEMP is foc used on detecting possible effects from two activities: • the entry of raw seawater via the intake structure on the sea bed, and • the discharge of the seawater concentrate (the main by-product of the desalination process) .

O ne of the key challenges of detecting impacts (caused by a specific activity, such as the operation of the desalination plant) in a complex, open marine environment is distinguishing the potential impact signal of the desalination plant from other impacts.

In designing the MEMP, a range of tools were employed to maxi mise detection of possible impacts attributable specifically to the desalination plant. These include sire selection according to BACI principles, and a weight-of-evidence approach.

The original Environmental Assessment of Sydney's desalination project predicted minimal environmental impacts of the plant's operation, and initial laboratory studies (toxicity testing) support this. The MEMP is designed to verify these predictions with field evidence, once the fu ll-scale plant is operational.

The Authors Dr Renee Kidson is rhe Program Manager fo r the Waterways and Environmental Science portfolio, emai l Renee.Kidson@ sydneywater.com.au; Susan Trousdale is the Environmental Approval and Interface Manager, Desalination Project; Julie Wood is a P roject Officer in the Waterways and Environmental Science group; and Dr Greg Allen is the Manager of Science and T echnology, all with Sydney Water.

References

0

Reduces CJ!ioEnergizer Desludging Costs !

Reduces Trade waste costs I BioEnergizer is a non-bacterium liquid biostimulant that accelerates the breakdown of sludges in flowi ng and static wastewaters

BioEnergizer provides nutrients and biostimulants to balance and accelerate existing microbial activity.

Lowers BOD/COD levels Reduces Odours Advantages Include: • No heavy machinery on site

Used In:

• Municipal wastewaters

• No dewaterlng required

• Industrial wastewaters

• No sludge stockpiles • No sludge transport costs • No sludge disposal Issues

• Grease traps • Pulp and paper mllls

• Enhanced wastewater biology

• Food processing plants

• Reduced foaming

• Isolated communities

• Dairy fanns and piggerles

Before BioEnergizer Solids virtually filling entire lagoon. Lagoon out of effective service.

Australian Drinking Water Guidelines (2004) ANZECC Guidelines (2000).

Department of Planning (2006) Condition of Approval - Concept Approval for Proposal, Project Approval for Desalination Plant and Project Approval for intake and discharge system website and Statement of Commitments - Preferred Project Report Sydney's Desalination Project. Website (August 2006)

solids beginning to break up, water beginning to flow.

EPA (2000) Proposed Marine Water Qualiry Objectives for NSW Coastal Waters.

New South Wales Government (2004) Metropolitan W ater Plan and revised (2006) Einav R, Harussi K. and Perry I. (2002) T he footprint of the desalination processes on the environment, Desalination 152 : 141-154 Condition of Approval - Concept Approval for Proposal, Project Approval for Desalination Plant and Project Approval for intake and discharge system - Department of Planning website and Statement of Commitments - Preferred Project Report Sydney's Desalination Project (August 2006) Port CL, Roddy SJ (2008): Sydney Desalination Plant: Pre-treatment Pilot Testing. Water, 35 No 2.

After 2 months significant reduction in solids, water now flowing through system.

After 4 months

Sydney Water Corporation (2005) Environmental Assessment of the Concept Plan for Sydney Desalination Project (Nov) Sydney Warer Corporation (2007) Marine and Estuarine Monitoring Program (MEMP).

Underwood, A.J. (1991) Beyond BACI: experimental designs fo r detecting human environmental impacts on remperal variations in natural populations. Australian journal ofMarine and Freshwater Research 42, 569 - 587.

Underwood, A.]. (1992) Beyond BACI: the detection of environmental impacts on populations in the real bur variable world. journal of Experimental Marine Biology and Ecology I 61 , 145-178.

Underwood, A.]. (1994). On beyond BACI: sampling designs th at might reliably detect environmental disturbances. Ecological Applications.

Emili: blo1nerglzer@ozmotech.com.1u

Journal of the Australian Water Association

water

MARCH 2008 89

tecnn1ca1 teatures .fereed paper

SYDNEY DESALINATION PLANT: PRE-TREATMENT PILOT TESTING C L Port, SJ Roddy Introduction

The Pilot Testing Program

Table 1. Elements of the pilot testing

Sydney Water is constructing a desalination plant to supplement Sydney's water supplies during the prolonged drought conditions and to provide for population growth. The plant will be a two-pass reverse osmosis desalination plant with a capacity of 250ML/d, upscaleable to 50 0M L/d.

The aim of the pilot resting program was to determine the most efficient pre-treatment process to produce filtered seawater su itable to be fed to a reverse osmosis plan t. Many membrane manufacturers require the SDI 15 (Silt Density Index-15 mins) of the filtered seawater feed to be below 3.5. T he pilot p rogram was to develop a process to reliably produce filtered seawater with an SDI 15 of below 3.

program.

Sydney Water Corporation has conducted a pilot-testing program to determine the most appropriate process for the RO pretreatment. The Pilot Plant Testing Program was required to optimise a generic conservative design to the most appropriate and economic pre-treatment process for seawater off the Kurnell peninsula.

The pre-treatment pilot testing p rogram was conducted over l O months, commencing in October 2006 and concluding in July 2007. This paper reports how concl usions from the pilot plant rest results directly impacted the design of the plant, leading to significant capital and operating cost savings.

This paper won a Best Paper Award at the Young Water Professionals Conferen ce, Brisbane, February.

The testi ng program was divided into phases focusing on the major process parameters as shown in Table 1.

Testing Phase Process Parameters Tested Coagulation

• Coagulation pH; • Secondary coagula nt (polyDADMAC) type and dose Flocculation

Deciding on the most appropriate and economic pre-treatment process.

• Flocculant type (polyacrylamide) and dose; • Flocculation conditions (G value and detention time)

Filtration

• Filtration rate; • Filter media size, type and configuration;

Results and Discussion The program tested the parameters o f conventio nal granular media filtratio n (GM F) pre-treatment shown in Table 1.

The conclusions from each phase of the pilot testing program form the basis of the optimum pre-treatment process, shown in Figure 2.

• Type of coagulant and dose;

• Two-stage filtration

Results have shown that these treatment co nditions provide an effective pretreatment process for the range of raw seawater qualities experienced (5D13 of 1 1 - 25). Typically, the fi lter ripening time is approximately 4 hours with a filter run rime in excess of 110 hours, depending on raw seawater quality. T he test results obtained from the program provide a good level of confidence that reasonable filter run

Temporary Pilot Plants Compound Cronulla STP effluent outtetto ;Potter Point

Drill launch site ,Caltex :Refinery

Intake Structure

Trenched

outlet pipeline

Pump

below ground

Figure 1. Seawater Intake to Pilot Plan t - Combination HDD Tunnel and trenched pipeline.

90 MARCH 2008

water

Journal of the Australian Water Association

Trojan UV-Oxidation Delivering Purified Recycled Water

The Proven Barrier to Contaminants of Concern In making recycled water safe, water providers use multiple treatment barriers to eliminate microbial and chemical contaminants. In installations worldwide, Trojan UV-oxidation is proven as a final treatment barrier to such contaminants, providing superior microbial disinfection while simultaneously destroying trace chemical contaminants. Using a combination of UV light and hydrogen peroxide, the TrojanUVPhox™ delivers UV disinfection and a proven way to address d ifficult-to-treat contaminants such as N-nitrosod imethylamine (N DMA), other nitrosamines, and a wide variety of pharmaceuticals and personal care products.

Only one technology has demonstrated its ability to cost-effectively remove NDMA and other nitrosamines from water: Trojan UV-oxidation. Only one com pany has the necessary experience in municipal, large-scale, high-efficiency UV-oxidation systems for water recycling: Trojan Technologies.

For cost-effective UV solutions that deliver Purified Recycled Water, Australia is turning to Trojan. www.trojanuv.com

•

-

Tro1an Technolog,es 1s represented by:

AQUATEC - MAXCON PTY LTD A B.N. 45 002 250 482

WATER TREATMENT TECHNOLOGY AND EQUIPMENT

119 Toongarra Road; Leichhardt OLD 4305 Phone: + 61 (7) 3813 7100 Email: ronh@aquatecmaxcon.com.au www.aquatecmaxcon.com.au

..

..·····..

TROJAN uv· WATER CONFIDENCE "

technical features

times will be achieved under poor raw seawater conditions, while providing an efficient operating regime for times of average raw seawater conditions.

pH co rrection pH = 6.5

T he results of the Pilot Testing Program led to the fo llowing conclusions that have had a direct impact on the design of the pre-treatment plant and resulted in a significant reduction in the plant cost: • Second Stage Filtration not required: Throughout the pre-treatment testing program single stage filtration proved to be adequate to achieve the objectives of a 'steady state' SDI 15 of the filtered water of under 3 with a range of raw water quality (raw water SDI 3 up ro 25) . Therefore, second stage filtration does not appear co be required. T he removal of these process units from rhe design has resulted in significant capital cost savings.

Coagulation

Single Stage Filtration

FeCI, = 3mg/L CIBA LT425 = 0.2mg/L

H2 S0,

l

l

Raw seawater

Filtration rate = 1Om/hr

To RO

Media: 1200mm, 1.7mmES Coal, 400mm. 0.65mmES Sand 3 x 100mm gravel

Figure 2. O ptimum pre-treatment process.

• Flocculation not required: Results from flocculation rests did nor show a significant benefit co the process. T he removal of these process units from the design resulted in furt her signi ficant capital cost savings. The dosing of a flocculanr d id nor demonstrate a consistent benefit in fil ter ripening time, 'steady state' SDI or fil ter head loss accumulation. Removal of flocculant dosing provides an additional saving of operational coses.

• Increased Filtration Rate: Tests have shown chat a filtration race of IOm/hr provides a robust filtration process which, even with a raw seawater SDI 3 of 20, can provide filtered water with an SDI 15 of less than 3 in around 4 hours. Therefore, ! Om/hr was determi ned as rhe optimum filtration race. The original conservative design required a first stage filter area 25% greater than that required with a filtration rate of 1Om/hr. T he significant reduction in filter area has resulted in substantial capital cost savings.

Figure 3. Inside GMF pilot plant.

Conclusion Each key parameter of co nventional pretreatment was tested as part of the pilot testing program. T he results enabled the optimisation of a generic conservative des ign to the most appropriate and

economic pre-treatment process for seawater off the Kurnell peninsula.

Conclusions from the pilot plane test results directly impacted the design of the plant, leading co significant capital and operating cost savings.

The Authors Catherine Port is the Program Delivery

Figure 4. Pilot Plant Facility (L-R) - Seawater Feed & Discha rge tanks, Zenon Zeeweed l 000 UF Pilot plant, GMF pilot pla nts and SWRO pi lot plant.

92 MARCH 2008

Water

Journal of the Australia n Water Association

Officer for the Sydney Desalination Project. She is a Chemical Engi neer (UNSW) with six years experience with Sydney Water in water treatment, plant design, asset planning and project management, email: catherine.port@sydneywater.com.au; Steve Roddy is Manager, Water Treatment, Desalination Project, for Sydney Water and also is Manager, Water Treatment Process in Sydney Water's Asset Management Division. Email steve.roddy@sydneywater.com.au

CALL TO DISCUSS YOUR NEEDS TODAY

Uohnsonscreens· AWmtherford Compmy

Australia: (07) 3867 5555 New Zealand: (09) 818 6108 Email: salesaus@ap.weatherford .com www.johnsonscreens.com

Con'tra Shear™

DESALINATION IN AUSTRALIA: A REVIEW G Crisp, EA (Bob) Swinton PART 1. SEAWATER DESALINATION The Need Recent climatic changes and population growth throughout Australia have highligh ted the need for more diverse water sources, chat aren't dependent upon climatic conditions, to be introduced to secure local and regional water supplies.

Australia is the world's d riest inhabited continent and the unpredictable climate means that the Australian population requires five times the water storage than does an equivalent population in the United Kingdom. Although 85 per cent of its people live within 50 km (31 miles) of the coast, the country has only recently begun to consider large-scale seawater desalination.

Western Australia The Perth Seawater Desalination Plant, PSDP, located at Kwinana on the Cockburn Sound, Western Australia, could be regarded as a world-leading model for future sustainable seawater desalination plants. The plant was completed in December 2006 and commissioned in early 2007. At a peak capacity of 144 MLD the A$387 million plant is the largest seawater desalinatio n plant outside the Middle East, and Australia's first large scale desalination facility.

Its capacity will make it the biggest single water source feed ing into the city of Perth, p rovid ing 17 per cent of water needs.

The plant was built by MultiplexDegremont Joint Venture, in all iance with the Water Corporation. It will be operated for 25 years by Degremont in all iance with the Corporation.

An associated 82 MW wind fa rm is injecting over 272 GWh r per year into the grid from which the PSDP will be abstracting 185 GWhr per year making the PSDP the world largest desalination plant using renewable energy. Coupling this energy with the low specific energy consumption achieved from the plants novel design, incorporating isobaric energy (PX) recovery devices from ERI, ensured 94 MARCH 2008

water

chat it is the worlds 'greenest' plant. Taking this into account and considering its small physical footprint, this plant has to be one of the most sustainable water sources in Australia.

Its success has led to Perth considering two more plants.

The Sou thern Seawater Desalinatio n Plant (SS DP 1) for 150 ML/d is currently being contracted. It will be built north of Bunbury. Two Spanish-led consortia Acciona Agua Australia Pty Led/United Utilities Australia Pty Ltd and Technicas Reunidas S.A/Valoriza Agua, S.L have been short listed. The two short-listed proponents have until the end of August 2008 to develop competitive bids and submit chem to the Water Corporation .

Both proponents are expected to establish themselves in Perth and begin discussions with design engineering and co nstruction.

Similar to the Kwinana plant, it will be powered by energy from renewable sources.

A third plant, SSDP 2, is envisaged for 2015.

CITIC Pacific Sino Iron, at Karratha, in the Pilbara, will be one the world's largest mines. To supply both the mine and township it is planning to build a 175 ML/d plant in stages at Cape Preston. The desalination plant is still in the planning stage and it is subj ect to government approvals o n a number of fronts.

Acciona/United Utilities have put forward a proposal to build the Sou thern Ocean SWD plant, for 45 ML/d, near Esperance, on the south coast, around 20 11.

Queensland Desalination is a key component of both the Gold Coast Waterfuture Strategy and the Southeast Queensland Regional Drought Strategy Contingency Supply Plan.

The plant at Tugun, close to the Gold Coast Airport, is well on track. It will produce 125 ML/d and is expected to

A quick review of what's going on around the continent.

Journal of the Australian Water Association

provide water by the end of 2008. A 25 km pipeline wi ll deliver the water to the network.

The plant is being b uilt by Veolia-Joh n Holland Joint Venture, in alliance with Gold Coast Water and the State Governmen t. This joint venture is registered as the Gold Coast Alliance. It will be operated for l O years by Veolia in alliance with the Gold Coast Water and the State Government.

An expansion of 45 ML/day was put forward as an emergency contender should dam levels not reach 40% of capacity in 2008. Due to good rainfall this summer it was likely this limit would be achieved and consequently fu rther development o f this expansion option was put on hold.

New South Wales Sydney Water has completed preliminary studies to build what could be the world's largest RO desalination plant. All ancillary components are sized to provide capacity fo r 500 ML/d i.e. about a third of Sydney's drinking water needs. The site is the Kurnell Peninsula, south of Botany Bay, which already hosts an oil refinery, and environmental studies for the site, the intake and saline outfalls, as well as the 15 km delivery pipeline have been completed, The first RO stage, under construction, will be for 25 0 ML/d. The total cost is estimated at some $2 billion. The Blue Water Joint Venture, consisting of J ohn Holland Group Pry Ltd and Veolia Water, will design and build the plant and Veolia Water will operate and maintain the plant.

Sydney Water will own the plant.

The plant will be powered effectively using renewable energy so there will be no greenhouse gas increase as a result of this plant.

Victoria The State Governmen t has committed itself to building a plant, up to 400 ML/d , to ensure reliable su pplies for southern Victoria. The site being investigated is near Wo nthaggi , G ippsland, with intakes and saline outfalls into Bass Strait, rather than into the Port Phillip or Westernport Bays, or the Surf Coast.

technical features

It wi ll be delivered by the private sector using Victoria's Public Private Partnerships fram ework. Expressions of interest are expected to be sough t in late 2008 and construction o f the p lant is scheduled to commence in 2009 in order to start delivering water by the end of 201 1. The $3.1 billion p roject will include an 85 kilometre pipeline to connect the plant to Melbourne. I t will be capable of provid ing around a th ird of Melbourne's ann ual water supply. The plant is estimated to use about 90 mega watts (MW) of power, which will be purchased fro m renewable energy sources.

Australia's Present and Planned SWRO Plants State

Location

Owner

WA

Kwinana Bun bury ?

Karratha Esperance Adelaide Whyalla ?

WCWA WCWA WCWA CIT Iron ?

SA Water BHP Roxby Mel Wat

WA SA Vic NSW Qld

South Australia A I 50 ML/d plant is planned for Adelaide on a site in the former Mobil refinery site at Port Stanvac .. Contracts are being closed in April fo r independent operation of a pilot plant, which should be installed by the end of July. Tenders will be sough t late 2008-9 from experienced contractors for implementation of the desalination plant.

BHP Bill iton, one of the world's largest resou rces companies produces copper, u ranium, gold and silver from its O lympic Dam mining and processing plant near Roxby Downs, South Australia.

As a result of expansion, additional water resources are required. The preferred water supply involves che followin g infrastructure: • Seawater desalination plant with a capacity of 180 ML/d. Infrastructure includes offshore intake and outfall structures, pre-treatment facilities, desalination plant d evelopment and pose treatment/storage. The preliminary p lant locat ion is adjacent to W hyalla, South Australia.

• Transfer pipeline system capable of transferring potable water to O lympic Dam. The estimated pipeline length is 320 km, with a nominal pipeline diameter in excess of I. Om. • T hree or four potable water booster pumping stat ions with power capacities up to 6 MW. • Construction of additional water storage faci lities in th e region of 1,000 ML (264 MG) . The most likely form of construction

Wonthaggi Kurnell 1 Kurnell 2 Tugun

Planned capacity

144 150 150 175 45 150 180

Syd Wat Syd Wat Gold Coast

will be lined and covered water storage dams .

Thermal Plants Western Australia also boasts two thermal desalination plants for industrial application. These are a 3.6 ML/d MVC plant on the Burrup Peninsula for Burrup Fertilisers Ammonia Plant and a 7.2 ML/d MED plant at Ravensthorpe for BHP Billiton' Ravensthorpe Nickel Plant.

Small Plants There are numerous small plants servicing coastal and island communi ties. The first was installed on H eron Island as long ago as 1980. Brackish water desalination is being used at a number of townships and mine sites in inland Australia. One problem is the varying quality of the groundwater, with a number of supplies bei ng high in d issolved silica and hardness, which cause fouling. T h e Water Corporation of Western Australia, in conjunction with GE and OsmoFlo, is p iloting two desalination options for h igh recovery from brackish ground water with high silica content.

These two plan ts are both located in the mid west and are an EDR (electro dialysis reversal) p lant from GE and a HERO (High Efficiency RO) plant at Yalgoo, to d eal with this problem. The HERO plant has been completed and initial results are proving to b e very exciting.

water

Journal of the Australian Water Association

Operating Tenders Aug 2008

400 250

2015?

Planning Proposed 2011 Planning, 2009 2009 EIS, 2009 2008

250 125 +45

? soon, 2008 2009

PART 2. MEMBRANE TREATMENT OF WASTEWATER The application of membranes for treatment of recycled wastewater has two advantages. Firstly membranes provide a barrier against micro-organisms, and also a large proportion of pharmaceutical chemicals, secondly, if extended to reverse osmosis, salinity is reduced. Th e former is important if indirect potable use is the target, the latter is more important for industrial re-use, as in boiler make-up and cooling towers.

Queensland Brisbane's WWTP at Luggage Point has been delivering about 8 ML/d of recycled water to th e BP refinery since 2002, and has a fi nal stage of RO to reduce salinity Starting in 2007, the Western Corridor Recycled Water Project is on-scream to deliver over 200 ML/d of purified water for two power stations, industry, agriculture and finally back to Wivenhoe Dam where it will be added to Brisbane's main potable water supply. The WCRWP will collect secondary treated wastewater from Brisbane's and Ipswich's major treatment planes at Luggage Point, Gibson island , Bundamba, Goodna, Waco! and Oxley.

T he created wastewater will then b e further created within three new Advanced Water

EQUIPMENT www.h2orx.com Ph: 0409 784 236 Fax: 02 9436 0491 96 MARCH 2008

Projected date

Reduce algal bloom, minimise evaporation and save more water with Synthesis

Water .storage shade structure â&#x20AC;˘ image courtesy of Superspan

In 1999 we recognised t he potential of our shade fabrics to provide highly effective protection for vu lnerable w ater resources in storage basins and dams.

In conjunction with the CS IRO and East Gippsland Water, we undertook world-f irst tria ls that p roved covering water storage basins w ith Synthesis shadecloth sign ificantly reduces alg al blooms, evaporation, noxious weeds and w ater contamination.

Synthesis shade fabrics 'breathe' to provid e t he perfect balance of light t ransmission and airflow. And because they aren't solid, there's no risk of biofil m growth on inner surfaces and they pose virtually no risk to people working inside a covered structure due t o possible b uild-up of hazardous gasses.

Under t he Synthesis brand, Gale Pacific designs,

-----

manufactures and supplies a range of premium , . '.

fabrics t hat protect and nurture so many of t he natural resources w e rely on for life.

Water is just one.

To find out how Synthesis fabrics can work for you in so many ways, contact us on 1800 331 521, email synthesis@galepacific.com or visit synthesisfabrics.com.

I

technical features

Treatment Plan ts (AWTPs). The AWTPs will treat the wastewater to the highest standard through a multi-barrier treatment system includi ng microfiltration (MF) and reverse osmosis (RO). T he final output will be over 200 ML/d. T he project incl udes some 200 km of pipelines .

T he first stage was the commissio ning of the AWTP ar Bundamba. Stage IA has a design output capacity of 20 ML/day. Stage 1B will follow w ith a design output capacity of 80 ML/day. Flows into rhe Bundamba AWTP will initially be sourced from Bundamba wastewater treatment plant (for Stage IA) and then supplemented by additional flows from Goodna, Waco! and Oxley wastewater treatment plants (Stage IB).

Stage 2 of rhe project involves the construction of rwo A WTP's at Gibson Island and Luggage Point. T he Gibson Island AWTP will have an initial capacity of 35 ML/day with an ultimate capacity of 50 ML/day. The Luggage Point AWTP will have an initial capacity o f 82 ML/day and can be upgraded to I 02 ML/day. Both AWTP's will initially provide water for transfer through the distribution pipelines with rhe additional capacity to produce water for local reuse to be added as local reuse demand arises.

New South Wales About 20 ML/d of high quality recycled water is being delivered to BlueScope Steel from a new recycled water plant at Sydney Water's Wollongong Sewage T reatment Plant.This replaces 7.3 billion litres o f drinking water per year previously drawn from the local Avon Dam, a 57 per cent reduction of drinking water consumptio n

by Sydney Water's largest customer.

The plant at Wollongong uses microfiltratio n and reverse osmosis membrane p rocesses to produce high quality recycled water suitable for a range of industrial purposes such as cooling systems. An upgrade in the near fu ture is being discussed.

The Western Sydney Recycle Water Initiative will recycle 27 millio n ML/year.

T he additional recycling will occur in the new growth areas to rhe no rth west and south west of Sydney where housing will be supplied with recycled water for non drinking househ old purposes as well as irrigation for agriculture.

Sydney Water has identified seven large water users in the C amellia and Smithfield areas to be provided with recycled water for the replacement of their drinking water usage. This recycled water will be used for industrial and irrigation purposes saving up 98 MARCH 2008

water

to six million m 3/year of drinking water a year. Treated effluen t from the Liverpool to Ash field Pipeline will be further treated at a recycled water plant which is likely to be based at Fairfield. The recycled water plant may use micro filt ratio n and reverse o smosis to treat the water to a very high quality for industrial use. A pipe network wi ll be constructed to distribute the recycled water to Sydney Water's customers.

The Replacement Flows Project involves connecting three sewage treatment plants (STP) at Penrith, St Mary's and Quakers Hill and transferring the created wastewater from these plants to a new advanced water treatment plant at St Mary's. This plan t is likely to have micro filtration followed by reverse osmosis with break point chlorination. Th is will ensure total nitrogen levels are less then I mg/ L. The new plant will treat the wastewater to an even higher standard than the created wastewater curren tly released into the HawkesburyNepean River from the th ree plants. This will improve the quality of the downstream reaches of the river and reduce the amount of potable water which has to be released from Warragamba Dam for environmental purposes.

Victoria Melbourne's Western Treatment Plant supplies recycled water of quite high salinity, typically 1200mg/L, to a near-by market garden complex, where it has to be blended with river water to reduce its sali nity to an acceptable level. Pilot trials of two types of RO and EDR have been conducted to assess the feas ibility of partially desalin ating the recycled water at so urce. Curren tly some 3000 ML/yr are supplied, with a further 1000 ML/yr fo r industrial, recreatio nal and residential purposes.

Melbourne's Eastern Treatment plane supplies treated water to new residential developments. It is treated by ultrafiltratio n as a fi nal pathogen barrier, but the salinity is only ca 500 mg/L so R O is not necessary. A target of 5000 ML/yr has been set.

The G ippsland Water Factory is on track fo r completion in 2008, to treat 8 ML/d of combined domestic and industrial wastewater by MBR and RO for supply to a major industrial user, the Maryvale paper mill. The eventual target is for 35 ML/d.

Apart from rhe treatment plant, the overall transfer system requires 75kms of pipeline and 8 sewer pump stations to be constructed. To d are approximately 37kms of pipeline has been completed . The target d are for commissioning of rhe Gippsland

Journal of the Australian Water Association

Water Factory system is late 2008 with commercial handover to Gippsland Water expected by mid 2010.

Western Australia In April 2005, rhe Water Corpo ration against a backdrop of d rying climate and a rapidly growing population - released rhe "Source Development Plan for the Integrated Water Supply Scheme". This formally recognised recycled water via groundwater replenishment as a potential d rinking water source, with the earliest d are of implementation being 20 I 4. A scheme treating and replenishing the entire flow from Beenyup wastewater treatment plant (WWTP) co uld allow an increase in groundwater abstraction of up to 30 million m 3/year fo r drinking water supply.

A trial of the approach is planned, wh ereby 5Ml/d of high quality water would be injected to rhe Leederville aquifer after ultra-fil tration, reverse osmosis, and possibly advanced oxid ation. Design definition is under way for the trial, and ir is planned to have the plant in operation by early 20 IO and operate it for three years.

Water Corporation has also constructed stage o ne of rhe Kwinana Water Recycling Plane (KWRP). This treats fl ow from th e Woodman Point WWTP to produce up to l 7ML/d of reverse osmosis treated water for use by various industries in the Kwinana industrial area, including oil refi neries, fert iliser and paint manufacturers and steel smelting operations. Demand now exceeds supply for the product water, an d Water Corporation are close to completing design o f the Stage 2 ! 0 Ml/d expansion of rhe plant, and are liaising with existing and potential customers to increase or commence supply of the high quality RO water.

The Authors Gary Crisp was in charge of desal ination for the Water Corporation until 2007, when he was appo inted Manager of Desalinatio n projects for G HD, based in the Brisbane Office. Gary.C risp@GHD. co m.au . He has recenrly been seconded to the Gold Coast Desalination Plane's owner, Sure Smart Water, as Chief Technical Officer. In October 2007 he was elected to the Board of the International Desalination Association; Bob Swinton, Technical Editor of Water, has updated Gary's paper and compiled rhis report by reference to numerous sources, mainly web sires, and would appreciate corrections or up-to-dare information, bswinton@bigpond.ner.au.

MAKING MODERN LIVING POSSIBLE

VL T® AQUA Drive - extended power range to 1.2 MW The ultimate solution for wastewater treatment & recycling The growing need for clean water and energy conservation is rapidly increasing the pressure on global fresh water resources, wast ewat er treatment, recycling and power generation.

VLT• AQ UA Drive is designed to enhance system operation, protect eq uipment, help reduce chem ica l consumption and water loss, wh ilst providing significant energy savi ngs.

,;:,

1

;:10

VLT* AQUA Drive is the ultimate solution for all water, wastewater and recycling processes and is supported by local experts in over 100 countries world wide.

0.25 kW - 1.2 MW power range Compact enclosures with seg regated cooling

EMC filter & DC Bus cho kes built-in Advanced harmonic filter optio ns Save on system & insta llation costs Protect eq uipment & piping Save energy Reduce operating & maintenance cost s

Visit Danfoss Drives at Enviro OB Exhibition at stand No. 172, Melbourne, 5 -7th May For further derails please conracr your local Danfoss Drives specilisr: Melbourne: Tel. (03) 9703 Sl00 • Sydney: Tel. (02) 8845 1800 • Brisbane: Tel. (07) 3630 1899 • Adelaide: Tel. (08) 81S0 7400 Perth: Tel. (08) 9333 3800 • Auckland: Tel. 64 9 259 2500 • Christchurch: Tel. 64 3 36S 6123

www.danfoss.com/pacific • Email: motioncontrols@danfoss.com.au

technical features

SOUTH GIPPSLAND WATER SALINE WASTEWATER PROJECT E Foley Abstract A 450 DN, below seabed, o cean outfall, 750 m long, plus a 100 m diffuser with 16 pores, was installed in an environmentally sensitive area of a National Park by horizontal directional drilling with minimal effect on the surroundings, despite the challenging conditions of Bass Strait.

Background to the Project South Gippsland W ater owns and operates an outfall pipeline which conveys process water from milk processing facto ries in Leongatha and Korumburra to Venus Bay where it discharges to Bass Straight. The 47 km outfall pipeline was commissioned in 1969, and at chat rime was the world's longest large d iameter PVC pipeline.

The outfall pipeline runs below ground to an outfall structure, which supports the pipeline about 5 metres above the intertidal beach and 100 metres into the surf zone of Venus Bay mid-way between Anderson's Inlet and Cape Liptrap. The structure was erected in the early eighties to raise and lengthen the buried outfall co address blockages caused by tidally induced sand movement. T he wastewater discharges continuously from the end of che pipe at a rate of about 4 ML per day but at peak times is close to pipeline capacity. T he volume and quality of the wastewater is regulated by an EPA waste discharge licence.

The project entails substantial upsizing of works, especially with respect to che pipeline and outfall, co cater for future expansion of ocher industries discharging saline effluent within the regio n. T his upsizing is being undertaken in recognitio n of the EPA's stated preference fo r the locarion of such saline waste industries close to a marine environment.

Construction of the pipeline upgrade and new o utfall structure was implemented under 2 separate contracts. COE Drill ing Pry Led was awarded the contract for constructing the new outfall structure at Venus Bay.

Trenchless technology reduces environmental impact.

New Venus Bay Ocean Outfall Structure The new Venus Bay ocean outfall structure comprises:

• An 11 metre depth reinforced concrete de-aeration chamber and air release pipeline located within the sand dunes of Cape Liptrap Natio nal Park • Connection pipework co the existing o utfall pipeline • 750 metre long 450mm N B H PPE below sea bed pipeline from the de-aeration chamber co a new seabed diffuser • New 108 metre long 450 N B steel seabed diffuser with 16 No. ports pinned co che sea bed.

The location of the works The New Venus Bay ocean outfall is located adjacent to the existing outfall structure on Venus Bay, mid-way between Anderson's Inlet and Cape Liptrap. The construction works site for construction of the de-aeration chamber, associated pipework, and drilling for the below seabed pipeline was located within sand dunes in the C ape Liptrap National Park.

100 MARCH 2008 Water Journal of the Australian Water Association

Environmental management Due co the environmental sensitivity of the constructio n location, consents and approvals from government agencies were required prior co commencement of construction of the works. The consents and approval included: • Section 27 (1) - Consent from Parks Victoria under the National Parks Act -

1975 • Sectio n 40 (1) - Consent from the D epartment of Sustainability and Environment under the Coastal

Management Act • Section 11 - Consent from the Department of Sustainability and Environment under the Conservation Forests

and Lands Act • Works Approval from the Environment Protection Authority As pare of the process of obtaining the consents and works approval, a derailed Environmental Management Plan was prepared co managing rhe environmental risks of construction work in the Cape Liptrap sand dunes and Venus Bay marine environments. COE Drilling cook a proactive role in preparing the

X-Flow leading in purification

For information on our Seaguard system please contact Australia - 02 9918 5088

Singapore - +65 6338 9919

rt..~ ~ WWF

for a living planet'

technical features

Environmental Management Plan, which required approval from the above governmen t agencies.

COE Drilling successfully implemented the construct ion component of the Environmental Management Plan without incident.

Construction COE Drilling commenced construction of the works in February 2005 and completed in April 2006, within So uth Gippsland Water's budget and time objectives.

Mobilisation, Design and Engineering Detailed project engineering and design was required co satisfy the client and to confirm chat the proposed methodology could achieve a successful installation of the produce pipes and conduits. Comprehensive management plans were prepared for che project covering OH&S and environmental management as well as all aspects of the d rilling and marine procedures including project risk analysis, engineering analysis covering stress analysis and marine mooring analysis for the marine spread.

Dri lling commences.

Work commenced with mobilisation of the CoeDrill 180 Drilling Spreads direct from another successful Shore Crossing project in Western Australia where two parallel 1000 m oil and water injection pipelines where installed for Roe O il as pare of the C liffhead Development Project. A drilling

water supply system was installed from a nearby dam to the drilling site co provide a concinuous supply of water for mixing drilling fluids at the site.

The fi nal design of the outfall involved installing a ON 450 HOPE 10 0 PN 16

HORIZONTAL DIRECTIONAL DRILLING

102 MARCH 2008 Water Journal of the Australian Water Association

(i)veouA WATER

technical features

pipeline from onshore behind the coastal dune system ro terminate 750m offshore in 10m water depth. The offshore diffuser was selected as a DN 450 16mm coated steel pipeline 108m in length secured to rhe seabed with galvanised piles ar 9m centres and with 16 outlet diffuser ports. The diffuser was coated both internally and externally wirh a liquid epoxy coating.

The locatio n of the drill ing site within Cape Liptrap Coastal Park managed by Parks Victoria and the conditions of consent for the project meant rhar fab rication of the HDPE pipeline and steel diffusers were nor possible at the HDD Sire. This work would be co mpleted remote to rhe drilling sire at Andersons Inlet near Inverloch some 20km along the coastline.

Selection of the fabrication site was based on access and approvals by Gippsland Ports.

Fabrication Operations On site ar Maher's Landing Road Inverloch, Coe Drilling fabricated the 750m long D N 450 HDPE 100 PN 16 pipeline onshore in readiness for rhe 20 KM seaward voyage to rhe offshore HDD exit location. A marine spread was mobilised to site complete with air diving spread to complete the underwater connection of the H DPE pipeline to rhe HDD drill sering.

The DN 450 steel diffuser complete with outlet porrs and pile support brackets was fab ricated by Coe and coated both internally and externally prior to delivery to sire. The steel diffuser was designed ro be rowed to site in rwo sections and bolted together on rhe seabed. The fabricated steel diffuser was joint coated both internally and externally and was fitted with buoyancy tanks to assist during the seaward row ready for use and positioned on Coe's pipeline rollers at the shore.

Drilling Operations D rilling operations commenced in early November using the Coedrill 180 drilling spread.

The pilot hole was drilled with a 12.25" mill tooth bit and a jetting assembly using a conventional downhole wireline magnetic guidance system. The drill sering used was a 6 5/8" Grade S 135. The ground conditions, as expected, consisted of very dense fine and coarse sands.

Following completion of the pilot hole the borehole was opened up in progressive stages (17.5" & 24") by back- reaming operatio ns. The offshore marine spread and d iving operations were in place during all stages of the crossing fitting and breaking

104

MARCH 2008

Water

our downhole tools offshore. T he water depth at the offshore HDD exit was 1 Om.

HDPE Pipeline Installation Once all hole opening had been completed and a anchoring system was installed at the HDD exit site the prepared DN 450 750m long HDPE pipeline was rowed out from Maher's Landing through the channel at Andersons Inlet with a tide window of less than three hours turn around. The pipeline row could only proceed with all tide and weather conditions complying with the requirements of the safety plan developed for the project. In Bass Strait chis can be difficulc ro program and execute, as Coe would know havi ng completed three ocher shore crossings along the Victorian coastline in recent years. The marine support spread continued ro pull the HDPE pipeline offshore until in position and aligned with t he HDD exit. The DN 450 HDPE pipeline was flooded prior to the subsea hook up to the drill string pullback swivel assembly using the diving spread.

Steel Diffuser Installations and Civil Works After installation of the DN 450 HDPE pipeline within the borehole, Coe cowed the steel diffusers from the fabricatio n site at Andersons Inlet to the offshore HDD exit site using buoyancy tanks and once at rhe site che tanks were removed and the diffusers fl ooded, submerged and secured ro rhe seabed. The 108m long section of DN

Journal of the Australion Water Association

4 50 steel d iffuser was secured using steel pin piles, which were jeered into rhe seabed 6m.

Onshore rhe HDD spread was demobilised from sire in preparation for construction of rhe de-aeration chamber. The chamber consisted of a 1 l.0m deep x 7.7m long concrete chamber which requi red a sheer piled cofferdam to be constructed with dewatering ro allow construction of the base and rising walls to commence. Once completed, the existing outfall pipeline was isolated and connections to the new outfall completed.

Conclusion The successful completion of the new Venus Bay outfall shore crossing by horizontal directional drilling demonstrates how che HDD process can be used in environmentally sensitive areas such as the Cape Liptrap Coastal Park without incident and ensures the landscape is untouched by the trenchless methods. The landscape still bears the scars of rhe original works from the open cue method used back in 1969.

The Author Eamon Foley is the Contracts Manager of Coe Drilling Pry Ltd. H e is a civil engineer with over 15 years specialist experience in HDD and has overseen several HDD shore crossing installations in Bass Strait as well as Western Australian waters. H e has completed in excess of 150 drilled crossings for pipelines by trenchless technology, email: e.foley@coedrilling.com.au

.fereed paper

PRE-TREATMENTS FOR SEAWATER REVERSE OSMOSIS B Bolto, T Tran, M Hoang Abstract Membrane life is extended significantly when organic and biological species and turbidity particles are fi rst removed from rhe raw water. Pre-treatment has been found to be essential for at least half of rhe major RO seawater desalination plants chat have been installed around che world. The choice of conventional coagulacion/sedi mentacion/filcracio n or low pressure membranes is dependent on local factors rhar need to be assessed in pilot studies; in extreme cases both are used.

Introduction Pre-creacmenc is prudent in any reverse osmosis (RO) or nanofilcrarion (NF) application to fi rstly remove turbidity and dissolved organ ic matter and so reduce rhe fo uling porencial of che feed water. le can be che most critical item in the process train.

Much attention has been given to pretreacmenc of wascewarers, including recent articles in chis journal (Davey et al., 2007a, 20076). Pre-treatmenc is also highly important in seawater desalination because of che marine flo ra, algal blooms and phytoplankton chat cause high turbidity.

The quality of raw seawater is exceedingly variable because of geophysical conditions such as currents, evaporation rates and weather. It can be affected by the local environment because of storm water ru noff or pollution in general, notably from hydrocarbons in some areas, which can lead to increased sedimentation or extra loads of nutrients, pathogens, pollutants or salinity (EI-Azizi and Omran, 2002). The salinity variation can be from 49,600 mg/L for Persian Gulf seawater to 18,000-31 ,000 mg/L for the Eastern United States (Adham et al. , 2006).

A very good example of what can happen without appropriate pre- treatment is che situation at Tampa Bay, Florida. T here the largest seawater desalination facility in the United States was intended to produce an initial 95 ML/d of drinking water, with a fu tu re plan ned expansion of 37 ML/d to help reduce che growing demand on the area's aquifers. It first produced water in 2003, bu t had to be placed in standby mode nearly a year later. The plant used a pre-treatment system comprising two-stage dual sand fil tration to remove turbidity, algae, organic material, and other particulate matter from the incoming raw seawater. The system was not able to consistently meet the silt density index (SDI) target set to protect rhe RO system.

As a result, rhe filte rs and membranes

rapidly clogged up, the mem branes fou led rapidly and the plant could nor achieve peak water production. Ic faced more frequent membrane cleaning, significantly increasing operating costs because of higher energy consumption, increased chemical usage, and more freq uent membrane replacemenc (Siverns et al., 2004).

Conseq uently it on ly operated intermittently and produced less drinking water than designed. The plane req uired major repair and remediation work,

Conventional coagulation/sedimentation/ filtration or low pressure membranes.

····---.-...-..

VYI\ I I:. K.\.U Water, the liquid of life CAST IRON AND STAINLESS STEEL PUMPS

Capable of processing up to 396 m'/hr.

CARTRIDGE AND BAG FILTERS

3.65 kW to 11 kW.

ELECTRONIC COAGULATION

Inadequate or no pre-treatment before an RO system can be serious as it can give rise to: Capable of withstanding up to 600kPa.

• rapid membrane fouling

Precipitates and coagulates a wide range of contaminants.

• excessive cleaning requirements NSW (Head Office) QLO 02 9898 8686 07 3299 9900

• lower recovery rates • higher operati ng pressure • shorter membrane life • lower quality product water (Siverns, 2006; Fritzmann et al., 2007):

VIC/TAS SA / NT WA 03 9764 121 1 08 8244 6000 08 9273 1900

www.waterco.com

@

~

~~f.~ [~ :.

Journal of the Australian Water Association

Water

MARCH 2008 105

technical features refereed paper

including modifications to the chemical facilities and the pre-treatment system, which now uses ferric sulphate addition, flocculation, and sedimentation. le became fully operational in late 2007, some six years behind schedule.

Fouling of NF systems has been extensively reviewed recently (Schafer et al., 2005), as have pre-treatments to control it, which are similar co those used for RO, in chat there is a need co (Tanninen et al., 2005): • remove suspended material, colloids (silt, colloidal Fe, Al and silica) and organic matter • reduce microbiological fouling • add pH adjustment and anciscaling chemicals • remove oxidants in the feed if the membrane type requi res it.

Chemicals are often added co protect against the format ion of membrane scale, made up of CaCO3, CaSO4, BaSO4, SrSO4, CaF2 or SiO 2, and then there is usually filt ration through a cartridge fil ter (Bodzek, 2000). Acidification to lower pH and scale inhibitors such as sodium hexametaphosphate are applied, decreasing the formation of carbonate and sulphate deposits. The deposition of silica is minimised by operating at pH >8 or <7 and raising the temperature. Eucrophicacion problems with phosphates have meant chat they are largely replaced by polymers (Fritzmann et al., 2007).

Generally, pre-treatment approaches include conventional coagulacion/sedimen cacion/ filtration, microfilcration (MF) or ultrafilcration (UF) membranes, combinations of conventional and MF/UF, or simply the use of cartridge filtration (Adham et al., 2006; Burbano et al., 2007).

A survey has been carried out of 62 planes worldwide, mostly in the United States, 84% of them RO plants and I 6% NF planes. Ir was confined to plants of 4 ML/day or larger constructed since 1990.

They were used for seawater (11 %), surface water (6%) and brackish water (60%) desalination, and wastewater reclamation (23%). le was found chat simple cartridge filtration was universal, and was sometimes the sole creacmen t. The percentages of the various methods over all applications were: • Camidge Filtration Only - 55% • Camidge Filtration Plus Ocher - 45% • Membrane Filtration, via MF - 21 % • Coagulation/Sedimentation/Filtration 13% • Combination of Conventional and MF 11% 106 MARCH 2008

Water

Table 1. Contaminants generally rejected by various membrane processes.

PROCESS

Monovalent ions

Microfiltrotion Ultrofiltrotion Nonofiltrotion Reverse osmosis

Divalent ions

Micro·organisms

..J

.J .J .J

Suspended solids ..J

..J

..J

..J ..J ..J

Conventional Coagulation/ Sedimentation/Filtration

• were effective over a broader dose range, and

Conventional coagulation/sedimentation/ filtration is involved in most pre-treatment for seawater applications, with ferric chloride as the preferred coagulant in I 00% of rhe plants surveyed, although alum is certainly used elsewhere. Some 85% of the plants made use of dual media filters. The pre-treatment is optimised co enhance rhe efficiency of the RO membranes by removing as much as possible of the particulate, microbial and organic matter.

Ch lorine is typically dosed at 0.5 mg/L upstream of the coagulant co reduce biofilm formation in the downstream filters; copper sulphate is also employed. The 140 ML/day RO plane in Perch makes use of an ironbased coagulant, followed by single stage dual media filtration and 5 µm cartridge filters (McCann, 2007).

• had greater organics removal.

Polymer may be dosed with the inorganic coagulant to help destabilise colloidal matter, but the practice varies with location .

Thus in one study the addition of 1 mg/L of a cationic polymer as coagulant aid was tested, but in the first year of operation, optimising the coagulation process eliminated the need for polymer altogether (Irwin and Thompson, 2003). Ocher work reported significant advantages from the use of 0.1 mg/L of cationic polymer as it reduced the coagulant dose, making fo r worthwhile cost savings (Al-Sheikh, 1997).

The selection of the optimum polymer type has been explored (El-Maharawy and Hafez, 2000). Comparison of cationic, anionic and non-ionic polymers showed that the nonio nic polymer gave the best results, although all organic polymers significantly improved the coagulation process. In general, it has been found char such polymers resulted in: • lower doses • longer filter runs

The effects of coagulant residuals on polyamide RO membrane performance have been investigated (Gabelich et al., 2006). A wide range of inorganic and organic polymeric additives was examined. It was found chat aluminium residuals, most notably from alum coagulation, caused colloidal fou ling because of interactions with silica co form aluminium silicates, and with anciscalants. The effectiveness of the latter reagents was also reduced. Although ferric chloride lowered the tendency towards colloidal fou ling, the presence of ferric ions enhanced the oxidation of polyamide membranes by chloramines. Aluminium chlorohydrate also minimised colloidal fouling and reduced aluminium-antiscalant interactions, but seemed to enhance oxidation of the membrane surface. The fouling of membranes by organic polymers appeared to be minimal.

The effects of pre-treatment with coagulants on UF performance with turbid surface waters vary with the mode of operation, depending on whether coagulation was conventional, formed a dynamic layer on the membrane, or was in-line (Sharp et al., 2006). The dynamic or secondary, coagulant-based layer achieved higher steady state flux values than unassisted UF, while in-line coagulation gave the worst flux decline.

Low Pressure MF or UF Membranes An integrated or hybrid membrane approach has been reviewed, making use of some of the membrane technologies summarised in Table I , which gives a general indication of the species that are rejected by membranes of increasing rightness.

but had issues associated with backwashing (Hagen and Comstock, 2004, quoted by Adham et al., 2006; Patterson et al., 2006; Franks et al., 2006).

Using low pressure (70-600 kPa) MF or UF membrane technology offers several benefits (Adham et al., 2006). The pre-created seawater has a higher quality in terms of suspended solids and biological content (Henthorne and Quigley, 2003), so there is:

Inorganic coagulants, on the ocher hand, offered:

• less cleaning of the RO membranes required

• better filtrate quality

• less foulin g so lower pressure drops

• better solids retention,

Journal of the Australian Water Association

2 heads are not always better than one.

New Technology from KMS

Old Technology PURON™ MBR Single Header

Double Header MBR Heads clog due to non-central air flow

• Reduced Energy Demand Centrally located air nozzles improve aeration efficiency by up to 50%.

High system stress Costly maintenance

• Red uced Capital Costs Simple pre-screening due to low risk of module clogging

More cleaning required Less effective filtration

• Higher Reliability Minimal maintenance

Head into the future of wastewater treatment with KMS. Choose PURON™ MBR.

141 KDCH MEMBRANE SYSTEMS ""

• longer RO membrane life

Table 2. Comparison of benefits, conventional versus membrane pre-treatment

• increased flux

(Vedavyasan, 2007).

• reduced footprint • lower overall chemical usage

Item

Conventional

MF/ UF

Capital costs

Competitive with MF/UF

Although 0-25% higher, life cycle costs are comparable

Foot print

Larger

30-50% smaller foot print

Energy requirements

Lower, usually gravity flow

Higher

Chemical costs*

High; coagulant and polymer always required

Low, depending on raw water quality; coagulant only for organics reduction

• less sludge handling.

A recent example of UF before RO is from China, where operation of a UF pilot plant in a high recovery (80%), low flux (60 L/m 2h) mode was found co be best (Xu et al., 2007). From seawater having a coca! suspended solids content of 4.8-8.9 mg/L and a chemical oxygen demand of 1.151. 76 mg/1, che feed for the RO plant had a turbidity 0.01 NTU, with no chemicals being used for disinfection, flocculation, enhanced chemical backwashing or cleaning.

Pilot studies have shown chat disc filtration in lieu of fine-m esh strainers prior co lowpressure membranes used in RO plants avoids the operational challenges encountered when seawater contains high levels of suspended solids, debris, seaweed and so on (Huehmer et al., 2006). There was reduced membrane fai lure caused by sharp and serrated debris.

Removal of the biodegradable organic content of seawater by means of a fl uidised bed biological activated carbon o r membrane bioreaccor (MBR) has been tested (Visvanathan et al., 2002). The MBR gave better removal and yielded the higher flux improvement.

High Pressure NF Membranes NF membranes are used for preliminary softening and organic removal, caking out material chat has MW >200 Da. The reduction of NaCl is usually about 50 %, and that of CaSO4 about 90%. The pressures required are 0.3-3 MPa, versus 47 MPa for RO. The application of NF before RO has been investigated as a means of reducing the concentration of divalent ions that are responsible for membrane scaling (Hassan et al., 2000). NF provides excellent feed water for RO, allowing operation at much higher fluxes and recovery rates. Other viewpoints however consider that the approach merely transfers the scaling problem further upstream (ElMaharawy and Hafez, 2000).

Recovery*

96%

93-95%

Average turbidity*

0.13 NTU

0.05 NTU

Cartridge replacement*

2-3 months

3-4 months

RO flux

Lower

Higher

RO capital cost

Higher

Less as flux can be 20% higher

RO operating costs

Fouling potential greater, more frequent cleaning

Less fouling potential and longer membrane life

* For San Francisco Bay (Frenkel et a/., 2006)

Table 3. Capital cost comparison, conve ntional versus membrane pretreatment at 20 ML/day (Pearce, 2007a).

Item

Conventional in kUS$

MF/UF in kUS$

RO

2000

2000

Pre-treatment

800

1700

Process equipment, pumps, site works

14,050

1,850

Indirect

590

5900

Total

22,750

23,450

Other coupling of NF and RO for the desalination of brackish water is reported, with the emphasis on the effects of di- and monovalent cations on operation (M'nif et al., 2007). At a pressure of only 0.6 MPa when run in sequence with RO before NF, the brine reject from the RO unit was desalted with respect to divalent ions, so chat the recovery of 40% for RO was increased to 80% with coupled RO/NF. When done in the reverse order, and conducted in parallel with recirculation so that the divalent ions were reduced by NF, the osmotic pressure co the RO decreased and the RO flux and recovery were also increased. However, so far pre-treatment with NF has not proven co be economically feasible (Ericsson, 2005, quoted by Fritzmann et al., 2007). H igher recoveries

Table 4. Chemical cost comparison, conventional versus membrane pre-treatment, 20 ML/day (Pearce, 2007a).

Item

Conventional in kUS$

RO cleans per year

MF /UF in kUS$

MF/UF in kUS$

3

2

Dosing & MF/ UF cleaning

61.4

24.1

24.1

RO cleaning

83.5

55.7

27.8

Total

144.9

79.8

51.9

108 MARCH 2008

water

Journal of the Australian Water Association

can in special circumstances make up for the additio nal investment cost, but these are unlikely to occur in the practical operation of a desalting plant.

Combined Technologies The prior addition of a coagulant before UF or MF can help in the removal of natural and synthetic o rganic compounds.

In extreme cases such as Jeddah, Saudi Arabia, the quality of the seawater may be so poor because of the facility being within a port area fo r example, where frequent ship movement and algal blooms can upset the qualiry, a combined approach such as coagulation, dual media filt ration and UF may be necessary (Buisson, 2006).

A large RO plane in C hina chat is co desalt highly variable poor quality seawater for use as boiler feed water and for potable purposes. It will use enhanced coagulation and clarification, fo llowed by immersed membrane UF co produce 76.8 ML/day of exceptional quality feed water (Siverns, 2006).

Comparison of Conventional and Membrane Pre-treatment The water quality obtained from MF pretreatment is better rhan that from the conventional approach, with in one example fo r seawater from the intake of the Doha East Power Station in Kuwait showing respective SDis of 2.1 and 3.3 (Bou-Hamad et al., 1997). Another report is on high-fouling seawater from G ibraltar char is subject to seasonal algal blooms, and has an SDI of 13-25. UF pre-treatment with coagulation at low dose gave an SDI of< l , whereas the conventional method failed co reduce the SDI below 2.5 (Brehant et al., 2002). A higher fl ux race and higher recovery were obtained. RO membranes

desalting seawater from the Gulf of M exico require cleaning after 980 h operation on conventionally pre-treated water, whereas those exposed to membrane pre-treated water needed no cleaning after the 1300 h evaluation period (Henthorne and Quigley, 2003).

Membrane pre-treatment with hollow fibre membranes is claimed to be cost competitive with conventional technology in one study (Wilf and Scheirach, 2001). The higher flux rate and higher recovery followi ng MF make for better reliability and overall economics because of lower investment and operating costs, which would result in about a 10% reduction in total water cost. Another investigation found pre-treatment with hollow fibre UF to be more expensive than the conventional, but feasible for sites that require very extensive conventional pre-treatment, or where there are wide fluctuations in raw water quality (Glueckstern et al., 2002).

qua~a.~

SCl€ff!lifJC

Aqualab Scientific is recognised as a leader in the environmental industry, supplying quality brands that customers have known and trusted for years.

,,.:,ffl

Water Quality Instruments

Water Quality Multiprobes measure and store temp, pH, conductivity, LDO (luminescent dissolved oxygen), self-cleaning turbidity, chlorophyll a, bluegreen algae, redox, depth etc.. in one rugged housing.

Used for monitoring lakes, rivers, estuaries, oceans, groundwater and wastewater.

Sensors ore built to lost, so your total cost of ownership is ultimately the lowest in the industry.

Overall, the advantages of membrane pre-treatment over conventional are (H enthorne and Quigley, 2003; Kumar et al., 2006)): • better feed water quality, especially the suspended solids and biological content • less RO membrane cleaning, so less chemicals usage • lower pressure drop in RO arising from fouling • longer RO membrane life • increased flux rates and smaller plant footprint, and therefore lower capital investment • lower overall chemical and sludge handling coses • better recovery of product water if NF precedes RO (M'nif et al., 2007; EI-Zanati and El-Khatib, 2007).

utomatic Wat....

AMCIIICAlf

Summaries of the situation have been published recently, as shown in T able 2 (Vedavyasan, 2007). Data for a specific example at San Francisco Bay are included (Frenkel et al., 2006). Many more publications have appeared that substantiate these claims (Bonnelye et al., 2004; Pearce et al., 2004; Franks et al., 2006; Patterson et al., 2006; El-Zanati and El-Khatib, 2007; Pearce, 2007a; Bonnelye et al., 2008).

A comparison of the cost of conventional and membrane-based pretreatment has been published (Cote et al., 2005). Estimated capital costs for a 75 ML/d plant desalting seawater were 944 and 1005 $/kL/d respectively for media fi ltration and immersed UF as pretreatment, and the total life cycle costs were 0.6 15 and 0.637 $/kL. Thus for UF there was a premium of 6.5% and 3.6% for capital costs and total lifecycle costs over that for conventional pre-treatment.

H owever there is the potential to reduce costs for the all membrane arrangement taking into account the better and more rel iable pretreatment and consequent operation of the RO at increased flux, with higher recovery and longer membrane li fe. The overall smaller footprint is another advantage.

A capital cost comparison for feed water from the eastern Mediterranean Sea having a salinity of 38,000 mg/L is shown in Table 2. A similar conclusion can be drawn.

Authors

Dr Brian Bolto, Dr Thuy Tran and Dr Manh Hoang (email: brian.bolto@csiro.au; thuy.tran@csiro.au; manh.hoang@csiro.au) work for CSIRO Materials Science and Engineering, Clayton, Victoria.

SD900 Portable Sampler sets up easily and quickly in the field. Reduced maintenance and reliable results ore assured. Automatic water samplers, flow meters, rain quality instruments, communication products & data management software.

Water Level Do

u

a..v~~- •

DIVER®

ii i!l i!, i51

The chemical cleaning of the RO membranes receiving coagulated seawater that has been MF/UF pre-treated are minimal, and this reduced frequency of membrane cleaning is the main advantage. The major benefit of MF/UF is that it provides a more reliable and stable system, tolerant of variations in the feed water quality and of upsets in the operating system. It is estimated that some 40% of UF/MF systems sold are for RO pre-treatment (Pearce, 20076).

The

S~AU

la !: a

vv

---

Diver Doto Loggers hove become on industry standard in the measurement of water level and salinity in groundwater and surface water applications. The pressure transducer, temperature sensor, salinity sensor, data logger and battery ore integrated into a single robust device.

02 9894 4511 Journal of the Australian Water Association

water

MARCH 2008 109

References Adham, S., Burbano , A., Chiu, K.-P. and Kumar, M. (2006). Development of a reverse osmosis/nanofiltration (RO/NF) know ledge base. Final Report for

California Energy Commission, Montgomery Watson Harza, San Diego.

Al-S heikh, A . H . H. ( 1997) . SWRO pretreatment with an emphasis o n t he Jeddah plant operation experience. Desalination 110, 37-48.

Bodzek, M. (2000) . Membrane techniques in water t reatment and renovation. In: Water

Management, Purification and Conservation in Arid Climates (ed. M. F. A. Goosen and W. H. Shayya), Technomic, Lancaster, Pe nnsy lvania, pp. 45- 101.

Bonne lye, V. , Sanz, M. A., Durand, J .-P., Plasse, L. , Guegue n, F. and Mazo unie, P. (2004). Reve rse osmosis on ope n intake seawater: P re-treatment strategy.

Desalination 167, 191 -200.

Bonne lye, V ., Guey, L. and De l Castillo, J. (2008). UF/MF as RO pre-treatme nt: the real benefit. Desalination 222, 59-65.

Bou- Hamad , S., Abdel-Jawad, M., Ebrahim, S, Al-Mansour, M. and AI-Hijj i, A. ( I 997).

Performance evaluation of three d ifferent pre-treatment systems for SWRO technique. Desalination 110, 85-92.

Brehant, A., Bonnelye, V. and Perez, M. (2002). Comparison of MF/UF pretreatment with conventional filtration prior to RO membranes for surface seawater desalination. Desalination 144, 353-360.

Bui sson, H . (2006). Pretreatment experiences and lessons learnt from over 25 years of constructing large SWRO plants . Proc.

AMTA Conf, Amer. Membrane Technol.

Assoc., Anaheim.

Burbano, A. A. , Adham, S.S . and Pe arce, W. R. (2007). The state o f full-scale RO/NF desal ination - results from a worldwide survey. JournalAWWA 99(4), 116- 127.

Cote, P. , Siverns, S. and Monti, S. (2005) .

Comparison of membrane-based solutions for water reclamation and desalination.

Desalination 182, 25 1-257.

Davey, A., Schumann, R. and Hoehn, K. (2007a) . Treat ment of wastewaters by RO: Operating issues - Part I . Water 34(5), 4447.

Davey, A., Schumann, R. and Hoehn, K. (2007b). Treatment of wastewaters by RO: Operating issues - Part 2. Water 34(6), 4656.

EI-Azizi , I. M. and Omran, A. A. M. (2002).

Design criteria of 10,000 m3/d SWRO desalination plant of Tajoura, Libya.

Desalination 153, 273-279.

EI-Maharawy, S. and Hafez, A. (2000).

Technical manageme nt of RO system.

Desalination 131, 173- 188.

110 MARCH 2008

water

EI-Zanati, E. and El-Khatib, K . M . (2007) .

Integrated membrane-based desalination system . Desalination 205, 15-25.

Ericsson, P. (2005). Evaluation of nanofiltration as pre-treatment to reverse osmosis in seawater desalinati on. Proc.

IDA World Congress, S ingapore .

Franks, R. , Wilf, M., Voutchkov, N., Murkute , P. and Kizer, J. (2006). A pilot study using seawater reverse osmosis membranes in combination with various pre-treatments to meet the cha I lenges of Pacific seawater desalination . Proc. AMTA Conf, Amer.

Membrane Technol. Assoc., Anaheim .

Frenkel, V. S ., Reynolds, T. K. , Kennedy, P. E. and Macinni s, G. (2006). Membrane (UF/MF) vs. conventional pre-treatment for SWRO in California . Proc. AMTA Conf , Amer. Membrane Techn ol. Assoc., Anaheim .

Fritzmann, C., Lowenbe rg, J., Wintgens, T. and Me li n, T. (2007). State-of-the-art of reverse osmosis desalination. Desalination 216, 1-76.

Gabel ich, C. J., Frankl in, J. C., Cohen, Y. and Suffet, I. H . (2006) . Reverse osmosis pretreatment: Chall enges with conventional treatment. In: Membrane Treatment for

Drinking Water and Reuse Applications (ed. K. J. Howe), Amer. Waterworks Assoc., De nver, pp. 149-168.

Gl ueckstern , P., Priel, M , and Wilf, M. (2002).

Field e valuation of capillary UF technology as a pretreatment for large seawater RO plants. Desalination 147, 5562.

Hagen, D. and Comstock, D. (2004). Granular media filtration pre-treatment. Proc. !DA

Conf Congress: RO Pre-treatment Workshop, San Diego. Internati ona l Desalination Assoc., T opsfield, Mass.

H assan, A. M. , Farooque, A. M., Jamaluddin , A. T. M. , AI-Amoudi, A. S., Al-Sofi , M. A. K. and AI-Rubaian, A. F. (2000). A demonstration plant based on the new NFSWRO process. Desalination 131, I57171.

Henthorne, L. and Quigley, R. (2003).

Evaluation of MF, UF and conventional pre-treatment for seawater RO applications. Proc. IDA World Congress, Sep 28-O ct 3, Paradise Island, Bahamas.

International Desalination Assoc., Topsfield, Mass.

Hue hmer, R. , Henthorne , L. and Guendert, D. (2006) . Increasing MF/UF reliability in seawater desalination pre-treatment applications using enhanced pre-filtration.

In: Membrane Treatment for Drinking Water and Reuse Applications (ed. K. J. Howe), Amer. Waterworks Assoc. , Denver, pp. 209-230.

Irwin, K. J. and Thompson, J . D. (200 3).

Orinoco fluctuations fail to make filters fail. Int. J. Desalination & Water Reuse 13(3), 12-16.

Kumar, M. and Pearce, W. (2006). An integrated approach to pre-treatment evaluation for seawater reverse osmosis using bench- and pilot-scale testi ng. In:

Membrane Treatment for Drinking Water and Reuse Applications (ed. K. J. Howe ),

Journal of the Australian Water Association

Amer. Waterworks Assoc., Denver, pp.

489-505.

McCann , W. (2007). Perth marks A ustrali a's progress to desalination. Water 21, June 2007, 17- 18.

M ' ni f, A., Bouguecha, S., Ha mrouni, B. and Dhahbi, M. (2007). Coupling of membrane processes for brackish water desalination.

Desalination 203, 33 1-336.

Patterson, A. W. , Johnson W. T. , Sundstrom, G. P. and Gallagher, P. M. (2006).

Seawater pretreatment us ing low-pressure membranes. Proc. AMTA Conf, Amer.

M embrane Technol. Assoc., Anaheim.

Pearce, G. K. (2007a). T he case for UF/MF pretreatment to RO in seawate r applications. Desalination 203, 286-295.

Pearce, G. (2007b). Introduction to membranes: Manufacturer 's comparison.

Part I. Filtration+Separation, October, pp .

36-38.

Pearce, G. , Talo, S., Chida, K. , Basha, A . and Gu lamhusein. A. (2004). Pre-treatment options fo r la rge scale SWRO plants: Case studies of UF trials at Kinsada, Saudi Arabia, and conventional pre -treatment in Spain. Desalination 167, 175-1 89.

Sharp, M. M. and Escobar, I. C. (2006).

Determining the e ffects of coagulant pretreatment on ultrafi ltration. In: Membrane

Treatment for Drinking Water and Reuse Applications (ed. K. J. Howe), Amer.

Waterworks Assoc. , Denver, pp.169- 186.

Schafer, A. I., Andritsos, N ., Karabelas, A. J., Hoek, E. M. V., Schneider, R. a nd Nystrom , M. (2005). Fouling in nanofiltration. Nanofiltration - Principles and Applications, A. I. Schafer, A. G. Fane and T. D. Waite (Eds.), Elsevier, Oxford, pp. 169-240.

Siverns, S ., (2006). Us ing ultrafiltration as a pre-treatment before RO. Ultrapure Water, 23(3), 36-39.

S iverns, S., Maness, R . and Movahed, B. (2004) . Alternative pre-treatment tests with UF yield excellent results at Tampa Bay. Journal A WWA 96(12), 30-32.

Tanninen, J. , K amppinen, L. and Nystrom, M .

(2005) . Pretreatment and hybrid processes .

Nanofitration - Principles and Applications, A. I. Schafer, A. G. Fane and T . D. Waite (Eds.), Elsevier, Oxford, pp.

24 1-262.

Vedavyasan, C. V. (2007). Pre-treatment trends - an overview. Desalination 203, 296-299.

Visvanathan, C., Boonthanon, N., Sathasivan, A. , and Jegatheesan, V . (2002). Pretreatment of seawater for biodegradable organ ic content removal using membrane bioreactor. Desalination 153, 133-140.

Wilf, M. and Scheirach, M. K. (2001).

Improved performance and cost reduction of RO seawater systems using UF pretreatment. Desalination 135, 61-68.

Xu, J., Ruan, G., Chu, X., Yao, Y., Su, B, and Gao, C. (2007). A pilot study of UF pretreatment without any chemicals for SWRO desalination in China . Desalination 207, 216-226.

tectinical ft:iatures

RESEARCH ON ADVANCED MEMBRANE TECHNOLOGIES S Gray Introduction The profile of water issues and water research in Australia has increased enormously over the last 10 years of drought, and many believe the recent changes in climate to be the effects of global warming and permanent cl imate change.

T he change in rainfall patterns across Australia over the last 50 years shows that our major capital cities are becoming drier (Figure 1). T he high level of urbanisation in Australia means that 90% of Australians are living in drying climates.

T he response to this change in rainfall patterns has been to source our water supplies from poorer quality water sources, with water recycling and seawater desalination systems being implemented.

Indeed, 5 years ago Australia had no seawater desal ination plants operating.

Today we see desalination plants planned for Melbourne, Sydney and South Australia, a plant under co nstruction in South East Queensland, and Perch has one operating plant and another in rhe planning phase. This move to desalination plants has been rapid and along the way there has been co mmuni ty discussion regarding the additional greenhouse gases emitted from utilising seawater for our fresh water supplies. T he most efficien t process for desal ination in the absence of large quantities of waste heat is reverse osmosis, but even this process uses 3.4 kW.h/ m3 of energy for seawater desalination which is still fou r times rhe theoretical limit of 0.8 kW.h/m 3 . By comparison, wastewater recycling is currently achieved for an energy requirement of 0.8 kW.h/m 3, rhe rheorerical limit for seawater desalination.

Community concern around the energy required fo r desalination has lead to rhe inclusion of renewable energy systems as pare of these projects, so that energy required for desalination is offset by the

Australia at the forefront of novel membrane development.

This article is based on rhe cired media release from CSIRO in 2007, updarcd and expanded.

800

600

:;o 0 400 30.0 200 100

u.u - 100

SE Qld

20.0

1.5 M

-30.0

~00 ·50.0 --000

0 ;I--

800

Trend In Annual Total Rainfall Melbourne 3.6~ ~ · 1950-2002 (mm/10 years)

Figure l. Australia' s change in rainfall since 1950.

Specify and use Link-Seal® seals to seal pipe penetration

Slide the assembly into the space between the pipe and wall opening

PR •jEX

When the bolts are tightened, Link-Seal modular seals expand to create a gas and water tight seal.

Available in Australia and New Zealand

Projex Group Pty Limited Telephone: 02 83361666, Facsimile: 02 83361670 www.projex.com.au, email: mail @projex.com.au

Journal of the Australian Water Association

Water

MARCH 2008 111

technical features

addit ional renewable energy in the electricity grid. W hile this is commendable, this power could be used to offset the power requirements for other uses if water with lower embedded energy were used.

Recycling of wastewater is also a reliable supply of water, and Brisbane is embarking on large scale indirect potable water reuse.

Additionally, the number of non-potable water recycling schemes has also increased d ramatically over the last 10 years and with Australia's largest cities all having recycling targets, chis trend in the use of reclaimed water will remai n fo r many years to come.

While energy use associated with treatment for water recycling schemes is signi fica ntly lower than fo r seawater desalination, water recycling does need to contend with higher health risks and potentially higher membrane fou ling races. The fouling races of reverse osmosis membranes treating ind ustrial wascewacers is unpred ictable, and several of the large recycling fac ilities have been unable to reach their design capacity because of unforeseen foulants. T his is a major concern fo r operators of these plants.

Boch seawater desalination and wastewater recycl ing are key co providing reliable water supplies for Australia bo th now and into the fu ture. Such schemes rely heavily on membrane technology to provide the required water quality, and therefore, membranes are a key strategic technology fo r Australia's fu ture prosperity. With ch is in mind, CSIRO's Water for a Healthy Co untry Flagship established the research cluster on Advanced Membrane Technologies for Water Treatment (http:/ /www.csiro.au/ partnerships/ps30e. hem!).

Research Projects The Advanced Membrane Technologies for Water Treatment cluster project is aiming to reduce the energy required for desalination from 3.4 kW.h/m3 co 1.7 kW.h/m3, and to place Australia at the fo refro nt of novel membrane development.

In order to reach these goals, a multid isciplinary research program has been established involving 9 universities and CSIRO . The program has a strong focus on new membrane materials, but includes molecular modelling, performance testing, characterisation of membrane materials and feedwaters, thermodynamic analysis of membrane processes, membrane distillatio n and collation of data into a National Membrane Library. The cluster program runs for three years and it is anticipated that the program will produce new membrane materials with the potential to find commercial applicat ion.

112

MARCH 2008

Water

Each of the 8 su b-proj ects of the program are described in brief below, and contact details for the projects leaders given.

Molecular Modelling - Atomistic Simulation of Membranes and their Fouling T his project will imp rove o ur understanding of membrane technology at its most basic level. With the aim of determining an atomistic understand ing of the interaction of both water and dissolve species with the surfaces of membranes, these insights into the atomic scale details of membrane processes will be fed back into the design and modification of mem brane materials for improved performance and to decreased fou ling. This research is lead by C urtin University of Technology.

Julian Gale - j.gale@curtin.edu.au Molecular Modelling - Transport and Separation in Membranes This project aims to improve our understanding of the processes that go on at the molecular and mesoscopic levels of membranes. By developing a suite of computer simulation programs based on existing molecular dynam ics and smooched particle hydrodynamics programs, we intend to model transport and separation processes occurring in membranes, and improve our understanding of the relationships between the microscopic characteristics of membranes and their performance. This research is led by

RMIT. Peter Daivis - peter.daivis@rmit.edu.au Feedwater Characterisation T his project aims to develop innovative structural and surface characterisation technologies fo r organic and inorganic nanofiltracion and reverse osmosis membranes. As small deviations in pore size and surface conditions can have a significant impact on separation and filtration, the project will provide q uantitative correlation between the structure/surface and the performance of membranes, which will subsequently be used for the new materials development and molecular modelling. This research is led by Deakin University and Murdoch U niversity.

Lingxue Kong - lingxue.kong@deakin.edu. au Rob Trengrove R. Trengove@murdoch.edu.au Performance Testing This project aims to characterise the elements and mechanisms that cause membrane fo uling, first by laboratory-

Journal of the Australian Water Association

testing to provide baseline data fo r existing membrane materials and an experimental database, along with the development of a range of standard experimental protocols.

We will be exami ning the adhesion of organic compounds on membrane surfaces, and their interaction with ocher organ ic compounds, resting newly developed chlorine-resistant membranes and membrane d istillation membranes, as well as provid ing assessment of new membrane materials developed by our other Research C luster projects. This research is led by University of Melbourne.

M embrane distillation as a desalination process is also being researched. T his technology has seen the recent co mmercial development of higher fl ux membrane materials, and the engineeri ng of a high efficiency process is requi red fo r membrane d istillation to find niche applications in water recycling and desalination.

Sandra Kentish; Reverse osmosis testingsandraek@unimelb.edu. au Stephen Gray: Membrane distillation Stephen. Gray@vu.edu.au National Membrane Database and Energy Analysis T his proj ect aims co establish a national database containing fundame ntal in formation o n a suite of membranes and surfaces, and further, to broadly apply the concept of exergy to the performance of d ifferent membrane structures operating under d ifferent processes conditions.

Analysis of the thermodynamic targets for d ifferent applications will allow us to identify rhe membranes and applications char offer the most promise for improving the efficiency of desalination. This research is led by University of New South Wales.

Greg Leslie - g.leslie@unsw.edu.au New Membrane Materials - Biomimetics Th is project aims to apply the designs founds in nature to new innovations in membrane technology. Using microfluid ic flow visualisatio n methods coupled with numerical models, we will exami ne the role of pore architecture in the filtratio n mechanism of select species of the unicellular phytoplankton diatom, as well as other biological structural membrane filters. By investigation of the role of the pore structure in separation and filtration, the project aims co develop methods to mimic the d iatom pore structure in manufacture membranes. T his research is led by University of New South Wales.

Gary Rosengarten g. rosengarten@unsw.edu. au

technical features

New Membrane Materials Multifunctional RO Membranes This project looks to develop the next generation of membranes uti lising new fu nctional materials incorpo rated into polymer membrane materials to improve their salt selectivity, flu x and anti-fouling properties. New fun ctional materials will be developed based on zeolites, a class of crystalline ox ides that have been widely studied as molecular sieves in the processes of separation and adsorption for their ability to distinguish molecules based on their sizes and shapes, to deliver significa nt improvements in the salt rejection, flu x and thermal stability of polymer membranes.

T his research is led by Monash Un iversity.

Huanting Wa ng haunting. wang@eng. monash.edu.au New Membrane Materials - InorganicOrganic Nanocomposite Electrodialysis Membranes for High Performance Desalination This project will be using nanotechnology to develop a new suite of membranes fo r electrodialys is (ED) technology, which may lead to breakthrough technologies in costeffective and high efficient water recovery

systems. By incorporating oxide nanoparticles into ion-conducting polymers to form new nanocomposites, the project is expected to enhance the chemical stability of co mposite materials and reduce the fouling tendencies of membranes allowing ED technologies to compete as an efficient and cost-effective means of water generation. T his research is lead by University of Q ueensland.

Max Lu- max!u@uq.edu.au Complementary Research in CSIRO CSIRO is also conducting complementary research into the development of new membrane materials based on hybrid inorganic/organic mate rials (Manh H oang: Manh.Hoang@csiro.au). T his technology has shown promise fo r developing gas separation membranes and the transferability of this approach to desalination is being studied. T he use of carbon nanotubes in mem brane materials is also being investigated (Niall.Finn : Niall. Finn@csiro.au) as part of a broader program on carbon nanotube production and applications of carbon nanocubes. Both the hybrid membrane and carbo n nanotube membrane projects are undertaking in

collaboratio n with Australian universities.

CSIRO is using their materials characterisation (Anita. Hill: Anita.Hill@csiro.au) and water treatmen t expertise to collaborate with the other cluster parties to enhance the capabil ities of each sub-project.

Collaboration with overseas research organisations is also part of the cluster program, and the development of organic membranes with specifica lly shaped pores (h ttp://www.cs iro .au/ news/ Fan tasticPlastic. html) was achieved as a result of collaboration between CSIRO, the University of Texas, USA and Hanyang University, Korea. This collaboration is also producing greater oppo rtunities for the university cluster participants, as the membrane performance project wi ll inco rporate testing of new materials developed overseas.

The Author Prof Stephen Gray is Director, Institute for Sustainability and Innovation, Victoria University. Prof Gray is the cluster leader of the CSIRO Cluster on Advanced Membrane Technologies.

Email: Stephen.Gray@vu.edu.au

BIO MESH TUB ING The ideal filter medium 100% Australia owned and manufactured •

high surface area for bacterial production

• •

encourages excellent biological film growth modular design for easy maintenance and handling perfect for all aqua cultural environments cost effective, efficient, reliable chemically inert used in trickle filters, aeration and degassing systems, submerged filtration and treatment systems ideal as gutter guard, cooling tower fill pack and in oil separation units

• • • •

•

Acacia Filtration Systems Pty Ltd Tel: 02 9756 6077 Fax: 02 9756 0173 sales@acaciaproducts.com.au www.acaciaproducts.com.au Journal of the Australian Water Association

Water

MARCH 2008 113

technical features fereed paper

OPTIMISING MIXING IN FULL-SCALE MBRs: CFD MODELLING AND VALIDATION M W D Brannock, Y Wang, G Leslie Introduction Memb rane Bio reacrors (MBRs) have been successfully used in biological wastewater treatment co solve the peren nial problem of effective solids-liq uid separation .

Co nventional wastewater treatment systems use solids seed ing methods fo r separation, h owever these processes are difficult co control and can produce highly variable effluent q uality Qudd, 20 06). MBRs on che ocher hand use memb rane m icrofilcracion comb ined with biological treatmen t (see che process flow d iagram in Figure 1.1 - Left) co consiscencly produce very high quality effluent. For chis reason, M BRs are an excellent op portunity co simultaneously manage m un icipal wastewater while producing valuable recycled water co meet che current water crisis (Core et al., 2004).

Scientific literature is silent on the relationship between membrane configurations, such as flat sheets and hollow fib res, and the energy required fo r mixing in membrane bioreaccor (MBR) in wastewater treatment and reuse applications. Consequencly, the energy required co achieve the desired mixing in MBRs is frequencly overestimated and not op timised. T his is a significant p roblem as energy contribu tes co over 50 % of an MBR plane's operatin g coses and over 90% of energy usage is dedicated co mixing (Cote et al., 2004).

Methods I n light of the energy usage implications o f mixing, a fundamental analysis of mixing co nditions was initiated comb ining Residence T ime D iscribucion (RTD) measuremen ts (a "mixing fi ngerprint") with Computational Flu id Dynamics (CFO) .

RTDs are measu red using tracer study experiments. This involves pulsedosing an This paper won a Best Paper Award at the Young Water Professionals Conferen ce, Brisbane, February.

114 MARCH 2008

Water

0.300 Colour Map: Veloci ty (mi s)

0 Inlet

Figure 1. 1 (Left) Example of a n MBR process flow diagram w here th e process units are the 1) Anoxic zone 2) Aerobic zones 3) Membrane zones 4) De-aeration zone; (Right) Example of C FD output w ith stream lines coloured accord ing to velocity w here the design aspects are a) M ain inlet b) Mixer c) Overflow weird) Aerator e) Underflow .

inert tracer ac che inlet and measurement of che response at the ouclec. CFD however, is a m odelling method which can outp ut an RTD and predict how vessel features affect the mixing and energy usage; an example of qualitative CF O model output is given in Figu re 1.1 (Right) . CFD achieves chis through n umerical resolution of the fu ndamental equations o f fluid dynamics .

The CFO model developed h ere incorporates m ultiple phase hyd rodynamics (i.e. liquid mixture, aeration and sludge transport) and can pred ict pollutant removal through inclusion of b ioreactions.

achieving che optimum mixing ac the lowest possible energy inputs. Given the complexities of wastewater creacmenc, and the inherent p roblems of simulating system performance in che laboratory (i.e. scaleup), the work was cond ucted on two fu llscale MBR planes located in Australia (one using Flat Sheet membranes and the ocher H ollow Fib re).

These techniq ues are being used co assess the im plications of membrane configuration on energy consumption with the view of

To reach the decision-makers in the water field, you should consider advertising in Water Journal, the official journal of Australian Water Association.

For information on advertising rates, please contact Brian Rault at Hallmark Editions, Tel (03) 8534 5000 or email brian.rault@halledit.com.au

CPD will prove to be a valuable design and optimisation tool.

Journal of the Australian Water Associotion

Water Advertising

technical features

1.0 0.8 u10.s 0.4 0.2 0.0

0

Experimental Power Sink

1.0 0.8 uf0 .6 0.4 0.2 0.0

o

MBR

Mixer

7.1 kW

2.2 kW

Bioreactor Aeration

3.3kW

8.SkW

Membrane Aeration

29.5 kW

13.8 kW

Recirculation Pump

16.0 kW

18.SkW

Total

55.8 kW

42.9 kW

Experimental

- Simulation

0 .0

1.0

Flat Sheet Hollow Fibre

MBR

- Simulation

2.0 8

3.0

Figure 1.2 (Left) Experimental & CFD Modelled RTDs whe re the top figure is the Hollow Fibre MBR and the bottom is the Flat Sheet M BR; (Right) Comparison of energy usage of two MBRs.

Results and Discussion RTDs were generated for each MBR sire with high reproducibility (correlation between repeated crials being R2 > 0.98) and tracer recovery (>95%); both are important measures of tracer scudy success. As shown by Figure 1.2 (Left) rhe mixing characteristics were very similar for each sire and membrane type.

However, it was found chat the use of flat sheet mem branes is attended by a 20% increase in mixing energy (see Figure 1.2 - Right).

Regarding the CFD modelling of the MBRs, extremely good agreement between the CFD simulated RTDs and the experimentally derived RTDs was ach ieved (see Figure 1.2 - Left) . The correlation coefficient between model RT D and experimental RTD was R2 > 0.95. This shows chat CFD will prove co be a valuable design and optimisation cool for MBRs.

Conclusions The mixing within MBRs and its effect on energy consumption has been an insufficiently understood aspect of MBR design. A CPD-based model has been developed co optimise mixing energy in MBRs and assess the effect of membrane configuration. This research actively lowers the barriers co the use of MB R technology for creating water and recycling Ausrralia's valuable water supplies. "Recent work has enabled the inclusion of the Activated Sludge Model No. I, therefore pollutant removal

under di fferent design conditions (e.g. mixer sizes, baffle positions ere) can also be eval uated. The methodology and final results will be made available through the MBR-Necwork (www.mbr-nerwork.eu)."

~ SprayNozzle ENGINEERING Total Spraying Solutions

WE Can Help You Save WATER! Reduce your water usage through water saving solutions such as: • Water efficient spray nozzles • Water saving washdown equipment • Trigger gun nozzle systems

Acknowledgments This project is proudly supported by the [ncernarional Science Linkages programme established under the Australian Government's innovation statement, "Backing Australia's Ability".

This work was also supported by rhe European Commission under the 6 Framework Programme (AMEDEUS project, contract 0 I 8328).

The Authors Dr Matthew Brannock is a Senior Research Engi neer, email: m.brannock@u nsw.edu.au; Yuan Wang is a PhD scudenc; and Associate Professor Greg Leslie is Deputy Director at the UNESCO Centre for Membrane Science and Technology, The University of New South Wales, email: g.leslie@unsw.edu.au

• Patented tank cleaning nozzles Ideal for use during high water usage activities such as:

-Cleaning, sanitising and washdown -Cooling, quenching and dust suppression

References Cote, P., Masini, M. and Mourato, D . (2004) Comparison of membrane options for warer reuse and reclamation. Desalination

167, 1-11.

Judd, S. (2006) The MER book: principles and

applications ofmembrane bioreactors in water and wtZStewater n·eatment, Elsevier, Amsterdam; Bosron; London.

Journal of the Australian Water Association

Water

MARCH 2008 115

fereed paper

EFFECTIVENESS OF DOMESTIC RAINWATER TANKS RE Collins Summary

Introduction T his work was stimulated by local observations chat most domestic rainwater tanks are quite small and chat many such tanks are fed from one downpipe so chat only a small p roportion of the available rainwater enters the tank. This work initially aimed to determine the amount of water that could be saved with such small tanks and inefficient rainwater collection. It was subsequently extended to a more general assessment of the relevance of rainwater tanks in achieving significant savings in domestic water u se.

Sydney Wateri provides estimates of water consumption rates in typical "water efficient" households during water restrictions. Fo r medium sized suburban blocks (501 -700 m 2), these average consu mption rates for dwellings with l , 2, 4 and 6 occupants are respectively 242, 365, 550 and 699 L/day. Consumption is typ ically 2-3% greater than these values in

Most domestic rainwater tanks are not connected to household plumbing.

116 MARCH 2008

Tanks with ln -ho uso connoctlon

Tanks without ln-houso connoctlon

A method is described for estimating approximate contributions that rainwater tanks can make to total domestic water needs. Results for two locations in the Sydney region are presented in normalised form, such that they can be applied at other location s and to systems having differen t roof areas, tank volumes and number of occupants. As expected , sign ificant water savings are only achieved if the tank is large eno ugh, if rainwater is collected from a large enough roof area, and if the system is connected to the household plumbi ng.

Most do mestic rainwater tank systems are no t so connected. Responses to a questionnaire indicate chat most rainwater tanks are in stalled to provide water fo r restricted uses duri ng dry periods. This is quite different fro m the aim of rainwater tank rebate programs of reducing consumption of town water. Suggestions are made to improve the effe ctiveness of rainwater tank rebate programs.

Water

600

6000

°i

0 2000-3999 L

D 2000-3999 LI SOOO . •4000-6999L _ _ ___.

~

a >7oooL

j

I

i

f 4000 _,___ _ _ __

§

a >7000 L

g' 400

!1l.

!i z

• 400Q.6999 L

~

€ ;

500

3000 -t- - -- --t 2000 -1---- - - --t

j

300

~

200

,,

~

§

1000 +-- -I

z

100

0

2002/3

2003/4

2004/5

2005/6

2006/7

2002/3

2003/4

200415

Yoar

2005/6

2006/7

Year

Figure 1. Number of rainwater tanks that have received a subsidy from Sydney Water for the years 2002/3 to 2006/7 inclusive, for different ranges of tank sizes.

The two parts of the figu re show separately the number of tanks without, and with connection to the house plumbing . Note that the scale is expanded by a factor of l 0 for the numbers of tanks with in-house connection. the warmer months (November - April), and up to 4% less from May co July. Water co nsumption varies slightly (up to ±4%) with the size of th e block.

Recent papers by Coombes and Kuczeraii, Coombes and Kozarovskiiii, and Lucas et aJv discuss the p ossible contribution of rainwater to domestic use at many locations and for several roof areas, tank volumes and number of occupants. These authors correctly argue that significant water savings will o nly be achieved if the tank is connected to the domestic plumbing so that the stored rainwater is continuously drawn down. I t is assumed that water from the tank is used for toilet flushing, laundry and hot water, and for all outdoor p urposes.

Assu med usage rates are depend ent on the location and month. The tank is fed with mains water wh en the level is low.

Calculated water savings are presented from a simulated rain model" applied at six minute in tervals, and compared with savings estimated from ocher modell ing approaches.

Figure 1 shows the sizes of rainwater tanks that have been installed in the Sydney Water rainwater tank rebate programvi,vii,viii,ix,x to the end of the 2006/7 year. Nearly half (4 5%) of all subsidised tanks have a capacity less chat 4000 L. About 15% of subsidised tanks are greater than 7000 L in capacity. Only about 7% of

Journal of the Austrolian Water Association

these tanks are connected to the domestic plumbing.

Analysis of Contribution from Rainwater to Domestic Water Use In chis section a simple method is described for determining the approximate co ntribution chat rai nwater could make to total domestic water need s.

Method of analysis The analysis is based on three simplifying assu mptions. Firstly, the consu mption of water is assumed to be the same on every day. T his assumption ignores the fac e chat less water would be used outdoors o n average in rainy period s, and more during periods of low rainfall. It would be straightforward to extend the analysis to a variable rate of water use, as is done in ocher worki_ Secondly, it is assumed that stored rainwater is used wh enever it is available. T his assumption ignores the possibil ity chat at times a h ouseh older migh t choose not to use rainwater from the tank. Finally, all rainwater chat falls on the defined roof area is assumed to feed into the tank. This assumption ignores water losses d ue to roof evaporation and firstfl ush devices at every rain event, and from overflowing gutters d uring h eavy rain.

The input data used in the analysis are the daily rain fall records over the period 2000 to 2006 inclusive at Sydney city and

technical teatures II. refereed paper

Penrith which have average annual rai nfalls of 1202 mm and 784 mm respectively.

T hese data were ob tained from Bureau of Meteorology daily rainfall recordsxi and are complete. For an indi vidual calculation in the analysis, specific values are assumed for each of rhree parameters: dai ly water use, tank volume and roof area. On any particular day, the calculation begins with the volume of rainwater remaining in the tank fro m the previous day. Firstly, chis volume is red uced by the assumed daily water use or, if less water remains in th e rank, to zero. This quantity is added ro the cumulative contribution from rain for chat particular year. T he volume of rainwater (in L) rim fell on the roof is calculated by mul tiplying the roof area (in 111 2) by the amount of rain (in mm) on rhac day. The volume of water in rhe rank is then increased by chis amount, to an upper limit of the capacity of che rank. The calculation is rhen repeated for the next day and conti nued until the end of rhe period.

Virtually identical annual water savings are obtained if the order of withd rawi ng water from rhe tank and adding rainwater is reversed. For very large tanks, rhe estim ates of water savings fo r rhe first year (2000) can be significantly affected by the assumed volume of water in the tank before the first day of the calculation (Day I of 2000). T he results obtained for the yea r 2000 are therefore ignored here, except for the vo lu me of water in the rank at the beginning of 200 I. Results from the analysis

Figure 2 presents typical data obtained from the analysis at Sydney city fo r one year (200 I) for which rhe coral rainfall was 1359 mm. Figure 2(a) shows the cumulati ve rai nfall at chis location over rhe year. The data in Figure 2(6) and (c) are obtained from rhe analysis fo r an assumed water use race of 400 L/day, and for all combinations of cwo roof areas (50 m2 and 100 111 2) and two rank volumes (4000 L an d I 2000 L). Figure 2(6) and 2(c) respectively show che volume of warer in the rank and the cumulative rainwater conrriburion to water use during 200 I , for all combinations of each of these parameters. For comparison, the cumulative total water use during the year is also shown in Figure 2(c). These dara illustrate the benefits of a larger roof area and larger storage volume.

The calculated water savings in any particular year as a function of the rainwater collected in char year can be normalised in terms of rhe annual warer usage and rhe number of days' storage in rhe rank. Figure 3 shows rhe proportion of

1400 1200

I ~ C

(a)

1000 800

'f

., >

~

600

-; E ::,

400

u

200 0 51

101

151

201

251

301

351

Day Number 12000 _

10000

,.=C:

!!

8000

-

Roof area = 100 m2, Tank \Olume = 12000 L

-

Roof area= 50 m2, Tank IO!ume = 12000 L

-

Roof area = 100 m2, Tank \Olume = 4000 L

-

Roof area = 50 m2, Tank 10lume = 4 000 L

,,.5

~., ., j

6000 4000

C:

¡;.

II'.

2000 0 51

101

151

201

251

301

351

Day Number 150000

,--~---------------,----,-----,r----,........., (c)

,,;J:..

120000

$

90000

.,::,

;

=;::;

: : :e: ut;oa::~~a~:a~l~L~e = 12000 L

-

Roof area = 50 m2, Tank \Olume = 12000 L Roof area= 100 m2, Tank \Olume = 4000 L

-

~ of area = 50 ~ 2, Tank \Ol~me = 4000 L ,

~

I

'

C:

60000

.i!:

1'::, §

30000

(.J

51

101

151

201

251

301

351

Day Number

Figure 2. Daily results from the analys is of rai nwater tan k systems in Syd ney city fo r the year 200 l . (a) Cumulative rainfall over the year 200 l; (b) Da ily volu me of stored wa ter in the tank, for all combinati ons of roof areas of 50 m2 and 100 m2, and storage volu mes of 4000 L a nd 12000 L. The assumed water use is 400 L/da y; (c) Cumulative volume of water delivered from the tank over the year for all combinations of these parameters. The cumulative total water use during th e year is also shown in th is figure.

coral warer use rhar can be suppl ied from the rank for all years in the period 20012006 and fo r borh locations modelled in rhis work. The results for different years, and for both locations, depart lirrle from well deft ned trends wi rh scatter due co randomness associated with rai n evenrs, and warer remaining in the rank at rhe end of yearly periods. Ir will be of interest co determine the extent co which rhis normalisation approach, based on average daily water consumption, is useful for presenting data obtained ar orher locations,

and from more derailed models incorporating shorter rime srepsxi i, ri medependent water demand, and losses from first fl ush devices, roof evaporati o n and guttering overflow.

Several co mments can be made about rhese results: 1. For roof areas such rhar rhe average rate of collection of rainwater is about the same as rhe water consumption rare, it is possible to obtain slightly more than half of coral domestic water needs from rain if rhe rank volume scores rhe equivalent co 1 0 days'

Journal of the Australian Water Association

Water

MARCH 2008 117

technical features Ill refereed paper

water use. Tanks with capacity greater than ~30 days' use can supply over 70% of household water needs for such roof areas.

2. For small tanks having volumes corresponding co only a few days' water use, che fractional contribution from rainwater to total water needs increases with tank capacity if che roof area is large enough. This occurs because small tanks fill rapidly during most significant rainfall events.

3. When the average rate of collection of rainwater is very much less than the race of consumption of water so chat the bulk of che water is supplied from the mains, almost all of the collected rainwater is util ised by the household and rhe absolute amount of water saved increases little with increased race of water use or tank capacity.

This behaviour, illustrated in the dara of Figure 2(6) for che second half of the year, occurs because the tank usually empties in much fewer days than the average incerval between significant rainfall events.

As an example, consider rhe case of a family of four chat uses 550 L/day. For roof areas of 85 m2 ac Sydney city, and 130 m2 for Penrith, an average volume of rainwater equal to about one half of che coral annual water use would be collected. Achieving such roof catchment areas would normally require chat multiple downpipes feed into the tank. Figure 3 shows chat chis amount of rainwater would permit 25% savings of town water use wich a small tank having a capacity equivalent co about 3 days' use (~ 1500 L), and ~50% (essentially all of che water collected) for a much larger tank with capacity equivalent to ~30 days' use (~ 15000 L). If che roof area is so large chat the annual volume of rainwater collected is equal to the total annual water use, the same respective contributions can be achieved with tanks having capacity equivalent to only 1.5 days' use (820 L) and 10 days' use (~5500 L). However, the necessary roof catchment areas of 165 m2 for Sydney city and 260 m2 for Penrith respectively are unlikely co be achievable on most suburban houses.

Motivation for Installing Domestic Rainwater Tanks Governments and water supply authorities

The stated intention of governments and water supply authorities in all Australian cities chat encourage the insrallacion of domestic rainwater systems is to reduce the use of town water. For example, che Sydney Water tank rebate licerature•iii states: "rainwater tanks help conserve valuable drinking water .. ..". The Sydney Water 118 MARCH 2008

Water

'c'

""'

:, <D ~ 0 "0

is i ~ o o

...- "'E

·;.

~

....

••

• • • • • •• • • •

.,

••

'. ... .. ... .

.·'...... • • .. .•.. •... • ~

e ,..

C

• •

• • • • • • • •• . .~ •• • •

,1 . . . ,.........

0.6

~

0.3 0.2

•••

0.1

•

-

I • Storage= 1 day's use

E i,'.l 0.4

S .t:~ I-~

• •

•

• Storage = 3 days' use

0.7

•- -

~

I "' Storage= 10 days' use

0.8

~ e 0.5 g '::

-

-

Storage= 100 days' use • Storage= 30 days' use

1•

0.9

I

•

•

•

. ..

•

a

• •

....

..

• • •

•

•

•

•

•

•

0.2

0.4

0.6

I I

0 0

I

1.2

0.8

1.4

1.6

1.8

2

(Rain wate r from roof / Total water used) In ea c h year from 2001 to 2006

Figure 3 . Results of the analysis for ra inw ater tan k systems i n Sydney city and Penrith

for the years 200 1.2006, based on actua l daily rainfall recording s at these loca tions. This figure shows the fractional co ntribution from ra inw ater to total water use in any year. The hori zontal axi s gives values of the volume of rain that fe ll o n the roof in each yea r, expressed as a multi ple of the total water use in that year.

Conservation and Recycling Implementation Report 2002-2003"i provides a rationalisation for rhe rai nwater tank rebate program: "Various research studies have shown chat rhe capture and use of roof water in rainwater ranks can reduce che demand on reticulated water supplies.

Typically savings of between 20 and 50% can be achieved, depending on the end uses of the scored water, tank size, roof catchment area, and rainfall pattern . .... tank yields are increased when scored water is used for indoor as well as outdoor use."

This Report also states rhe aims of the rebate program: "The rainwater tank rebate program aims to assess whether a rebate incentive can stimulate an increase in tank installations . ... The rebate also aims to encourage rank purchasers to inscall largersize tanks and to connect chem ei cher to their toilets or washi ng machines." The data in Figure 1 indicate char the program has achieved the first aim, but not the second.

It is unfortunate chat the rebate program did not also aim to validate the resulrs of the (unreferencedvi) research studies showing chat rainwater tanks can reduce the consumption of town water "typically (by) between 20 and 50%". Water savings of chis magnitude are achievable by only a small proportion of currently installed rainwater tank systems - those chat are co nnected co the domestic plumbing.

Information on the devices that are connected to the tank is sought in che Application Form for che Sydney Water rank rebate'iv_ Presumably Sydney Water could access meter readings to determine whether significant water savings have occurred in households with rain water

Journal of the Australian Water Association

tanks. However, the validation would require the critical parameter of roof catchment area to be determined, since chis information is not sought in applications for a tank rebace•iv_ Ocher Sydney Water liceraturexv contains the recommendation chat rainwater tanks should be used: "in conjunction with water efficient devices like dual flush toilets ... " This is sensible advice because, as is well known in che energy field, conservation is the quickest and least costly method of reducing consumption of any quantity.

The NSW Government's BASIX (Building Sustainability Index) programxvi is aimed at reducing the environmental impact of dwellings. Conservation of water is an important BASIX strategy. T he BASIX liceracure'vii encourages che installation of rainwater tanks stating chat chey will "Reduce your consumption of mains water "

Sydney Waterxv provides guidance for deciding on che size of a rainwater tank: "the larger the tank, the more rainwater can be captured for use during dry periods. A minimum tank size of 5,000 litres is desirable." However, both Gosford City Councilxviii and Hornsby Shire Councilxix impose an upper limit of 10,000 Lon the size of tanks. This is inconsistent with achieving the largest possible reductions in the use of town water.

Apart from the reference to research studies quoted above"i, there is liccle indication in the literature about the magnitude of the overall reduction in che use of town water chat is sought through encouraging the installation of domestic rainwater tanks. le may cake many years before there is

lecnn1ca1 realures II. refereed paper

substantial (say, ~50%) penetration of water tanks into the domestic market.

Large water savi ngs could therefore not occur in the short term.

Householders

In the absence of any published data about the motivations of householders for installing rainwater tanks, a question naire seeking this information was di stributed by the author in mid-2007 to householders mostly residing on the Central Coast of NSW. T he question naire also sought information on tank sizes. Responses were received fro m 46 tank owners, of whom 40 had tanks sized 2000 Lor greater. On ly three respondents had tanks greater than 5000 Lin capacity. Within the limitations of the small number of respondents, rhe distribution of tank sizes is consistent with those in rhe Sydney Water tank rebate data shown in Figure 1.

Very few (3/46) of the respondents stared chat they had installed thei r rainwater tanks to make a contribution towards reducing the use of town water. The motivation of most was to have water available fo r purposes that are restricted fo r town water, such as (predom inantly) watering gardens, and also washing cars and windows, and topping-up swimm ing pools.

would be much more informative to such people.

The most recent Water Conservation and Recycling Implementation Report by Sydney Wate1¡xxi scares rhar rhe rank rebate program has paid "a total of 36,842 rebates si nce rhe program commenced in October 2002.

The program has saved an estimated 1485 ML/year." These claimed water savings greatly overestimate what is possible with these systems. Even if these savings are assumed to be current rather than, as implied, over the duration of the tank rebate program, rhey would average ~40,000 L/year for each system. Based on other data given in this Reportxxi, chis is ~ 19% of average residential water demand.

The analysis presented in this paper shows char savings of ch is magnitude are achievable by the few (~ 7%) rainwater tank systems char are connected to rhe household plumbing. However, the savi ngs from the majority of these systems char are nor so connected would be very much less than chis. Clai med water savings from rainwater tanks should be justified on the basis of reasonable argument or referenced method. Ideally, in order to evaluate che effectiveness of the Sydney Water rainwater

tank subsidy program an d to validate any claimed savings, systematic monitoring should be occurring of water consumption in a statistically significant sample of households having rainwater rank systems.

The results of such monitoring should be reported in a transparent way. O nly very limited monitoring of this type has been reportedxxii to date.

Strategy for provision of back-up water

Savi ng a substantial amount of town water requires the rank to be connected to the domestic plumbing so chat stored rainwater is drawn down continuously. O n the other hand, most householders have installed their rai nwater tanks to provide water for purposes that are restricted or forbidden by regulation - predom inantly for maintai ning gardens. T hey therefore try to retain as much stored rainwater as possib le in the rank fo r these purposes, and nor to use this water fo r uses that can be served by town water. The motivations and actions of rhe two stakeholders in rhis matter co uld hardly be more diffe rent.

For virtually all systems, there will inevitably be periods when town water is needed to maintain the operation of the household. The strategy discussed by

Although most (33/40) of the respondents with ranks sized 2000 Land above stated that their system had a pump, very few (3/33) of these systems were co nnected to the domestic plumbing. This is consistent with the Syd ney Water rank rebate data.

Slightly more than half (23/40) of the eligible respondents seated that they had obtained a subsidy when they installed their system. However, few (2/23) of these stated chat the subsidy had influenced their decision to insrall the rainwater tank. Some householders had not sought a subsidy because they did not want to be restricted in the use of their tank water. Because restrictions exist on the way in which scored rainwater may be usedxx, this could be a reasonable concern.

Discussion Reporting and validating water savings from rainwater tanks

Claimed water savings from rhe Sydney water rank rebate program are reportedvi,vii,viii,ix,xxi in terms of the total amount of water that has been saved. Dara presented in this way are not particularly useful in demonstrating the effectiveness of rai nwater ranks to individual householders who make rhe decision to install a tank.

Reporting savings as proportional reductions achieved in individual household water use Journal of the Australian Water Association

Water

MARCH 2008 119

technical teatures I. ,cfereed paper

Coombes and Kuczeraii of topping up the tank with town water when the level falls to a defined value is attractive in an operational sense. The house is always fed with water fro m the tank, pressurised by a co nventional pump. A float-operated valve controls rhe feed of town water into the tank when the level is low.

Systems of chis design have two significant disadvantages, however. Firstly, the water stored in the tank is no longer sourced only from rain. Much of the time chis water comes from the mains and its use is restricted. It would be a signi ficant disincentive to householders to expend a substantial amount of their own money on a storage tank and not to have the benefos of a small amou nt of water to use for restricted purposes during dry periods.

A second disadvantage of systems wirh topup by town water is the energy used by the pressurising pu mp during dry periods which would nor be required if che house is connected directly co mains water at such times. Th is disadvantage could be overcome if the rainwater systems were co nnected to the do mestic plumbi ng with automatic valvesxxiii, as requ ired by the South Australian Department of Water, Land and Biodiversity Conservacion""iv.

Conclusions This paper presents a simple analysis utilising daily rainfall records of the approxi mate annual contribution chat rainwater could make to total domestic water needs. T he analysis assumes a constant race of water use permitting the calculated ann ual water savings to be presented in normalised form so that they can be applied to systems with a wide range of design parameters. Ic would be interesting to determine the extent co which chis normalising process is useful for presenting data obtained ac climatically diffe rent locations, and fro m more sophisticated modell ing approaches char assume a variable rate of water use and utilise shorter modelling time seeps.

The proportion of domestic water needs chat can be replaced by rainwater is strongly dependent on the amount of rainwater collected, che size of che storage tank, and the race of water use in che household. In many households, savings of about 25% of total water use are achievable with the quire small ranks chat are currently being installed (up to ~5000 L) provided chat che roof area is large enough. Larger savings (~50% or more) require very large roof areas and often substantially larger tanks than are commonly installed at present. In all cases, significant savings will only occur if che rai nwater rank systems are connected to che

120

MARCH 2008

Water

domestic plumb ing. Mose current rai nwater rank systems are not so connected and therefore will therefore not achieve rhe seated intention of authorities in subsidising chem - co reduce significantly che domestic consumption of town water.

The installation of rainwater tank systems requires a substantial investment by the householder. It appears that such systems may only be attractive to most householders if chey are configured so chat at lease some of che stored rainwater can be used for purposes chat are restricted during dry periods. Ir is reco mmended chat rainwater ranks be connected so char che dwell ing is supplied by rainwater from rhe tank when it is available, and di rectly from rhe mains when the water level in the tank falls co a defined low level. For such systems, there need be no restriction on rhe way that scored rainwater can be used, so that the desires of householders and of water supply authorities can be simultaneously mer.

Connecting the systems in this way will lead to energy savi ngs compared with rhe topping-up method.

It is recommended that systematic mon itoring be undertaken of a statistically sign ificant number of rainwater tank systems chat are connected to the domestic plumbing. The resulcs from such monitoring will inform chinking relating co future policy initiatives on rainwater tanks, and will demonstrate the extent to which savings are consistent with predictive models.

Acknowledgments The author thanks Manfred Lenzen, Ian Wright and two referees fo r constructive comments on chis paper.

The Author Emeritus Professor Richard Collins (r.collins@exemail.com.au ) retired at rhe end of 2000 as Head of rhe School of Physics at the University of Sydney after a 40 year career as a scientist in industry, government and academia. For che past 25 years, he and his family have lived on properties with water supplied only from rainwater tanks.

References 1.

Sydney Warer, hrrp://www.syd neywarer. com.au/SavingWarer/InYourH ome (2007)

11.

Coombes P. J. and Kuczera G., Analysis of the performance of rainwater tanks in Australian capital cities. The lnstirution of Engineers Australia 28th International H ydrology and Warer Resources Symposium, Wollongong NSW (10- 14 November 2003)

111.

Coombes P J and Kozarovski P, Development of a regional model ro

Journal of the Australian Water Association

understand hydrological and economic benefits of rainwater ranks across New South Wales. 29th H ydrology and Water Resources Symposium, 2005 (2005) 1v.

Lucas S A, Coombes P J, Hardy M J and Geary P M, A comparative study of common uses of selected modelling rools for evaluating rainwater harvesting strategies.

Conference Proceedings of Water 2006, Grand Hyatt Conference Centre, Auckland, NZ (2006)

v.

H eneker T. M., Lambert M. F. and Kuczera G ., A point rainfall model for riskbased design. journal of Hydrology 247 pp 54-71 (200 I)

v1.

Sydney Water, 2002-2003 Water Conservation and Recycling Implementation Report (2003)

v11. Sydney Water, 2003-2004 Water Conservation and Recycling l m plemenration Report (2004) v111. Sydney Water, 2004-2005 Water Conservacion and Recycling lmple menrarion Report (2005) 1x.

Syd ney W ater, 2005-2006 Wacer Conservation and Recycling Implementation Report (2006)

x.

2006-2007 rank rebate dara supplied by Sydney Water

xi.

http://www.bom.gov.au

xn.

Coombes P J and Barry M E, The effect of seleccion of time steps and average assumptions on che continuous simulat ion of rainwater harvesting strategies. Water Science and Technology, 55 pp 125-133 (2007)

x111. Sydney Water, Rainwacer T anks - Rebates. http://www.sydneywarer.com .au (undated) xiv. Syd ney Water, Application for a rainwater rank rebate. h ttp://www.sydneywater.com.au (undated) xv.

Sydney Water, Rainwater ranks. http://www.sydneywater.com.au (undated)

xvi. http://www.basix.nsw.gov.au xvii. WacerWise NSW, Ger tanked and save a bundle. http:/ /www.waterwisensw.com .au (undated) xviii. Gosford City Council, Rainwater T anks Residential Rainwater Rebate Program. http://www.gosford.nsw.gov.au (2006) xix. Hornsby Shire Council, Water Efficiency Rainwater Tanks. http://www.hornsby.nsw.gov.au (undated) xx.

Water Restrictions for Internally Connected Rainwater Tanks. Gosford City Council and Wyong Shire Council (2006)

xx1. Sydney Water, 2006-2007 Water Conservation and Recycling Implementation Report (2007) xxii. Coombes P J, Kuczera G and Kalma J D, Economic, water quantity and quality impacts from the use of a rainwater tank in the inner city. Australian journal of Water Resources, 7 pp I 11-120 (2003) xxiii. See for example: http://www.ongawaterswitch.com.au; http://www.capindustries.com.au xxiv. South Australian Department ofWater, Land and Biodiversity Conservation. Rainwater T ank Policy -- Quick Reference Summary. http://www.dwlbc.sa.gov.au (2005)

-cecnn1ca1 Tea-cures

QUEENSLAND'S WESTERN CORRIDOR RECYCLED WATER PROJECT: ALLIANCES IN ACTION C Berry Summary This article outlines the Western Corrid~r Recycled Water Project - one of SouthEase Queensland's drought-proofing projects - and couches on the delivery of part of the scheme through the Western Pipeline Alliance.

DeCEl"T1014 IAY

Introduction With water supply levels at historic low levels, South-East Queensland (SEQ), as with much of Australia, is in the grips of a seven year dry period. Significant population growth, combined with the effects of climate change, is straining the ex isting water infrastrucrnre systems. SEQ has rhe largest growth race of any area in Australia with the population estimated co be rising at around 2.5 per cent per annum.

As part of a number of demand management measures, level six water restrictions are in place th roughout most of SEQ targeting residential water use co I 40 litres per perso n, per day. As well as demand management, the Queensland Government has instigated an ambitious program co "drought proof SEQ".

Western Corridor Recycled Water Project Australia's largest recycled water scheme and the third largest advanced water treatment project in the world, the $2.4 billion Western Corridor Recycled Water

A no-blame culture predominated the alliance.

Luggagt Poire lttaimtnc Pt rit

-.-.-.

Figure 1. Map of the area.

Project (WCRWP) is a visionary undertaking chat will use seco ndary created wastewater from Brisbane and Ipswich co produce purified recycled water for SouthEast Queensland. The project comprises a netwo rk of pipelines, storage tanks/reservoirs and pumping stations chat transports treated recycled water co various customers via three new Advanced Water Treatment Plants from the southern and western areas of Brisbane (Figure I).

Involving significant challenges in terms of technology and rhe management and regulation of a recycled water scheme, the WCRWP will deliver up co 232 megalirres of purified recycled water daily co two major power stations, industry and agriculrnre as well as co the public water

supply via the Wivenhoe Dam system co supplement Sou ch-Ease Queensland drinking suppl ies.

Central co the project is construction of more than 200 kilometres of underground pipelines and th ree advanced water treatment plants. The WCRWP will collect secondary treated wastewater from Brisbane's major treatment plants at Luggage Point, Gibson island, Bundamba, Goodna, Waco! and Oxley. The treated wastewater will then be further treated within three new Advanced Water T rearmen r Plan rs (AWTPs). The AWTPs will treat the wastewater ro the highest standard through a multi-barrier treatment system incl uding microfiltrarion (MF) and reverse osmosis (RO). The first stage • • • • • • •

GEORG FISCHER PIPING SYSTEMS

Ph: 07 3390 7166

Fax: 07 3390 7177

Email: info@allflowsupply.com.au Web: www.allflowsupply.com.au

Journal of the Australian Water Association

Water

Automation Data Logging Pressure Control PH/ ORP Flow Meters Plastic Piping Systems Automated Meter Reading Equipment • Under Pressure Tapping Equipment

MARCH 2008 121

technical features

AWTP came on line in August 2007 with remai ning stages fo llowing in 2008.

As well as meeting the short and long-term water needs of SEQ, the WCRWP seeks to utilise world best practice through sustainable pri nciples. T his has been already seen through reduced pressure on the region's dams and waterways, and reduced nutrient discharge inro the major river catchments and coastal areas.

Early design and development of che business case started in 2006 by consultants GHD with sub consultants Black and Veatch and Su nWarer who developed an optimised treannent and distribution strategy ro meet the variable demands of the industrial and commercial end users of the water.

The project is being delivered through five project alliances formed ro design and construct pipelines and advanced water treatment plants. Veolia Water Australia will operate the scheme once construction is complete. T he Western Pipeline Alliance (WPA) was formed in Ocrober 2006 between GHD and contracrors McConnell Dowell and Abigroup ro design and co nsrrucc rhe secrion of rhe projecr west of the AWT P at Bundamba.

The Western Pipeline The Western Pipeline involves pumping water from the AWTP at Bundamba and from the other AWTPs through a single pipeline ro a booster pump station 32km away at Lowood, via a Balance Tank. From the Lowood pump station the treated water is then delivered ro users at Caboonbah (at the top of W ivenhoe Dam).

Figure 2 . Schematic of the distri bution network, i nclud ing the two Power Stations.

The inirial 32km of pipel ine fro m Bundamba ro Lowood was 145 1mm dia x! 0mm MSCL pipe. From Lowood Pumping Station che pipeline reduces ro 1000mm GRP for a fu rther 49km ro irs connection wirh the end user, Tarong Power Srarion ar Caboonbah. A second 1200mm GRP pipe was also installed for 15 km between Lowood and the water srorage reservoir at Wivenhoe Dam.

A significant challenge ro rhe Western Pipeline Alliance was the timely procurement of materials and plant. Long lead items were identified by the alliance as part of the early design works in the firs t few weeks of the project. Major valves and pumps having lead times of over 12 mo n ths have been sourced from around the world, including T urkey, Sweden, Germany, Scotland, Canada and China.

The fi rst stage of rhe Bundamba AWT P was completed in August 2007 and commenced delivery of treated water ro the Swanbank Power Station.

From the srart of the WPA in early October 2006, che design has been fast cracked all owi ng the procurement of pipe and first installation to begin in late January 2007 .

By December 2007, more than 48km of I 000mm GRP pipe and 20km of 145 1m m MSCL pipe had been laid, with commissioning activities due to commence in February 2008.

Key aspects to the success of the project were that the WPA structure allowed the designers ro work with the consrructio n reams, reduci ng rhe lead ti me on pipe procuremenr and insrallario n ro a few weeks.

The project involves more rhan 95km of large diameter pipeline, two major pump ing stations and two balance ranks.

The Bu ndamba West Pumping Station forms a focal point in rhe WCRWP rransfer system delivering water ro end users. T he pumpi ng starion is divided inro cwo principal pump secs, the booster pumps receiving rreared RO water from rhe outlying AWTP and che transfer pumps lifting water from rhe rreaced water tanks ar Bundamba. These pumps pump inro a common header manifold and inro rhe single 32km 1451MSCL pipe ro rhe Lowood Balance Tank. Maximum pump and surge pressures in che pipeline are 290m. The cwo pump sers at the Wesr Pumping Srarion consisr of: • I 540kW booster pump morors: rwo dury + one standby, variable speed drive; • 1365 kW transfer pump morors: three duty+ one standby, soft starters.

122 MARCH 2008

Water

The majority of the pipeline nerwork is being constructed from rubber ring-jointed, mild steel concrete lined (MSCL) pipe.

Current demand for MSCL pipe is high, leading ro long delivery times and high procuremenr cosrs. Demand was also compounded by rhe Queensland Government's declaration of rhe WCRWP as a 'fast track' project. The use of alternative pipe materials such as GRP was proposed, in part, ro manage the demand on a single supplier and provide additional consrruction flexibility.

Journal of the Australian Water Association

T he use of the specialist software, PC pipeline, provided rhe abi li ty ro respond rapidly to changes on site an d produce a high level of derail for the pipe procurement and construction teams.

Other innovative solutions adopted duri ng rhe project included: • use of ' robotic' trench shields • elimination of th rust blocks and minimising welding • use of alternative materials and suppliers to more effectively manage supply chain risks • strategic modell ing co minimise hydraulic co nscra 1nrs • use of alrernarive materials such as GRP.

technical features

Pressure vs Chainage 300

I

..J

-

Pipeline Profile

-

Surge Profile

-

Mitigated Surge Profile

200

et:

] 100

0 0 The design team from G HD had to implement strategies fo r meeting the fast crack project schedule, one of which included improving the delivery time fo r MSCL pipe.

Conventional design of MSCL pipelines for water involves the specification of a maximum allowable operating pressure (MAOP), wh ich is derived from the design pressure plus an allowance for surge. An allowa nce of 25 per cent is usually specified. This often leads to a conservative wall thickness specification

20000

30000

Chalnage (m)

Figure 3. Pressure Surge Profiles: M itigated v's Unm itigated .

Pressure vs Time Response 2

The team foresaw chat the procurement of pipe would be a major risk and therefore implemented some specific cost and time saving strategies in the design.

One of chose strategies involved a computational surge analysis for the pipeline to determine the severity of water hammer and options to mitigate surge pressures.

10000

, ,- • "

1 .75 1.5

i

1 .25

I

f

;>

0.75 0.5 0.25

-

'1

I

0 -0.25

2250 2000 1750 1500 1250 1000

& ,:!.

-

Velocity

750 500 250

j

-

Pressure

0 -250

-0.5

-500

-0.75

-750 -1000

-1 -1 .25 50

100

150

200

-1250 250

Tire (S)

Figure 4. Cyclic Surge Pressures: Unmi tigated.

CONTr~INMENT Minimal Maintenance Systems

~11l PERtlASTORE* ~ r fitll.YiNJ±Wi:1 TANKS & SILOS

4/8 Leighton Place PO Box 240 Hornsby NSW 2077 Toll Free (from land lines) 1800 25 30 40 Fax 02 9477 7363

•

Journal of the Australian Water Association

Water

MARCH 2008 123

tecnn1ca1 teatures

and a pressure raring char is u nnecessarily high. Through the use of computational surge analysis, designers can instead optimise pipe wall thickness by the utilisation o f specific surge mitigating equipment. This could include the strategic placemen t of surge vessels, pressure relief valves or ch eck valves, as concl uded from the analysis.

Using a surge analysis software program known as Hytran , various operational cases were modelled. For each operating case, the results were tabulated and graph ed to illustrate pressure vs rime and pressure vs chai nage. Figure 3 is a plot that compares pressure surge profiles for mitigated and unmitigated surge scenarios and a pressuretime response plot without surge mitigation .

The results of the co mputational surge analysis indicated that a MSCL pipe with 16mm steel wall th ickness would be requ ired to contain the level of surge pressures produced during a power fa ilu re event. The conventio nal WT sizing

procedure verified this result, indicating a mini m um requi red steel thickness of I 6mm. Pressure vs time plots, as in Figure 4, also indicated a significant level of cyclic surge pressures, providing concern about long-term fatigu e fa ilure in the steel p ipe.

(Design life of the pipeline is 75 years.)

Using the computational surge analysis software, different methods of surge mitigation were modelled and analysed to d etermine the most effective solution for surge pressure reduction including the installation of surge vessels and p ressure relief valves. H owever the most successful option, as in most cases, mrned out to be the simplest. An in-line non-return valve positioned approximately midway along the pipeline section proved to mitigate the surge su fficiently to allow the pipe steel wall thickness to be reduced to 10mm. The non-return valve acted as a partition in the p ipeline, separati ng the upstream and downstream sections of water. In doing so the magnimde of the surge p ressure was considerably reduced. Cyclic pressure loading on the pipe was also reduced, minimising the risk of fatigue failure in the pipe.

T he use of alliance contracting was seen as another key strategy in the successful delivery of the project in the timeframes imposed by emergency drought legislation.

(Refer Box) .

THE WESTERN PIPELINE ALLIANCE Th e Western Pipeline Alliance consists of McConnell Dowell , Abigroup, GHD and pipeline subcontractors, Ecocivil. The allian ce is construct ing an 80 km pipeline, and associated pumping stations and structures, between Bundamba and Caboonbah as part of the 200 km Western Corridor Recycled Water Project. The WP A and subcontractor C LM are also con structing the Towns Water Supply Scheme; a 50 km network of pipelines that will feed regional towns from Esk to Yarraman with alternative drinking water supplies.

G HD is an international professional services com pany with a network of offices throughout Australia, New Zealand, Asia, the Middle East, U nited Kingdom and the Americas. GHD's architects, engi neers, drafters, planners, scientists, management consultants and economists service the

MARCH 2008

McConnell Dowell McConnell Dowell is an Australian-based multi-discipline engineering, construction and services company operating throughout Australia, Asia, New Zealand, the Pacific Islands and th e Middle East.

The company delivers civil, marine, electrical, mech anical, pipelines and tunnel construction solutions.

Abigroup

GHD

124

infrastructure, mining and industry, defence, property and building and environment sectors.

Water

Abigroup Limited is o ne of Australia's largest engi neering, building and asset development companies. Abigrou p invests in infrastructure projects, u ndertakes works in all areas of engin eering and has extensive building experience across a range of development styles.

Journal of the Australian Water Association

Alliances - the Delivery Method for

WCRWP The fast track namre of the project, high likelihood of scope changes, management of co mplex issues and risks, including effect ive stakeh older management and government control , favo ured Alliancing as the delivery method.

The Alliancing framework also ensured the cap ture of key skills and reso urces in an already busy market place. W ithin the Western Pipeline Alliance the participating teams were able to quickly take advantage of the early invo lvement developing a high performing team and promoting a number of key initiatives and innovations.

Motivation was also increased across all participants in the team structure through rhe shared vision to deliver the public's desire for a drough t breaking scheme for South-East Queensland.

A 'no blame' culture predominated the alliance while still recognisin g individual's responsibilities and accountabilities of their own perfo rmance, The alliance arrangements fo r the project are nor without their problems, such as defining the commercial framewo rk, KPis and fin ancial impact. Defining 'success' has taken some time, limiting the impact of the innovative work undertaken at the early stages of the p roject establish ment.

Despite the rapid pace o f construction safety, quality and th e environmental impact have all been a key obj ective for the p roject with extensive environmental restoration wo rks and community co nsultatio n being undertaken.

The Alliance strucmre provides an effective mechanism fo r fast delivery but also for the development of personal skills by young engineers and exposu re to sen ior engineers in the project team. Knowledge capture and retention formed an important part of the process.

The Author Craig Berry (email Craig.Berry@ GHD .co m. au) is employed by GHD in their Brisbane Water Group and was appointed as Design Ma nager for the Western Pipeline Alliance in October 2006.

ASIAN EXPERTS SHARE SA' s GROUNDWATER EXPERTISE J McKay Straddling rhe Victorian border in South Australia's south-ease, Terry Buckley's properties produce some of the biggest potato crops in Australia.

The third-generation Limescone Coast fa rmer also grazes prime lamb and cattle, and grows "ocher bits and pieces" over hundreds of hectares. Dotted in the background are vast tracts of pine and blue gum plantations. Up the road a few dozen country mi les, there are row after row of grapevines - cabernet sa uvignon , shiraz, merloc and a hose of white grape varieties some 100 years old, growing in the Coonawarra's famed rich cerra rossa.

It's a rural setti ng familiar co many Australians. W hether it's SA's south-ease, Viccoria's Gippsland, the New South Wales Riverina or Queensland 's Darling Downs, these regions all share the same defining qualities - they are sparsely populated, are carved up by expansive land holdings, and host a mix of agricultural activities.

But standing on Terry Buckley's farm in January, a group of IO high level academics and senior government officials from Bangladesh, India, China, Pakistan and Nepal couldn't have helped but notice how different ch is setting was to farming regions at home where rhe scale of farming is very different with average fa rm sizes of less than one hectare. However, what was similar was the issue of water sharing processes, plans, policies and laws where the same problems occur.

T he delegation was in South Australia as pare of the I ncernacional Water Management lnstituce's (IWMI) Groundwater Governance in Asia program.

Ir aims co build capacity in the lndoGangecic and Yellow River Basins through a research-based training program for professionals and senior managers actively involved in groundwater management. The professionals were here ro see a sophisticated system of groundwater planning which considers the eco nomic environmental and social sustainabili ty aspects in future water allocation decisionmaking.

Coming from countries where private land holdings might be limited to 10 acres, where access ro groundwater - lee alone a

pump or the energy to power it - is no guaranteed thing, where the livelihood of up co 86 per cent of the population depends on agricultu re, and whe re collectively more than one billion people rely on groundwater fo r irrigation, che Buckley farm might have seemed a world away.

And yec there are many similarities, coo increasing and co mpeting demands fo r groundwater, risi ng salinity, falling water cables and rhe expectations of farmers.

Professor Jennifer McKay, Direccor of the Centre for Comparative Water Policies and Laws at rhe University of South Australia (UniSA) has hosted visitors fo r two years now under the IWMI program.

She said the aim of the visit was to share information about current groundwater law policy and practice in South Australia, which is advanced by international standards.

A key issue in many parts of regional Australia is rhe sharing of water between agricultural users and other users such as city water supply, industry and rhe enviro nment - rhar is, rhe economic, environmental and social sustainability aspects of water sharing.

This issue is one that has been addressed by all Scare governments in a water sharing process which results in water plans being created by the local community. In all cases to dare in Australia rhe water planning

processes have resulted in revised water allocations co all sectors of the co mmunity.

The most severely affected have been broad acre growers in agribusinesses.

The water planning processes of late have been fostered by Commonwealth government initiatives called the COAG reforms ( 1994) and the National Water Initiative (2004) which were a response to international calls coming from rhe Rio and Johannesburg summits fo r sustainable development laws and policies.

In South Australia the water plan ning process is part of rhe Natural Resources Management Ace (2004) and rhe creation of eight NRM Boards in the Stace. There are 56 such regions in rhe whole of Australia.

"South Australia not only has the most coherent approach co water management in Australia," Prof McKay said, "we also lead the way globally when ir comes to groundwater management."

The two-week vis it included presentations

by UniSA water experts, representatives from the SA Department of Water, Land and Biodiversity Conservation and farmers such as Terry Buckley who cocally rely on groundwater. As well as che Stace's Sou th Ease, the visit cook in rhe Coorong and rhe Virginia marker gardens north of Adelaide.

Director of the IWMI in Delhi, Dr Bharat Sharma said che importance of the program should not be underestimated. T he

Journal of the Australian Water Association

Water

MARCH 2008 125

groundwater used for agriculture in these the Indo-Gangetic and Yellow River Basin states is worth around US$20 bill ion per year and sustains an agricultural o utput of between US$50-64 billio n annually.

Around 50 per cent of the populations of these basin countries directly benefit from this output and with proper targeting, groundwater offers big opportunities fo r poverty reduction in many areas.

Groundwater use in these areas had spawned a "colossal anarchy", with millio ns of small and scattered groundwater users operating in an unregulated environment.

"What is lacking is a p roper inst itutional structure and capacity to manage groundwater."

South Australia, particularly the State's south-east, is well advanced in this area.

The commu nity there has been consulted by the local N RM Board and the impacts of the policy is bei ng evaluated in a study funded by the CRC Irrigation Futures as part of the system harmo nisation project.

T his project has produced a photographic exhibition called Picturing water use and justice which was shown at ArtLab Australia lase month as part o f the Adelaide Fringe Festival and will tour other pares of Australia.

The visitors - Mr Anwar Zahid, Deputy Director, Groundwater Hydrology, Bangladesh Water Development Board; Mr Mahbub Ul Alam, Join t Secretary, Ministry o f Water Reso urces, Government of Bangladesh; Mr Fengxin Kang, Professor of Hydrogeo logy, Deputy Director of Hydrogeology Division, Shandong Provincial Bureau of Geology and Mi neral Resources, China; Mr Jiqun Zhang, Deputy Section Director, Water Resources Management Center, Ministry of Water Resources, Beij ing, China; Dr

The wonders of irrigati on.

Shakeel Ahmed, Depu ty Director, National Geophysical Research Institute, Indo-French Centre fo r Grou ndwater Research, I ndia; Kishore Chandra Naik, Superintending Hydrogeologisc/Central Ground Water Board, Central Ground Water Board, O rissa, India; Baiwinder Singh Sidhu, Director of Agriculture, Govern ment of Punjab, India; Avinash Mishra, Dep uty Adviser, Planning Commission, New Delh i, India; Mr Jeevan Lal Shrestha, Deputy D irector General, M inistry of Water Resources, Government of Nepal, Groundwater Resources Development Project; and Dr Allah Ditta Khan, Director (Hydrology), Pakistan Council of Research in Water Resource were certainly impressed by the regulation of grou ndwater in South Australia.

As Dr Sidhu, from the Punjab, explained, regulation of groundwater in pares of India - where only o ne in fo ur farmi ng households owns an irrigation well and the remaining three pay for pump irrigation has a very practical dimension.

The government has agreed to supply electricity for free co farmers in order co operate pumps. This way, the authorities can control where and how much water is pumped simply by switching off the electricity.

Dr Ahmed said that wh ile the scale of problems in India is very different from Australia, there are lessons to be learned.

India, of course, has a much greater population. The land hold ings per farmer are much smaller, and the geological formatio n is also very different - limestone is predom inant in SA's south-east, whereas two-thirds of India is covered in hard rock, which makes it geologically more complex.

Nevertheless, he was very impressed to see how groundwater is managed, taking into accoun t stakeholder input and scientific inpu t.

"In India, we need more of the scientific modelling that is do ne in Australia".

Water from underground.

126 MARCH 2008 Water Journal of the Australian Water Association

Professor Kang, from Shandong Province, China, where the aquifer is similar to that in South Australia's south-east, said that with many sources of surface water polluted beyond use, China will increasingly rely on groundwater. Water law in China relates to surface water only. "We need governmen tmanaged law fo r sustainable use of groundwater".

THE CHALLENGES IN ASIA Compiled by E A (Bob) Swinton At the ASPIRE Conference in 2007, four presentations outlined the supply and management issues facing their countries.

Readers will note that there are certain similarities to the Australian situation, but also some radical differences. Some comments on China' s problems have been added.

Japan

Japan faces rwo sets of issues, that of a sustainable water supply combi ned with the inertia of a highly decentralised ad ministration, a legacy of the past.

Professor Yoshimasa Watanabe, President of the JWWA, outl ined his respo nse to the possibility that global warming will reduce current rainfall in critical areas. For sustainability there are several solutions which depend on geological, eco nom ical and cultural fac tors. For most si tuations, recycl ing water is the most effective solution and considerable developments are taking place. Applications range from provision of super-pure water for rhe semiconductor and liqu id crystal factories, and 'fit for purpose' water for huge office complexes and commercial buildings.

Developments in MBR are the key to efficient water recycling.

Yoshihiko Sugiyama, from Tokyo, stated that water utilities in Japan are faced with similar but also some very different problems co mpared to their counterparts in Australia. They face a decline in fi nance, the need fo r rehabilitation of infrastructure but unlike Australian cities, a steady decl ine in population which is acco mpanied by an imbalance in demographics, with a decrease in the worki ng population and an increase in rhe aged .

In 2004, the government launched "Waterworks Vision", a national strategic goal to extend the water system.

There are about 10,000 water utilities in Japan. Most of them are small to medium and they are financially and technically unstable. In fu tu re they will face severe stress because of a decline in water demand, costly regeneration of infrastructure, and the big issue of technical skills succession caused by the mass retirement of the babyboomer generation.

To address these issues, the integration and expansion of water systems will contribute to strengthen their financial position, secure water resources, provide more efficient water supply control and build

stronger maintenance systems. Despite government initiatives to dare only about half of the utilities have been integrated.

Some reasons fo r the delays are the diffe rent management systems of the larger and rhe smaller utili ties, differences in customer charging systems, and d ifferent connection charges. Yoshihiko then analysed these reasons in more derail and suggested strategies to improve the s1 tuatton.

A case study of a large uriliry in Tokyo itself ill ustrated some of rhe complexities.

The Tokyo Bureau of Waterworks (TBW) directly manages the waterworks of central Tokyo. However, rhe Tokyo Metropoli tan Govern ment (TMG) has delegated rhe operation of the Tama Waterworks, which supplies the western zone, to the 25 local governments, although retaining the ownership. T o solve the dual management problem, TBW is proceeding with integration of these local utilities bur only by 2012 wi ll all 25 co me under the one ad ministration.

TBW has invested in a complex management system wh ich will supervise this reorganisation. T his will be the fitst case in Japan and will be applied to other cities for integration of their water systems.

Jc will involve outsourcing fu nctions to the private sector.

Jc is very important for Japan to lower costs and improve efficiency.

Malaysia

Teo Yen Hua, of the Ministry Of Energy, Malaysia, spoke on rhe two major issues facing che current water industry structure in Malaysia. Firstly, there is a lack of coordination between Scare-level and Federal-level govern ments, inter-agencies, agency-operator and agency-operatorconsumer. T his shortco ming has led to varied levels of industry performance across the State's governments. Secondly, rhe current industry structure allows private operators to rake on various pares of the water services but not all have met with economic success in doing so.

Consequently the major objective of creating an opportunity to raise fu nd ing for construction, operation and maintenance of future wate r assets to achieve national development goals appears to be at risk.

T he proposed industry model aims to ensure a more holistic approach to the management of the whole value chain by an injection of efficiency and effectiveness and the establishment of an economic regulation body. T he proposed model will take into consideration the interest of all stakeholders, especially the consumers, to ensure char every citizen has access to affordable, clean and potable water and efficient sewerage services.

The presence of a comprehensive regulatory framework is vital, to ensure that in dustry reform can achieve full cost recovery in the long run. Hence rwo pieces of legislation have been introduced: the Water Services Industry Act and the National Water Services Commission Act. The for mer secs out the direction for reform by introduci ng an integrated and holistic approach. The latter was enacted to support the need for a strong body to ensure proper reinfo rcement of the former .

T he ocher critical success factor is the need to overcome fund ing restrai nts. At che moment State govern ments largely depend on fu nding from the Federal government.

Yet there is a risi ng demand for water supply in terms of coverage an d qualiry from both domestic and ind ustrial sectors.

T here is a lot of pressure on State govern ments or operators to carry out their obligations with limited fi nancial resources.

As a long-cerm solution, a fu lly Govern ment- owned Water Assets Management Company (WAM Co) has been established to undertake the development of infrastructures on a build and lease principle. Through ch is arrangement, water operators become assetlight as they are relieved from financial responsibilities fo r infrastructure development, rhus allowing chem to improve water qualiry and service levels through che proper allocation of resources.

Journal of the Australian Water Association

Water

MARCH 2008 127

technical features

Korea

China

A team from Korea University, Zuwhan Yun and Euiso C h oi, and Kyungpook National University, Kyung-Sok Min, outlined the reaso ns for che develo p ment of wastewater recycling in South Korea .

Zhou Shengxian, Minister of the Stare Environmental Protection Administration addressed rhe National Meeting on Pollution Preven tion and Control of Rivers in Zhengzhou City, H enan Province on 14 November 2007 and noted char the p urpose was co implement che omcomes from the 17th National Congress.

The Korean economy will co ntinue co expand in the next decade, bm the development o f new fresh water resources to support public and industrial uses has been d rastically decreased.

The coca! per capita freshwater resource in Korea is on ly 1400 m3/yr, yec rhe public water supply in 2004 recorded 365 lpcd, or about 10% of the total. D espite various water-saving efforts, the need for alcernacive water supplies is increasing in the next decade.

However, citizen groups and environmental NGOs are opposed co co nstruction of new dams for ecological reasons. In any case, the available watershed for large scale multipurpose dam develop ments is limited , while small dam construct io n is not cost-effective.

For these reasons no major dam has been constructed in che lase decade.

The territorial area of Korea is on ly 99,600

km 2. The water supp ly for 47 million Koreans is largely dependent of five major river systems. In evaluating future water supplies there is also a need co preserve sustainable water quality in rhe river systems.

The options, including re-use of reclaimed wastewater, rainwater harvesting and desalination, have all been evaluated. Tr has been fo und that the use of reclaimed wastewater is che most attractive and feasible option. le is a more stable water resource in terms of supply side than che other options. le will reduce the Water Exploitation Index., an ecological indicator o f the health of ou r river systems. Wastewater reclamatio n will further allow sustainable development of ecologically sensitive areas, such as riverine areas, by reducing export of pollutants. It will further reduce export of n itrogen to the surroundi ng seas, which is currencly far higher than for ocher internatio nal zones.

However, in the assessment it was recognised chat extensive re-use will require, not only investment in suitable wastewater treatment systems, but also extensive distribution systems.

128

MARCH 2008

Water

The Cen tral Committee of rhe Co ngress sec prevention and con trol of water pollution as a major task in rhe strategy of sustainable d evelopment.

Although a number of effect ive measu res have been imp lemented, he said chat ic must be appreciated ch at water pollution is scill comm on, wastewater treatment is still lagging nationally and th e safety of th e water environment is still unsecured .

Water pollution is cu rrencly jeopardising public health, social development and affecting the nation's economy.

M inister Z hou indicated the ultimate goal is co return the rivers co the original p ristine condition and hen ce ensure the safety of the entire river basins, bu t in the short term co improve the drinking water indicators up to 100% by 2 008 , improve the enviro nmental water quality of river basins co suit plann ing requ irements, life the urban sewerage race co 70%, reduce the coral major water pollmants in the form of COD by over 10% from 2005 values and strengthen rhe capacity of Agencies in supervision of pollution control and emergency response systems by che end of 2010.

Minister Zhou en unciated rhe following p rinciples for pollution control: I. The environment and its preservation must serve as the basis for eco nomic and social development.

2. Ensure the safety of d rinking water co achieve public health and protect catchments used for drinking water.

3. Target responsib ility fo r river basins and sec policies co suit each river basin for pollution control.

4. Tighten system management o f river basins overall, so cha r ecological health is maintained and provide n ecessary treatment for domestic, industrial an d agricult ural wastes so ch ar water can be allocated co suit consumers' water quality.

H e seated chat there was a need for environmental planning fo r all projects, che discharge of toxic and hazardous substances monitored an d controlled with permits issued and effective industrial treatment of wastes enhanced co meet environmental cond itions . To achieve water quality and environ mental guidelines there is not only a need for legislation, bu t it muse be policed.

Water resources and environmental protectio n muse be coordinated and unified.

M inister Zhou co ncluded chat the targets and policies are in place, but now it is necessary co implement chem all. Starting from 2008, SEPA will monitor Agencies on their achievements and provinces failing co meet targets or exceeding pollu tion loads will have Scace suppo rted projects suspended and will nor gain approval for fu ture projects.

H e seated chat it is necessary co mobilise the community ac large co participate in environmental protection, educate on rhe environment, culcure co nservatio n and develop an attitude by che communi ty of appreciating and saving water.

Ac ENVIRO 2006, Mr Gai Yin, Secretary General, Advisory C ommittee for Science and T echnology of SEPA commented o n rhe current water pollution situation in China as follows:

'1/ man does not conserve water, then the last drops on earth will be his tears" Some statistics quoted: • Toca! water resources - 2,810 billion m 3

Water Advertising To reach the decision-makers in the water field , you should consider advertising in Water Journal, the official journal of Australian Water Association.

For information on advertising rates, please contact Brian Rault at Hallmark Editions, Tel (031 8534 5000 or email brian.rault@halledit.com.au

Journol of the Australian Water Association

• Average water resource/head - 2,400 m 3 (a ¼ of world average) • Sewage discharged is around 40 billion m3/year.

• Investment of 1300 billion Yuan for environmental protection during the 11 ch Five Year Plan (A$200 billion).

J ust u nd er half of its nearly 90,000 dams are dangerously unstable and need urgent repair, requiring a three-year plan costing 27 billion yuan (A$4 billion) (fro m abc.n er.au/ news).

.fereed paper

STORMWATER IN THAILAND: A POTTED HISTORY K Harriden Abstract "Srormwarer" reflects a specific attitude ro overland fl ow nor shared by all environmental knowledge systems.

Attempts to transfer the srormwarer construct to Asian countries has lead ro a range of unintended negative co nsequences.

Chonb uri Province, Thailand, fo r example, has experienced nor only changes in rhe physical environment bur at a local level there has also been a loss of local scientifi c knowledge, with limited knowledge about rhe new system provide in its place.

Introduction The srormwater construct is nor ubiquitous.

Rather, it is a strongly engineering focussed idea arising from the desire, common to the scientific approaches of the industrialised urban ised environments of Western Europe to control environmental processes for h uman convenience. The basic premise of srormwarer technology is to quickly remove overland flow from an area so char property and livelihood are nor impeded in any way.

The term encompasses both the channel and drain infrastructure and the water in, or on its way ro, the disposal in frastructu re.

The water is seen as a waste product with no positive values. A discussion of how excess overland flow came to be seen in these terms in beyond the scope of this shore article.

Srormwarer, as just briefly described, is a relatively new concept in Thailand, although European scientific thought has been fam iliar to the Thai royal circle since at least the mid 19th century (Winichakul, 2004). The concerted adoption of European scientific approaches and practices (science), including the science of srormwarer, has only come with the poseWorld War II natio nal economic and industrial development agenda (Thawrhong, 2001). The use of European scientific approaches in Thailand is nor the firs t example of rhe adoption, and cooption, of ideas and practices from ocher environmental knowledge systems.

Buddhism (Byrne, 2007) and Indian water practices (Rigg, 1992) are two influential examples.

Local Wisdom: Floods Versus Drought Before rhe widespread adoption of European science, a local environ mental knowledge system, called local wisdom (.f,Jili1f\jf\j1'Vlil'3tiU) , was used ro investigate and articulate the natural worl d. Local wisdom has a different notion of excess surface water to char of science's "srormwarer". Thai local wisdom sees the water, and resultant water bodies, as valuable collective mulriuse water sources.

The use of natural landscape features such as depressions and hollows ro collect water is common, creating wetlands, marshes and ponds char have varyi ng life expectancies bur sim ilar roles, including sto rage for d ry season water security, important food sources, especially of fish protein and green leafy planes and water and habitat for worki ng animals such as buffalo. Irrigation cooperatives, some still running today on agreements and practices handed down for many generations, also rook advantage of excess overland flow, th rough the use of simple technical and landscape modification practices, ro harvest water for rice production and other irrigation needs (Urivan, 1995) . Many of the water bodies fed by large overland flow even cs also fulfil important spiritual needs, as well as ameniry/aesrhecic roles. Underlying these local wisdom understandings of excess overland flow is a sense of interconnectedness between people, spirits and nature, and a willingness ro work with nature's cycles nor evident in rhe European "srormwarer" construct.

T his short o utline of local wisdom's approach ro overland fl ow suggests char in pre-development T hailand the notion of excess surface water was far less worrying than rhe scarcity of surface water. Thar historically these communities were more co ncerned about drought than flooding is demonstrated by one of the more evocative, if increasingly neglected, local wisdom measures for annual rainfall called "Naga

Wetlands as a key management approach to excess overland flow.

givi ng water" (irn11~ t.ii") Qumsai, 1997, p.

24). T h is measure places the Naga, a cosmic serpent strongly associated with water, on a scale of one ch rough seven, wi t h increments of one Naga. A one Naga year is considered good as only o ne Naga is drinking rhe available water, so there is !ors of water around. A seven Naga year is a bad year as rhe max imum number of Naga possib le are drinking all the available water, so th ere is little water around Qumsai, 1997). It is not possible ro have a no Naga year, as cosmologically speaking there must always be at leasr one around, so a one Naga year necessarily indicates significant overland flow events, possibly quire serious floo ding. Yer as it is possible to have all seven Naga's drin king all the water, water scarcity is clearly the focus of this measurement system. In such a case it seems feasible that a community concerned about drought would val ue all water sources positively and be reluctant to cons ider any portion of the available supply as waste.

Ir is worth quickly noting that the "Naga giving water" measurement is a good example of how the spirit-human-nature interconnectio ns underlying local wisdom are expressed in daily life.

Chonburi Province, Thailand Given the sustained emphasis on national economical and industrial development in contemporary Thailand, the understanding o f overland flow, particularly among the political, capitalist and technical classes, increasingly reflects those of the srormwater construct and overland flow is seen as a negative event. Chonburi Province provides a useful case study of how these changing understandings are playing out at a local level as it is, for h istorical reasons, more integrated into both Thai and global economic aspirations than ocher region of the Kingdom. This significance is demonstrated by the emphasis in national development programs on the industrialisation of chis part of the country (Pongsapich, Hafner & Veeravongs, 1979; Thawrhong, 2001). In many ways Chonburi represents the globalised developed economy to which the Thai political and economic elite aspire.

Journal of the Australian Water Association

Water

MARCH 2008 129

Located on Thailand's eastern seaboard, Chonburi Province relies on water for almost every aspect of life. The name literally means "water town". Historically, significant parts of the city have sat on stiles beside docks, ab ove che large tidal zone (Pongsap ich et al, I 979). T he contemporary town, capital of the same named province, is crammed between che landward side of the intertidal zone and the main highway heading ease from Bangkok.

The tropical climate and low lying land encourage temporally variable wetlands of various sizes. Given Chon buri Provi nce's tropical coastline, many wetlands are mangrove environments, as indicated by nineteenth century maps of the Bay of Thailand (Phasuk & Stott, 2004) . These maps indicate a heavily forested coastline.

Prior to industrialisation, C honburi was predominately a fish ing society with many activities related to the ocean includi ng fishing, fish sauce production and dried fish and squid production. Rice farming and mixed small scale agriculture were also important activities (Pongsap ich et al, 1979). The pace of li fe was tempered to eb b and flow of the tides, with the foc us much more on human-water than humanland relationships Oumsai, 1997). Although the introduction of stormwater infrastructure has been patchy during the national industrial and economic development planning process, it has still led to the loss of wetlands. A 1986 topographical map of the Chonburi coastline showed significant mangrove stands, although less widespread than earlier mapping. Now many of these coastal areas are cleared of mangroves.

The loss of wetlands on a large scale in C honburi is most obvious on the coastline near C honburi town and other large coastal towns of the province such as Pattaya and Sattahip , where housing estates, tourist resorts and indu stry have replaced them.

C learly this change in wetland distribution is not entirely stormwater related. However the influence of stormwater demands o n the total wetlan d area can been seen on smaller scales. For example Burapha Un iversity in Bang Saen, Chonburi, has a series of lakes and wetlands throughout the campus. Not only do they capture a substantial portion of the overland flow during the wet season, they also provide important spiritual sustenance an d amenity to staff and students, with one wetland in particular having strong spiritual significance for students seeking good results. The university was plumbed and connected to the local stormwater system in 1997/98, with some dis ruptio n to the wetlands system. The number of bu ildings

and people on the campus has grown substantially in the last 10 years, with lakes and wetlands being gradually being fi lled in to accommodate the associated land demands. A few years ago one of the larger lotus covered wetlands was converted into a sealed surface parking lot. The car park floods every wet season (pers comm. April, 2007), in spite of the scormwater drain installed at the lowest point. It seems evident that the wetland system was more effective at absorbing overland fl ow than the scormwater system in this case.

with no infrastructure component. By engulfing naam laak in the realm of technical/scientific language, without (re)educating the rest of the T hai community, it has been rendered mostly meaningless. Its everyday use is now confused with a highly specific definition with no immediate cultural or physical relevance co the end users of the scormwater system. Every day people are using an unfamiliar system with seemingly familiar words. It may have been better co make a new word.

Yet the attitudinal changes co excess overland flow, leading co con cern about excess overland flow on the part of the political, scientific and technical classes.

Field observations of the way the scormwater in frastructu re is used in Chonburi strongly indicates th is discontinuity in knowledge. I have seen sewage systems from hotels and housing being conn ected d irectly to stormwater infrastructure (during the construction phase), the greywater from commercial laundries drained directly into srormwater system and drain inlets so blocked from dry season dirt and debris you wonder how the first flush (es) of wet seaso n water can move through the infrastructure at all.

Now, while blithely using imported scormwater management approaches and technologies, many residents of Chonburi Province no longer value or understand wetlands, and by association, overland flow, the way they once did. This reflects an erosion of the local wisdom knowledge base and its application in the region. During field interviews, there was a noticeable age d ifference in the levels of local wisdom regarding overland flow capture and storage among those interviewed. The younger the interviewee, the less local wisdom they held and the less comfortable they were talking about local w isdom generally (pers comm.

April , 2007). The low levels of awareness about science's overland flow understandings and management pract ices across th e generations strongly suggests limited scientific environmen tal knowledge has been offered co replace the waning local wisdom (pers comm. April, 2007).

Different Thinking, Different Language My research suggests an important reason fo r this d iscontinuity in knowledge is due to the imposition of the scormwater construct by the political and scientifictechnical elite o f Thailand and the international development culture. This imposition is exemplified by the limited use of naam laak (1H1-1111n), the Thai language term fo r scormwater. Talking co everyday people exposes not only a deep ignorance of the fundamentals of the scormwater construct (some respondents believed it could be drunk) bur there were few people who were even familiar with the Thai language term adopted for srormwater by the scientific com munity. The lack of general awareness of the technical use of naam laak is not surprising, given it is an everyday word . A Thai word was needed co articulate the new scientific scormwater construct so the technical classes seconded naam laak to describe this non-local scientific concept. The term appears in the Environment Engineering Association of Thailand (EEAT) Prescribed Words

and the Environmental Definitions (EEAT, 2003) and the definition reflects science's understandings perfectly (p. 439) .

(~vnifucycy~LLol::{ltJ11J~'1LL16liHl1J )

The everyday use of the term holds a more positive understanding of overland flow,

130 MARCH 2008 Water Journal of the Australian Water Association

Australian Developments Interestingly as the scormwater construct is being increasingly applied in Thailand, it is coming under scrutiny in Australia. T his scrutiny is coming from local level natural resource management groups, as well as the scientific communiry and, to some extent, industry. Environmental, water quality and other issues are seeing the stormwater 'waste' tag increasingly challenged. Science is now trying co find ways co revalue and repackage srormwater so that it can be used, rather than rushed down concrete channels.

One of the more popular approaches is wetlands for scormwater management. The first attempt in the ACT co rehabilitate wetlands for scormwater management was undertaken by a community-led N HT funded project, resulting in the very popular David St wetlands. A more recent example is the planned government managed rehabilitation of some reaches of Yarralumla Creek from a concrete channel back co the chain of ponds they once were.

A Revival?

There is also a change in recent years in the level of accep tance of science in Thailand,

technical features asia pacific issues with an increasing n u mber of challenges, by bo th academics and comm u n ities, co the privileged status o f European science. These challenges have come with a general, and somewhat complicated, revival of local wisdom practices. T he difference is that the challenges to science in Thailand are motivated by cultural factors, rather than the environmental pressures driving change in the science of srormwater in Australia.

Th is article has described Thai local wisd o m's use of wetlands as a key management app roach to excess overland fl ow. In the course of technological development longstand ing wetlands were radically altered , often removed, to accommodate the sro rmwater co nstruct.

Now, as weste rn European science itself starts to challenge th is construct, it continues co influence T hai environmental knowledge so t hat the "technology" of wetlands for stormwater management is actually bei ng imported into Thai land. It seems that the local wisdom approach of valu ing excess su rface water and scoring it

rloPro

in temporally and spatially variable wetlands may prove better than the "fear it and fling it" approach of science.

The Author Kate Harriden is an independent researcher of water and Southeast Asia. She has just completed a Master of Ans (Asian Studies), where the main research foc us was the stormwater construct in the ACT and Chonbu ri, T hailand. She is currently worki ng, with D r Lahiri-Dutt at the ANU, on Australian rural intra-househo ld water use and , individually, on the uses of and differences in th e various Thai wo rds for wetlands. Email wolfkeeng@bigpond.com

Bibliography Byrne, D. (2007). The Forrress of Rat ionality: Archaeology and T hai Popular Religion. (unpublished) Environment Engineering Association of Thailand (ÂŁEAT) (2005) Prescribed Words

and the Environmental Definitions (~~\'lll"cyl\T~llll~ilu1lJA~u;~AillJJ Bangkok: EEAT (in Thai)

ra . . . . . . ...

Jumsai, S. (1997) . Naga: Cultural Origins in

Siam and the West Pacific. Bangkok: C halerm nit Press and DD Books.

Phasuk, S. & Storr, P. (2004) . Royal Siamese

Maps: War and Trade in Nineteenth Century Thailand. Bangkok: River Books Pongsapich, A., Hafner, J. & Veeravongs, S. (1979). Chonburi Project: Institution and

Human Resources Development in the Chonbttri Region. Bangkok: Chulalongkorn Un iversity Press.

Rigg, J. (Ed.). (I 992). The Gift of Water: water

management, cosmology and the state in Sotttheast Asia. London: School of O rien tial and African Studies, University of London.

Thawthong, R. (200 1). A Survey and Research of the Basic Sources of Arrs, Culcure and Local Wisdom in rhe Eastern Region BuraphaA rts journal, 6- 17. (in T hai) Uraivan, T. (1995). Muang-Fai CommttnitiesAre For People. Bangkok: C hulalongkorn University. (in T hai) Winichakul, T. (2004). Siam Mapped: A History ofthe Geo-body ofa Nation. C hiang Mai: Silkworm Books.

technical features

odour manag

PREDICTING HYDROGEN SULFIDE FORMATION IN SEWERS: A NEW MODEL K Sharma, D W de Haas, S Corrie, K O'Halloran, J Keller, Abstract A new model has been developed both as a research tool and in collaboration with industry partners. In this paper, the successful application of the model to simulate rising main systems on the Gold Coast was demonstrated and tested in systems with oxygen injection intended for control of dissolved sulfide and odour.

Introduction T he production and emissio n of hydrogen sulfide (H 2S) has long been known to be a major cause of cor rosion and odour problems in sewer systems (Latham , 1873; USEPA, 1974; Boon and Lister, 1975; Pomeroy and Parkhurst, 1977). When anaerobic conditions prevail in a sewer system, sulfate present in the wastewater is reduced to sulfide by sulfate-reducing bacteria residing in biofilms on the walls of the pipelines. This results in emission o f H2S to the sewer atmosphere, causing odour and corrosion problems in partially-full pipe sections, manholes, vent pipes and other places in contact with air. Rising mains, which normally operate fu ll and thus lack oxygenation, contribute considerably to H 2S production in a sewer system. In Australia the problem is generally exacerbated by the hot climate as well as relatively flat coastal terrain with high population densities, resulting in long rising sewer mains that collect sewage from catchments with sandy soils and where a high potential often exists for intrusion of sulfate- rich seawater.

Prediction of sulfide production in rising mains, especially if based upon understanding the hydraulic conditions, wastewater composition and fundame ntal processes o ccurring, would greatly benefit the development of appropriate strategies fo r controlling sulfide formation or H 2S emissions.

Cu rrently there are several strategies available for control of sulfide production in sewer systems. These include the injection This paper is a modified and updated version of that presented by de Haas et al. (2007) at the OzWater 2007 Conference, Sydney.

132 MARCH 2008 Water

of chemicals such as oxygen, ni trate, iron salts or alkal i to either p revent sulfide formation, remove sulfid e fro m wastewater once formed, or inhibit its emission as H 2S gas (Hobson and Yang, 2000; WEF, 2004; de Lomas et al., 2006) . Co nsidering the high chemical costs involved as well as the cost of conducting field studies to determine the effectiveness of such strategies, modelling provides a more eco nomical and rational means for assessing and optimising these strategies.

Significant p rogress has been made in recent years in u nderstanding the biological transformation processes that occur in sewers. In particular, a group at Aalborg University (Denmark) developed the WATS (Wastewater Aerobic/Anaerobic Transformations in Sewers) model (Hvitved-Jacobsen et al., 1998; 2002) based on the knowledge of in-sewer biotransformation processes at the time.

The model d escribes both the anaerobic and aerobic processes involving multiple carbon and sulfur species, and was a major step forward, compared with earlier models based o n empirical express ions (Thistlethwayte, 1972; Boon and Lister, 1975; Pomeroy & Parkhurst, 1977).

H owever, application of the WATS and o ther models has usually been limited to sewer systems under steady state conditions (Hvirved-Jacobsen et al. , 1998; Nielsen et al., 2005). Sulfide predictions are generally based on average flow and average wastewater characteristics, and the H 2S concentrat ion is pred icted as a function of location; temporal variations are completely ignored. Due to considerable variations with time both in flow and sewage characteristics, steady-state predictions have limitations in developing H 2 S control strategies.

This study describes the use of a new mathematical model that has the capability to predict sulfide production in sewer systems dynamically (i .e. as function of time

Simulating biological, chemical and physical transformations.

Journal of the Australian Water Association

z Yuan

and pipe fl ow rate). It can be used to investigate temporal and spatial variations of sulfide productio n in rising or gravity main sewers. It was calibrated and validated for rising mains using field data collected by Gold Coast Water in A ustralia. This paper summarises some of the model applicatio ns against fu ll-scale test data and d emonstrates its performance both in the presence and absence of oxygen dosing as a con trol strategy for sul fid e.

Model Development The dynamic rising main model was form ulated at the AWMC (University of Queensland), based on concepts and models previously established in literature for wastewater systems (Freudenthal et al. , 2005 ; Hvirved-Jacobsen et al., 1998; Hvitved-Jacobsen, 2002; Gujer et al., 1995; O 'Brien and Birkner, 1977). A number of improvements have been mad e to the model, which include: formulations for pipe flow conditions (flow rate, quiescent vs. flowing time, pipe length and diameter); boundary layer formation (in proximity to the biofilm on the pipe wall under nonmixed conditions); significant changes to processes for chemical/biological oxidation of sulfide and reduction of nitrate; and key equilibrium processes for predicting pH, carbonate, phosphate, sulfide and iron subsystem chemistry. Development of the model has been further described by Yuan et al. (2006) and Sharma et al. (2008).

Model Testing Field studies were conducted in several rising main systems belonging to Gold Coast Water (Queensland). Special sampling points were installed, usually consisting of a 16 mm diameter pipe tapped into the underground rising main at each locat ion and conn ected to a sampling tap at gro und level. Samp les were taken manually at differen t locations over d iurnal periods rangin g from approximately 8 to 36 hours. Key parameters measured were co ncentrations of dissolved sulfide, sulfate, thiosulfate, and sulfite using ion chromatography (IC).

Sample preservation and analytical

technical features II.

.fereed paper

Table 1. Summary of the sampling campaigns on Gold Coast Water rising mains.

Campaign & Date

1 15 Nov 2005

Rising Main System

Length (m)

UC09 (Single)

1084

Pipe diameter Daily average Oxygen (mm) flow (m3/ day) injection

150

79

Sample collection

No

Hourly somples for 29 hours at 5 locations (wet-well, 357, 547, 828, and 1002 m)

No

Hourly samples for 28 hours at various locations, including TH5 wet well (0 m) Elenora WWTP inlet (13900 m)

2 9-1 0 Moy 2006

TH5 to Elenora WWTP (Network)

3

13900 100 to 600 (Network led from 13 no.

(Ave. 440, branch pump stations) weighted by length)

14324

(Note 1)

D6

15 Moy to 25 Jun 2007

Si ngle (up to 750 m from orig in (Note 2)

5300

450

2079

Yes (Note 3)

Continuous (1 min. interval) on-line data for pH, DO, Temp. & dissolved sulfide at wet well (0 m) and 735 m Grab samples to validate on-line data

Note 1: Oxygen injectors exist at seven of the thirteen pump stations included in the TH5- Elenora network. Six of these had been used in the recent past (up to March 2006). Scenarios for operating in "current" configuration and proposed (opti mal) configuration were investigated.

Note 2: The 06 rising main is ioined by other minor rising mains /034 and 055 at 2500 and 3100m from origin at Pump Station 06) before connecting to other major mains from Helensvale and Eastern Forced Main that pump to Coombabah WWTP. Note 3: Oxygen injector on 06 operated in the latter part of the campaign period /7-25 Jun 2001).

WWTP: Was tewater Treatment Plant

methods for IC were critical co rhe success of the project and have been described by Keller-Lehmann et al. (2006).

Ocher measured parameters, which were used for sewage characterisation, included

cemperacure, pH, volatile fatty acids (VFA) by gas chromatography, soluble chemical oxygen demand (CO D) and flocculacedfilcered COD. On-line (real time) measurement of sulfide was made possible

TH5 flow data, 8 Nov 2005

The model was tested against data collected from three field campaigns, as summarised in Table 1.

300 U) 250 ::J 200 Cl> ~ 150 ~ 100 0 Li: 50 0 0

6

12

18

24

30

36

42

48

36

42

48

Tt-13 flow data, 9-10 May 2006

30 ~ 25 20 Cl>

~ ~ 0

Li:

15 10

I

5

0 0

6

12

18

24

I 30

C30 f low data, 9-10 May 2006

~

at selected sites by installation of a specialised spectrophotometer along with continuous sampling system developed inhouse by GCW staff. Similarly, on-li ne dissolved oxygen was implemented at one of che cesr sires chat had oxygen injection upstream.

T he risi ng mains were modelled as ranksin-series ro mimic plug flow in a real sewer pipe. SimulinkÂŽ in MATLABÂŽ was used as rhe simulation platform. Simulation was performed with various numbers of tanks ranging from 25 co 100 for a 1.084 km pipe (43 m ro 10.8 m length per rank), and rhe results did nor show any remarkable differences (<0.2 mgS/L predicted coral dissolved sulfide). As a compromise between mimicking rhe plug-flow nacure in a pipe and rhe simulation speed, rhe length of each tank was chosen co be between 40 m and 100 m, wirh a shorter length per tank for shorter rising mains and longer lengths per rank for larger network systems.

80 , - - - - - - - - - - - - - - - - - - -- - - - - - - -

Flow Data

60

Under dry weather conditions rhe flow pattern in rising mains is normally intermittent due co pump operation and varies significancly between different pump stations. Where available, flow meter data was used. Alcernarively, pump operating rimes were obtained from rhe GCW SCADA system and combined wirh wee well dimensions in order co calculate flow rares in che rising main. Three examples for small co medium-sized pump srarions in rhe T H5-Elanora network are shown in Figure l.

Cl>

~ 40

~ 20

Li:

0

6

12

18

24

30

36

42

48

Time (h)

Figure 1. Exa mples of rising main flow rates calculated from SCADA pu mp run time data. The retention ti me in risi ng main pipelines depends on both the interva l and d uration of pump ru n times, which co n vary widely depending on the size a nd characteristics of a given sewer catchment and capacity of the pump station servi ng the catchment.

Journal of the Australian Water Association

Water

MARCH 2008 133

technical features .fereed paper

a particular wastewater slug undergoes in a sewer pipeline. As a result, a large diurnal variation in sulfide production is expected along the length of a rising main pipe, both in terms of rate of sulfide p roduction and concentration at a given location. The data from full- scale rising mains clearly illustrate rhe dynamics of sulfide production, which a steady-state model would not be able to predict correctly for such systems.

Variation of Flow Rate and HRT

4

150

--;;

2.

100

~ r

2(IJ

0:::

Campaign 1 The fit between the model-predicted vs .

~

0

2

u::: 50

0

0

1.0

2.0

1.5

3.0

2.5

Time (days)

Figure 2. Diurnal variation in hydrau lic retentio n time (blue line) in a rising main (D6, Go ld Coast). Instantaneous (intermittent) flow rate from pump station in background (grey bars) .

As a consequence of rhe intermittent and variable flow rate, hydraulic retention rime (HRT) in che rising mains scudied (Table 1) shows considerable d iurnal variatio n in the range -1 to 10 hours, depending on the rime of day a given slug of wastewater enters the pipeline. Retention rimes are typically rhe longest fo r slugs of wastewater entering the pipe in rhe lace evening (-2100 to 2200), and emerging approx. 8 to 10 hours later at the end of the pipe. In

extreme cases (nor considered here) rising mains serving new (and under-loaded) catchments may have retention rimes exceeding 24 hours for slugs of wastewater entering at night. An example of the variation in HRT for the D6 rising main (refer to Table 1) over a two-day period is shown in Figure 2.

HRT d irectly affects the degree of biological or chemical transformations chat

25 1' 1~

~

I

15

C

c 1lC 0

u

Ir

d o

11

I I

2/

:I

1I

00 Q:) n..

rfa'o...,

I t) 10 I I tr)! \

'l~ II~ 1:/

0

-~

Location: 357 m

Location: 50 m

O

Ch :"R

I \

Cl

.s.

Do

I I

20

I

1 0 :J)

,,

oJ-.'

~

.'""Yt-"

. J~

',J', fH, e3

1,:

10

A schematic of the rising main pipeline system from Tugun (scarring from T H 5 pump station) to Elanora WWTP is shown in Figure 4. The system includes a balance rank (named BT in the figure), located at pump station C27. The ocher pump stations are marked in blue circles; pipeline lengths and diameter are annotated.

5

- - Diss. Sulfide (simulated ) Sulfate (simulated)

Locati on: 547 m

::;- 20 ci'i

0

â&#x20AC;˘

Diss. Sulfide (measured)

O

Sulfate {measured)

L oca tion: 828 m

Cl

15

C

0 :,,

i1l

10

C

0

u

5

o+--~-~----~-~-~--+-----~-~-~-~-----1 0

12

18

24

30

Time (h)

36

42

0

6

12

18

24

30

36

42

Time (h)

Figure 3. Mod el calibration agai nst measured sulfide a nd sulfate data at four locations along UCO9 rising main (Campaign l) (0 and 24 ho urs are midnight), w here the discharge pipe from the pum p statio n was the start of the rising main pipeline (0 m locatio n).

134

MARCH 2008

Based on rhe calibration experience gained in use of the model for Campaign 1 and subsequent campaigns it was possible to obtain one consistent set of model parameters chat retrospectively firred the measured data from all campaigns. The model data in Figure 3 for Campaign 1 reflects rhe model calibration gained from experience over approximately eighteen months during which the model was developed.

Campaign 2

0

.s.

measured sulfate and sulfide concentrations in UC09 at four locations resulcing from model calibration is shown in Figure 3. The model is able to predict the variations in sulfide and sulfate concentrations reasonably well. The first two sites (50 m and 357 m from the wet well), in particular, showed an excellent fit over the entire 29 hours. The model predictions generally followed the measured trend for the last two sections although some discrepancies were observed. The predicted sulfide concentrations were lower than the measured values in some periods, while in ochers the opposite was observed. We hypothesise chat the spatial variation in rhe biofilm activity along the pipe, which was not taken into account in the model, may be one reason fo r the less-than-ideal fie.

Water

Journal of the Australian Water Association

48

Figure 5 shows the measured and modelpredicted d issolved sulfide concentrations at Elanora WWTP with no control measures applied in the rising main system (oxygen injection turned off). Whilst the measured and predicted values did not match exactly, the fit may be considered good, taking into account the size and relatively complexity of the network. Due to limitations on resources, the diurnal changes in sewage characteristics at all the pump stations

technical features

· _ 9d _o _ur manag could not be measured. The measured va lues for one pump station (TH5 ) were applied to all others, except where grab samples from so me of the other stations allowed adjustments to be made. Significant differences in sulfate and sulfide concentrations at the wee wells were noted in some cases, probably due to seawater infiltration since parts of the catch ment are located close to the ocean or in marina-type developments.

!fereed paper

TIii

• $

-

E :;;

- .,, C31

Bl7

821 816

.

825

C urrently, oxygen injectors are located at E E seven pump stations (TH5, C27 or "BT", ~ ! rJ !ti C l , Bl 7, BI9, B21 & C30) in the Tugun450.

450' ""'' Elanora network. Of these, all except C30 200m t14tm 1484m S73m to m were operational until March 2006.

375 600 81911 -- - ·- - -....... ~ _, Oxygen However, modelling showed that the 21em 162m current location of these injectors is not 450. ; . : _ - - - -o-- - -""' .;;.;..;';.....--<l- -~- 5~9- -MWTP ideal for controlling dissolved sulfide at 21158m 955m Oxygen 4'Mm Elanora WWTP inlet. Figure 6 shows that Oxygen three injection points located closer to the WWTP (as denoted in Figure 4) could Figure 4. Schematic layout of Tug un-E lanora rising main network (Gold Coast) as significan tly reduce dissolved sulphide modelled . Proposed optimal locations for oxygen injectors denoted with arrows. concentrations at the WWTP inlet. The Note the modelled sections refer only to the rising main system serving the Tugu n average dissolved sulfide could be reduced and ad jacent sub-catchments. Other sub-catchments served by El anora WWTP via from ~8 mgS/L to 0.9 mgS/L using the separate rising main systems were not included here. proposed optimal locations. The pred icted mass of disso lved oxygen required Average Concen tration = 8.0 mg SIL (approximately 12 mg/L on ave rage) by injection at the optimal Model prediction g5 percentile Concentration = 13.5 mg SIL ::::, locations co rresponded closely wirh char actually injected (2005-6 • Measured SuffKie discharge = 112 kg Slday in o, 15 data from Gold Coast Water), assuming che injection efficiency .§_ was in the range 40 to 50%. Pure oxygen gas is inj ected into rhe C .Q pressurised main pipelines, with anecdotal evidence of significant ~c 10 off-gassing at air release valves downstream of th e inj ectors. The ~ C actual efficiency of injection is not known. The assumed value of 0 u injection efficiency was based on an estimate from a similar

.

f

-

t

20r-===-===:--=-:=::-:-:- - -;::::::========~

Cl)

oxygen injector in a different system (see Campaign 3 below) for which actual in situ sewage oxygen concentration data downstream of the injector had been measured.

Figure 7 shows how model predictions can be used to assess the optimal location of the point of oxygen injection. In chis case, che inj ection point located 873 m upstream of the rising main junction from pump station B25 (refer to Figure 4) was selected for illuscracion purposes.

Similarly, modell ing shows chat increasing oxygen doses produces a d iminishing return in terms of lowering dissolved sulfide co ncentrations (Figure 8). Complete elimination of sulfide is unlikely to be practical or financially feasible. The model can also be used to predict the effect of oxygen dosing on net loss of biodegradable COD (or volatile fatty acids) in the sewage due combin ations of competing biological transformations (sulfate reduction, fermentation and oxidation). T his may be important where che WWTP process depends on biodegradable COD for biological nutrient removal purposes. The model is currently being extended to include methane fo rmation in sewers (Guisasola et al., 2008), which may be important for greenhouse gas accounting purposes.

Campaign 3 The 06 rising main was used to refine model calibration and predictions against real-time data for dissolved sul fide and dissolved oxygen residual in che presence and absence of oxygen injection at the pu mp station. A special sampling point, with instruments for on-line measurement of dissolved sulfide and

'O

-=

:i

CJ)

• O+--~~~--~~~-- ~--~~~~---i 0.0

0.5

1.0

1.5

2.0

Time (da y)

Figure 5. Model vs. measured values (Campaig n 2 ) for dissolved sulfide at Elanora WWTP inlet works.

20

::::, u'i c,

.§_

=========---- -- - - - - - - - , -

Baseline Current 0 2 Injection

15

C

.Q

T§

c

10

~

§

u

Cl)

'O

-=:i

5

CJ)

0.0

0.5

1.0

1.5

2.0

Time (day)

Figure 6. Model predictions for dissolved sulfide at Elanora WWTP inlet, comparing Baseline (no oxygen, as per Figure 5) with Current a nd Proposed optimal locations of oxygen injectors. Proposed optimal locations of oxygen injectors are shown with arrows in Figure 4.

Journal of the Australian Water Association Water MARCH 2008 135

technical features !fereed paper

5..-- - - - - - - - - -- ----------,

--eI - • -

C

Average 95 percentile

I

/

/

'§

---- -----·-----.,...,,,.

c

g

2

0 (..)

, ,,.

rJl

*

,.

C

Q)

0,

/

/

0

/

-2000

-1500

-1000

\

50

c::, 0

' ....

' " . . . ... -....

E

<(

500

-500

1000

-.... -....

-....

0

0-1----------- ---------~ 0 -2500

~

''

100

~ 0

/ /

Average

- - 0 - - 95%ile

\

£

/

·------------------- •"

1

C, 150

/

----

\

(I)

/

(I)

'O

!i: ::,

'

"'

/

/

200 ~

:!::' 0)

I

/

3

.Q

~

1500

-- -.... -....

6

4

0

8

10

Average Sulfide Concentration (mg SIL)

Location (m from reference point) 0

2

Figure 7. Sensitivity of model predicted sulfide concentration at Elanora WWTP inlet i n response to moving the location of one oxygen injector [location of the two other injectors unchanged) . Reference point (0 m) is the junction with B25 rising main (see Figure 4 ).

20

, - -- ---------:======::;-i •

J

Simulated SCAN Measured

for the Tugun-Elanora system to achieve average (or 95%) compliance with selected dissolved sulfide conce ntrations at Elanora WWTP inlet. Note: Amount of oxygen injected (kg/d) in this figure assumes dissolved oxygen excluding potential oxygen losses to gas phase d ue to dosing inefficiency. Dosing locations are those proposed as optimal (see arrows in Figure 4). :::; iii

20 -

.s

rJl

•

Simulation Results SCAN Measured

C

0)

2 ~ c

§.

C

.Q

~ g

10

10

Figure 8. Model-predicted dissolved oxygen dose requirements

0,

15

8

6

4

95%ile Sulfide Concentration (mg S/L)

~

u

Q)

n

15

10

:2 "= ::,

8

(/)

al>

i i5

5

ni

~

4

0

Time (day)

0 0

10

15

20

Time (days)

Figure 9. Comparison between o n-line measurem ents an d model predictions of dissolved sulfide for D6 rising main (Campaign 3) without oxygen injection.

12

I

-

Probe Measured

Simulation Results

oxygen, was sec up 735 m downstream of che pump station in a single rising main before it joins ocher rising mains in a larger network chat pumps to Coombabah WWTP (Gold Coast Water).

I

10

~ .§.

.

8

I

-~ ~

~ 5

U 0 0

Figure 10. Comparison of mod el predictions w ith o n-li ne d issolved sulfide measu rements for D6 risi ng main (Campaign 3) with oxygen injection .

.

..

The results are summarised in Figure 9, Figure 10 and Figure 11.

'

I

6

'

4 11

II

I

l ,---...-

10

15

20

Time (days)

Figure 11. Compa rison of model predictions with on-line dissolved oxygen measurements for D6 rising main (Campaign 3) with oxygen injection.

136 MARCH 2008 Water Journal of the Australian Water Association

The model predictions for Campaign 3 showed very good correlation with o n-line data. In chis campaign, an in-line flow meter was also available to provide accurate flow data. The results from Campaign 3 increased confidence in che application of model co full-scale sewer systems. It also highlights the value of on-line real-time data (e.g. sulfide fro m the S:CAN® sensor and flow data from an in-line meter) for model verification purposes.

Moreover it suggests chat input data arising from grab sampling and calculation of flows from pump run d ata (as in Campaigns I and 2) pose inherent limitations for model calibration. Although beyond the scope of this paper, extensive tests (both laboratory and field) were conducted for verification of the S:CAN sensor sulfide resulcs against independent analysis of parallel samples from the same system using ion chromatography. For total dissolved sulfide, dependence of the S:CAN sensor results upon

tecnn1ca1 reatures ,fereed paper

accurate simultaneous pH measurement was highlighted in these cescs.

assistance with the use of the S:CAN on-line dissolved sulfide sensor.

Summary and Conclusions

The Authors

A new dynamic model has been developed at the University of Queensland that has imp roved capability for prediction of the transformation processes in sewers that lead, amongst other effects, to the formation of hydrogen sulfide. T he model uses both kinetic expressions and eq uilibrium chemistry to sim ulate biological, chemical and physical transformations chat occur in sewage when transported along sewer pipelines. le was developed both as a research tool and in collaboration with industry partners chat face significant operation and maintenance issues associated with odour and corrosion arising from sulfide formation in sewer systems. In chis paper, che successful application of the model to simulate rising main systems on the Gold Coast was demonstrated and tested in the case of systems with oxygen injection intended fo r control of dissolved sulfide and odour. The model can also be applied to gravity pipel ines. The modelling of gravity sewers also calls fo r quantitative relations about gas transfer between the sewage and the air in the headspace where the pipes are partially filled.

Keshab Sharma, David W de Haas (email David.deHaas@GHD.com.au), Jurg Keller and Zhiguo Yuan are all at che

In the following paper (de Haas et al., 2008), further application of che model for ocher sulfide control strategies in rising mains will be examined, along with financial considerations.

Acknowledgments The project was fu nded by an ARC Linkage grant, with Gold Coast Water and Sydney Water as industry partners.

We thank Ian Johnson (Gold Coast Water) for technical assistance with the field work and Dr Beatrice Keller-Lehmann (AWMC, University of Queensland) for support with laboratory analyses.

DCM Process Control is gracefully acknowledged for providing technical

Advanced Water Management Centre (AWMC), University of Queensland, Brisbane; Shaun Corrie and Kelly O'Halloran are with Gold Coast Water.

References Boon AG and Lisrer AR. ( 1975) Formarion of sulfide in rising main sewers and its prevenrion by injection of oxygen. Prog. Wat. Technol. 7(2), 289-300. et al. (2008) Paper (2) AWA Water journal, this issue.

De H aas

De Lomas JG, Corzo A, Gonzalez JM, Andrades JA, Iglesias E, and Monrero MJ . (2006) Nitrare promotes biological oxidation of sulfide in wastewaters: Experiment at planrscale. Biouchnol. Bioeng. 93, 801-811.

Freudenthal, K. , Koglatis, J., Otterpohl, R. and Behrendt, J . (2005) . Predicrion of sulfide format ion in sewer pressure mains based on the [WA Anaerobic Digesrion Model No. I (ADM I). Water Sci. Technol., 52( 10-11 ), 1322.

Gujer, W., Henze, M., Mino, T., Matsuo, T., Wentzel, M . C., and Vonmarais, G. (1995).

The Activated-Sludge Model No-2 Biological Phosphorus Removal. Water Sci. Tech., 3 1(2), 1-11.

Hobson J. and Ya ng G. (2000) T he ability of selected chemicals for suppressing odour

developmenr in rising mains. Water Sci. Technol. 4 1(6), 165-173.

Hvitved-Jacobsen T. (2002} Sewer Processes: Microbial and Chemical Process Engineering of Sewer Networks, CRC Press, USA Hvirved-Jacobsen T , Vollerrsen J and Tanaka N .

(1998) Wasrewater qualiry changes during transport in sewers - an integrared aerobic and anaerobic model concept for carbon and sulfur microbial transformarions. Water Sci. Technol. 38( I 0), 257-264 Guisasola A, de Haas D , Keller J and Yuan Z. (2008) Methane Formation in Sewer Systems. Accepted for publication in Water

Research.

Keller-Lehmann B, Corrie S, Ravn R, Yuan Z, and Keller J. (2006) Preservat ion and simultaneous analysis of relevanr soluble sulfur species in sewage samples. Poster presentation, Sewer Operation and Maintenance 2006 Conference, Vienna, Austria, 27-28 October.

Latham, B. ( 1873) Sanitary engineering-A guide

to the Construction of Works ofSewerage and House Drainage. E & FN Spon Led, London .

Nielsen AH, H virved-J acobsen T, and Vollerrsen J. (2005) Kinerics and stoichiometry of sulfide oxidation by sewer biofilms. rl7ater Res. 39, 4119-4125.

O' Brien, D.J ., and Birkner, F. B. (1977) .

Kinetics of oxygenation of reduced sulfur species in aq ueous solurion . Env. Sci. and Tech., 11, 1114-1120.

Pomeroy RD and Parkhurst JD . (1977) The forecasting of sulfide build-up rates in sewers.

Prog. Wat. Tech. 9, 621-628.

Sharma KR, Yuan, Z ., de Haas D, H ami lton G, Corrie S, and Keller J. (2008) Dynamics and dynamic modeling of H 2S production in sewer systems. Accepted for publicat ion in

Water Research.

Water Advertising To reach the decision-makers in the water field, you should consider advertising in Water Journal, the official iournal of Australian Water Association.

For information on advertising rotes, please contact Brian Roult at Hallmark Editions, Tel (03) 8534 5000 or email brian.rault@halledit.com.ou

Thistlethwayte D. ( 1972) The Control ofSulfides in Sewerage Systems. Butterworth, Sydney.

USEPA (1974) Process Design Manual for Sulfide Control in Sanitary Sewerage Systems. USEPA 625/1-74-005, Washington DC. WEF (2004} Control of Odors and Emissions from Wastewater Treamient Plants. WEF Manual of Practice 25, Water Environment Federation, Virginia.

Yuan Z, Sharma KR, de Haas D, Hamilton G, Corrie S, and Keller J. (2006) Dynamics and dynamic modeling ofH 2S production in sewer systems. Oral paper presented ar IWA Specialised Co nference on Sewer Operation and Maintenance, 26-28 Oct 2006, Vienna.

Journal af the Australian Water Association

water

MARCH 2008 137

-

...

•

technical teatures .-' Q ct g ~ r . rn..~-n a -· . . ,._

.f

.

-

. • _.,,'

-":.. ~- i,'';

g

ODOUR CONTROL BY CHEMICAL DOSING: A CASE STUDY D W de Haas, K Sharma, S Corrie, K O'Halloran, J Keller, Z Yuan Abstract The aim of this paper was to perform a financial analysis of chemical dosing options for the Tugun-Elanora rising main sewer system, caking into account the predictions from rhe kinetic model outlined in the preceding paper. The in tention was for chis research to serve as a guide to managers in implementing an appropriate and cost-effective chemical dosing strategy.

The approach could facil itate more derailed engineering of odour control in chis and similar rising main systems.

Introduction Sewer rising mains are a significant cause of sepcicicy in wastewater, which commonly results in major problems due to corrosion and odour emissions. In particular, the generation of hydrogen sul fide (H2S) is highly problematic. Ir is generated by sulfate reducing bacteria (SRB) char grow, fo r example, in slime layers (biofilms) on the internal walls of sewers under anaerob ic conditions. SRBs grow in close symbiosis with other bacteria (fermenters and methane formers) in anaerobic environments where organic compounds are plentiful, and together produce compounds causing related issues of odour, environ mental hazard and accelerated infrasrruccure decay. le is seldom feasib le ro completely prevent rhe bacterial biofilm activity rhar leads to problems in sewers. Control strategies usually focus on H 2S since it is a highly odourous (and poisonous) gas rhac can be readily detected and exists in aqueous equilibrium as dissolved (or particulate) sulfide ions. Dissolved sulfide is readily oxidised under aerobic conditions to other forms of sulphur char may be equally problematic (e.g. sulfuric acid chat leads to corrosion of concrete or metal sewer infrascruccure, or solid sulfur chat may cause pipe or pump blockages). The common control strategies for H 2S (and dissolved sulfide) involve dosing chemicals char either oxidise ic to less problematic forms (e.g. rhiosulfare) or "lock" ic into forms char are not volatile (i.e. HS'/S 2 - ions char dominate the eq uilibrium at alkaline pH, or metal precipitates such as iron sulfide). However, dosing chemicals is

138

MARCH 2008

Water

expensive, resulting in operating coses typically in the order of $30 to $100/ML of flow (see below), which is in rhe order of one-sixth to one-third of the operating cost of sewage treatment plants in South-East Queensland (De Haas et al., 2001) .

In rhe past, chem ical dosing regimes for H 2S control in sewers have relied main ly on empirical formulae, based on anecdotal or industry experience and the lireramre (e.g.

WEF, 2004; USEPA, 1985) . Most recently, ki netic models have been developed both overseas (Hvicved-Jacobsen et al. , 1998; Hvirved-Jacobsen, 2002) and locally (Yuan et al., 2006; De Haas et al., 2007; Sharma et al., 2008) that have the capability of predicting sulfide formation in sewers more accurately than emp irical methods. T hese models take into account the dynamics of flow, pipe geometry, retention rime and bacterial growth or chemical processes (such as diffusion into biofilms). T he models have been based on (or tested against) rigorously collected data from laboratory or fu ll-scale sewer systems and open rhe possibility of beccer optimising chemical dosing regimes for odour control.

A recent joint project by the University of Queensland and industry parcners has used a rising main system on che Gold Coast (named che Tugun-Elanora system) as a case smdy for che application of a kinetic model to select and optimise a chemical dosing strategy for odour control. The model development has been described by Yuan et al. (2006) and Sharma et al.

(2008). The rising main system used in the case scudy has also described by Sharma et al. (2008) (this issue) . In summary, the network modelled included almost 14 km chainage of rising main pipelines ranging in diameter from 100 to 600 mm, with thirteen major contributing pump stations, of which seven are equipped wich oxygen injectors (currently nor all are operational), pumping an average dry weather flow of approximately 14 ML/d to the Elanora Wastewater Treatment Plant (WWTP) .

Oxygen probably the most cost-effective if dosing is optimised.

Journal of the Australian Water Association

The aim of chis paper was to perform a financial analysis of chemical dosing options fo r the Tugun-Elanora rising main sewer system, caking into account the predictions from che kinetic model. The objective was to enable Gold Coast Water (GCW) to pre-select the opcion(s) most likely to be cost-effecti ve.

Methodology The kinetic model developed at the Advanced Water Management Centre (A WMC, University of Queensland) was set up for the Tugun-Elanora (T-E) network on rhe Gold Coast, as described by Yuan et al. (2006) and Sharma et al.

(2008). This model includes biological, chemical and some of the physical processes char occur in the sewer system, based on current understanding. The model was used to predict rhe chem ical doses required to minimise odour at Elanora WWTP when rhe sewage is discharged for crearmenc.

The reasons for including the AWMC dynamic model predictions in chis study were as fo llows: • Oxygen dosing had histo rically been used in rhe Elanora WWTP catchment without a clear understanding of whether the dosing locations we re optimal for minimising sulfide emissions at the treatment plant inlet. On-goi ng issues with odour emission, corrosion and high sulfide measurements at the plane suggested that oxygen dosing alone was not proving satisfactory for control of che problem. As described in che previous paper, modelling provided a more convenient and effective means with which co determine whether oxygen dosi ng could be optimised and, if so, how (e.g. dosing location, flow-pro portional and/or timebased dosing control).

• The modelling option for oxygen was cheaper than carrying out field trials, which would involve costly relocation of che oxygen injectors and little opportun ity to optimise location (trial-and-error, lack of flexibility etc.). Hence the model predictions fo r optimal oxygen injection in che Tugun-Elanora sub-catchment, which had been based on derailed field trials for calibration purposes, was used as point of reference fo r comparing other chemical dosing alternatives. This approach evolved

technical features ~fereed paper

odour manag during the course of study; other approaches for comparative purposes might have been equally valid under other circumstances.

Table l. Oxygen dosing assumptio ns based on Mod el (base case compa riso n for

• The use of the model and associated field data collected for calibration and testing highlighted the problem of apparent low efficiency of oxygen injection. This problem had not been documented before

by Gold Coast Water despite anecdotal evidence from operators of gas-release valves operating frequently downstream of the injectors. Allowance for inefficiency obviously strongly influences the consumption and cost of injected oxygen.

Average concentration

• Con ventional experience (published and anecdotal) existed for estimating likely chemical doses for magnesium hydroxide, iron salts or calcium nitrate. The model provided a useful tool with which to test the doses of conventionally used for these chemicals. We are not aware of any other model commonly used for sewer systems in Australia that integrates a kinetic approach for estimating sulfide formation rates with an equ ilibri um chemistry approach for control of dissolved H 2S using pH and/or metal precipitation.

• Data were collected from a laboratorybased small-scale sewer system for the purposes of testing the model. T his study suggested that chemical doses recommended from conventional experi ence (published or other sources) may be either higher (for iron salts) or lower (for calcium nitrate) than that from laboratory data and associated model predictions.

Therefore, both the "experience" and "model" estimates were included to assess sensitivity of the financial analysis to different dose scenarios.

Chemical Dose T he model data was compared to "rules of thumb" for expected chemical doses, based on conventional experience from suppliers, operators and literature sources. The four alternative chemicals compared were: • Dissolved oxygen (injected in pure gas form into the pipeline); • Magnesium hydroxide; • Ferrous chloride; and • Calcium nitrate Note that oxygen inj ection has been used in the past in the T-E network, whereas the other chemicals could be considered possible alternatives to oxygen. For equivalent cost comparison purposes between the alternatives, it was assumed that new oxygen injectors would be installed at optimal locations, based on model predictions.

Sce na rios 1 & 2) vs. Actual (2005 Gold Coast Water) data.

Parameter

Actual (2005)

Model

Unknown Unknown Unknown

17 42 40% 1.40

mg/ Los dissolved 0 2 mg/ L0 2 injected Dosing efficiency (%) (Note 1l mol dissolved 0 2/ mol S

Unknown 369 up to 6

176 440 3

kg/ d os dissolved 0 2 kg/ dos 0 2 injected No.

22 !Notes 2 & 3)

(initial) Doi ly mass dosed No. of dosing points Delivered product active ingredient(% w/w) Delivered product specific gravity

Units

as pure compressed 0 2 gas kg/ Sm 3 (Note 4)

>99.5% 1.429

Note 1: Dosing efficiency assumed (40%), based on model estimates Fram model calibration against full-scale data, where available. Significant loss of undissolved oxygen as gas is suspected to occur via air/ gas release valves downstream of iniection points.

Note 2: Based on oxygen iniector orifice settings (information supplied by Gold Coast Water, Nov. 2005) for 6 no. pump station iniection points, assuming oll were operational {only five were operational in Nov. 2005). Oxygen can only be iniected when the pumps are operating. Actual pump run times for the period modelled (9-10 May 2006) were applied. At the actual doses applied, oxygen dosing was only partially effective in practice in lowering sewage sulfide concentrations at the Elanora WWTP discharge point during the 2005-6 study periods. Refer also to Note 1.

Note 3: Average actual dose range (5 to 60 mg/ L) varied significantly between pump stations, depending on the settings for oxygen gas iniection role ond flow rates through respective pump stations. The tabulated value is the flow-weighted average.

Note 4: Gas phase density of compressed oxygen= 1.429 kg/ m3 at STP {0°C, 1 aim). Sm3 refers to "standard cubic metre" at o·c, 1 aim.

Table 2. Mag nesium hydroxide dosing assumptions bosed on Experience (Scenario l) vs. Model (Scenario 2) data.

Chemical & Dose

Average concentration

Doily mass dosed No. of dosing points Typical delivered product active ingredient (% w/w) Delivered product specific gravity

Experience

Model

Units

100 150 87 150 1432 2

69 103 60 103 986 2

L/MLos delivered product mg/L as delivered prod uct (Note 5) mg/Las Mg(OH)z mg/L as CaC03 L/d as delivered product No.

58%

as MglQHb kg/ L

1.5

Note 5: Range from industry experience: approximately 50 to 150 mg/ Lmagnesium hydroxide as delivered product to achieve 80 to 90% H2S reduction respectively.

The model dynamic flow inputs (based on real operating data collected from the various pump stations in the TugunElanora system) resulted in a totalised daily flow from this system of 14.32 ML/d. In the absence of any chemical dosing, the model predicted an average total dissolved sulfide (H 2S/HS·/S 2·) concentration of 8.8 mgS/L at the discharge to Elanora WWTP. The model predictions in respect of key parameters (e.g. VFA, to tal dissolved sulfate and sulfide) had been previously calibrated to real measurements taken at several sampling points in the Tugun-

Elanora system (De Haas et al. , 2007; Sharma et al., 2008 in this issue) .

The chemical dose required to minimise odour was based on a benchmark target of an average total dissolved sulfide co ncentration at the discharge ro Elanora WWTP of< I mgS/L fro m model predictions or experience. ln the case of magnesium hydroxide (which does not remove dissolved sulfide but shifts its species eq uilibrium disrriburion with pH), the adopted target was < I mg dissolved H 2S. T he model predicted an average dissolved of sulfide concentration of 0.3

Journal of the Australian Water Association

water

MARC H 2008 139

technical features fereed paper

mgS/ L, at a median pH of 8.6, for a dose of just over 100 mg/L as delivered magnesium hydroxide product.

Table 3. Ferrous ch loride dosing assumptio ns based on Experience (Scenario 1) vs.

As a "sanity check", the model predictions were compared with either accual (historical) oxygen doses for this network, or, in the case of ferrous chloride, magnesium hydroxide and calcium nitrate, with literature and experienced-based estimates of che likely chemical doses. Due to the many different fo rmats of dose units in the "rules of thumb" found from the li terature or industry experience, several units have been listed for comparative purposes, where possible. A summary of the adopted doses (Model vs. Actual in the case of oxygen or Model vs. Experience in the case of the ocher chemicals) is presented in Tables I to 4.

Chemical & Dose

A comparison was made between the various options in cwo scenarios for chemical dosing, as listed in Table 5. The reason chat model data was used for oxygen dose assumptions in both scenarios is chat the actual Gold Coast Water data (from 2005, refer to Table I ) came from a period when the average sulfide concentration at Elanora WWTP was not reduced to che benchmark level applied in chis study, namely an average dissolved sulfide concentration of <l mgS/L. To keep the cost comparison valid in all scenarios, the chemical dose applied was expected to produce chis benchmark. The model suggested that the main reason fo r oxygen injection being non-ideal and failing to meet the chosen benchmark for controll ing odour at the WWTP is that the pump stations fitted with oxygen injectors are located too far up the catchment. Oxygen is rapidly consumed and can only be injected when the pump stations are running. W ith sufficient retention time in the rising pipelines downstream of the points of oxygen injection, sulfide reforms and produces odour at the WWTP inlet.

Note 6: Indicative range from industry experience: -30 to 50 mg FeC/2 as delivered product per mg S (initial).

Note 7: Optimal value suggested by WEF /2004) .

Note 8: Laboratory data used for model suggests range 0.5 to 1.0 (average -0.7) mo/ Fe (dosed)/ mo/ S (removed). Literature and industry experience suggests that around 2 mo/ Fe/mo/ S (initial) is required in practice.

Ic is worth noting chat both oxygen and nitrate control sulfide by oxidation. In the case of oxygen, both biological and chemical oxidation mechanisms take place, whereas for nitrate only the biological mechanism occurs. The biological mechanism is slower and located mainly in the biofilm, to che extent that dissolved oxygen (or nitrate) from the bulk liquid penetrates che biofilm and sulfide-ox idisers are present, with oxygen or nitrate serving as electron acceptor. The chemical oxidation mechanism for oxygen is faster and cakes place mainly in che bulk phase and to a limited extent in the biofilm.

Hence, research has shown chat longer retention times are required to achieve the 140 MARCH 2008

Water

Model (Scenario 2 ) data .

Average concentration

Experience

Model

203 250 70 31 28

79 97 27 12 11

Units

3.5 2.0

1.36 0.78

L/ML as delivered product mg/Las delivered product mg/L as FeCl2 mg/Las Fe kgFeCl2 delivered product/kg S (initial) (Note 6) kgFe/kgS (in itial) (Note 7) mol Fe/mol S (initial) (Note 8)

2908 2

1134 2

L/d as delivered product No.

Daily mass dosed No. of dosing points Typical delivered product active ingredient(% w/w) Delivered product specific gravity

28% 1.23

as Fe(Clb kg/L

Table 4. Calcium Nitrate dosing assumptions based on Experience (Scenario 1) vs.

Model (Scena rio 2) data.

Chemical & Dose

Average concentration

Daily mass dosed No. of dosing points Typical delivered product active ingredient(% w/w) Delivered product specific gravity

Experience

Model

Units

39 57 29 5.0 0.57 4.4 1.30 554 2

60 89 46 7.8 0.89 6.9 2.0 865 2

L/MLas delivered product mg/Las del ivered product mgCa(NO3)2/L mgNO3-N/L kg NO3-N/kg H2S-S LCa(NO3)2 solution/kg S (Note 9) mol NO3-N/mol S initial (Note 10) L/d as del ivered product No.

8.75% 1.48

as NO3"N (Note 11 ) kg/L (Note 11 I

Note 9: Range -5 to 7.5 LCaN03 solution/ kg S (WEF, 2004) Note 10: Laboratory data used for model suggests range -0.6 to 1.6 mo/ N03-N/mol S (removed) for oxidation of sulfide alone, with partial oxidation at the lower doses to some intermediate with a lower oxidation state than sulfate (e.g. polysulfide or molecular sulphur). Increasing sewer retention times tends to promote anoxic growth of heterotrophic biofilms with sustained nitrate dosing, which tends to increase nitrate consumption with liHle added benefit of greater sulfide oxidation. For practical purposes, the molar ratio of nitrate dose to sulfide oxidised is likely to be limited to a maximum of approximately 2 mo/ N03-N/mg S. For example, WEF /2004) suggest doses in the range -5 to 7.5 LCa{N03/2 solution/kg S {initial), which corresponds to a molar ratio of -0. 9 to 1.3 mo/ N03-N/mg S for the stated concentration of solution (42% as calcium nitrate).

Note 11: Calcium nitrate delivered product typically ranges 8.4 to 9.1 {adopt 8.75) w/ w active NOr N. SG = 1.45 to 1.55 (adopt 1.48) kg/L. The adopted values are equivalent to 51 % w/w as Ca{N03)2. equivalent degree of sulfide oxidation by nitrate than by oxygen. For chis reason the ideal dosing point for nitrate needs to be further back in a rising main system than for oxygen, relative to the odour-sensitive discharge point.

Journal of the Australian Water Association

An unfortu nate consequence of the longer retention rime required for sulfide oxidatio n by nitrate (compared co oxygen) is that greater consumption of nitrate occurs by competing growth from heterotrophic bacteria (non-sulfide oxidisers). Noc only does chis necessitate a

technical teatures [i]

higher nitrate dose than oxygen (when compared on an electron-equivalent basis) bm it may result in thicker biofilm growth.

Ironically chis may lead to accelerated sulfide production due to sulfide reduction in rhe event char the nitrate is depleted and anaerobic conditions resume.

.fereed paper

Table 5. Chemical doses applied in scenarios for cost comparison.

SCENARIO

CHEMICAL

Scenario 1 Scenorio 2

Oxygen

Magnesium Hydroxide Ferrous Chloride Calcium Nitrate

Model dose (Table 11 Model dose (Table l )

Experience (Tables 2,3,4) Model (Tables 2,3,4)

Assumptions for Cost Comparisons The following assumptions were made in the fi nancial analysis: • Average daily flow (14.32 ML/d) and chemical doses as defined above for the Tugun-Elanora case study.

• Capital cost estimates fo r each dosing station (3 no. for oxygen; 2 no. for each the other chemicals) included road easement to allow access for ranker delivery, concrete slab, security, lighting and other services (power, water, drain age etc.) for the dosing facility. In the case of ferrous chloride, due to the higher chemical use and delivery requi rements, a larger bulk storage rank (30 kL) in a bunded area was allowed for ch is chemical. For the other liquid chemicals (i.e. except oxygen), a smaller storage rank size ( 13 kL) with bund provisions to prevent spillage to the environment was allowed. Oxygen was assumed to be injected in gaseous form and delivered from a pressurised tank storing pure liquid oxygen, which is typically supplied and maintained by the suppl ier.

• Operati ng cost estimates included delivered costs of bulk chemicals, hi re of dosing equipment, as well as preventative and breakdown maintenance plus spares.

Routine maintenance of the dosing systems was assumed co occur at a frequency of every six months for all the chem icals.

Table 6. Results of fi nancial analysis for SCENARIO 1.

Chemical Dosed

Capital Cost

Operating Costs (per year)

Net Present Value (at 7% discount rate)

$293,000 $196,000 $357,000 $196,000

$161,700 $563,000 $478,900 $351 ,200

$2,440,000 $7,838,000 $6,834,000 $4,905,000

Oxygen injection Magnesium hydroxide dosing Ferrous chloride dosing Calcium nitrote

• Go ld Coast Water staff costs were assumed co allow for one operacor attending each dosing facility for 2 hours per week (plus 0.5 h travel rime). One operaror was assumed co have a base salary cost of $55,000 pa, with allowance of an additional 25% made for overtime, annual leave and 40% of the base operaror salary cost co cover staff overheads and laborarory services.

• Estimated power costs were relatively insignifica nt bur were taken inco account for compl eteness (approximately $ 1000 ro $2000 pa).

• The analysis was carried ou r over a 30year rime period (2007-2037), with civil structures assumed co have a life of 30 years and mechanical and electrical equipment a life of 15 yea rs. Replacement of the dosing equipment was not separately included since these are typically costed on a "hire" basis from the chemical suppliers.

NPV Analysis for SCENARIO 1 at 7% discount rate $11,000,000

r;===========--=============:;-, __. Ferrous Chloride (2.0 mol Fe/mol S)

._.Magnesium Hydroxide (100 L producUML)

-+- Calcium Nit rate (1 .3 mol NO3-N/mol S) -+- Oxygen (1.4 mol O2/mol S, 40% efficiency)

a, ::,

~

....C

$7,000,000 +-- - - --+-- - - - - - - - - - - - - - - - - --<

a,

"'e

11.

$5,000,000 +-- ---1-----,'---------------::,.......=---- - - - - --i

I

ai

z

I

$3,000,000

/

//

1

___,-~ --- - - - - - - - - - - - - ---j

-t---,-----,.-/ ____

.,,,.,.----

.,...

S1 ,000,000

~

.... -- --------02

~--~-------------~-------< so S200 $400 $600 $800 $1 ,000 S1,200 $1,400 Cost of delivered chemical($/ tonne liquid product or $/1 000 Sm3 for pure oxygen)

Figure 1. Chart showing Net Present Value and sensitivity to delivered cost of bulk chemicals for the four options of chemical dosing in the Tugun-Elanora rising main case study defined as Scenario 1 .

Maintenance an d replacement costs of the dosing facilities were therefore covered in the hire and maintenance component of the operating cost for these units.

• Chemical dosing costs were escalated annually in proportion co flow rare. The flow rare through the sewer network was assumed co increase at a rare of I% pa, which is very close co the weigh red average projected by Gold Coast Water planners for the Tugun-Elanora catchment in the period adopted for the analysis here (2007-2037) .

• Inflation was excl uded from the fi nancial analysis.

• The discount rate in the Net Present Val ue (NPV) analysis was assumed co be 7%.

Results The resu lts of rhe NPV analysis are summarised in Table 6 and Figure 1 for the following dosing assumptions, named SCENARIO I (refer co Table 5 for definicion): • Oxygen injection at 12 mg/L dissolved 0 2 (1.4 mol 0 2/ mol initial H 2S), or 440 kg/d injected at 40% efficiency • Magnesium hydroxide at I 00 L delivered product/ML (87 mg/Las Mg(OHh, equivalent co I 50 mg/Las CaC03) • Ferrous chloride at 203 L delivered produce/ML (2.0 mol Fe/mol S as initial

H 2S) • Calcium nitrate at 39 L delivered product/ML ( 1.3 mol N0 3-N/mol Sas initial H 2S) Ir is apparent from Table 6 that the whole of life cost (or NPV) for chemical dosing co control odour is heavily dominated by the operating cost, which, in rum is dominated by cost of the delivered chemical consumed.

In fact, for all rhe options in Scenario I, approximately 90% co 98% of the N PV is contributed by operating and maintenance

Journal of the Australian Water Association

Water

MARCH 2008 141

technical features [ii ?fereed paper

odour manag (O&M) costs. The cost of the chemicals contributed approximately 70% to 80% of the O&M cost, depending o n the chemical and projected year.

The results of th e fi nancial analysis for Scenario 1 suggest that oxygen dosing is the most cost-effective option for odour con trol, assuming the dosing regime can be optimised in p ractice to produce the target outcome defined (<l mg/L dissolved H 2S).

For the dosing assu mptions based on "Experience" (Scenario 1), calcium nitrate dosing was th e next most cost-effective option after oxygen, followed by ferrous chloride and magnesium hydroxide (the most expensive option).

Figure 1 sh ows that for the assumed dose rates in Scenario 1, oxygen is the most costeffective ch emical over the whole range of likely currenr costs of delivered b ul k chemical. T he cost range examined was the approximate (indicative) delivered price of chemical in South East Queensland ±20%, based on Gold Coast Water information and industry experience. It is obvio us chat if the cost of chemical is heavily d iscounted (dotted trend line on the extreme left in Figure 1), then the NPV for all options cluster around a rough common poim, reflecting che similarity in capital and operating coses (ocher than chemical) between optio ns.

In order to further test the financial model sensitivity to chemical dose, SCENARIO 2 was examined, based on the following dosing assumptions (refer co Tab le 5): • Oxygen injection at 12 mg/L d issolved 02 (1.4 mo! 0 2/ mol initial H 2S), or 440 kg/d injected at 40% efficiency (same as Scenario 1) • Magnesium hydroxide at 69 L delivered produce/ML (60 mg/Las Mg(OH h, equivalent co 103 mg/Las CaC0 3) • Ferrous chloride at 79 L delivered produce/ML (0.78 mo! Fe/mo l Sas initial

H 2S) • Calcium nitrate at 60 L delivered produce/ML (2.0 mol N03-N/mol S as initial H 2 S) The financial resul cs fo r Scenario 2 are presemed in Table 7 and Figure 2.

From the results in Table 7, oxygen remains the most cost-effective chemical in Scenario 2, but calcium nitrate becomes the most expensive (similar to magnesium hydroxide). Ferrous chloride is the next most cost-effective after oxygen, assuming the lower dose (based o n laboratory/model observations) works in practice at fu ll-scale.

Similar conclusions may be drawn from Figure 2 for the range of bulk chemical costs considered.

142 MARCH 2008

Water

NPV Analysis for SCENARIO 2 at 7% discount rate $11 ,000,000

r;:==::=========================i ....- Ferrous Chloride (0.78 mol Fe/mol S) -+- Calcium Nitrate (2.0 mo\ NO3-N/mol S) -M-' ag~ne_s_ l um _ H~ yd_ro_xl_ d e~ (_ 6 9_L...c.p_rod _ u_c_ VM_L-'-)_ I~---

-+__ O...c xyg -=en (1.4 mol O2/mol S, 40% efficiency)_

$9,000,000

t-- - - - - - - - - - - - - - - - - - - - - - - ---1

I

$7,000,000

t---- - - - - - - - - - - - - - - ~,-c._ ': 1/ ______---l

n.

$5,000,000

~

~

Ca(N0 3),

./

Mg(OHI, _ //

~

/

_/ .

1--- - - - - - - - ,~-- ~ -- ~ ~ - - - - - - - - - - - ----l

j

Fe~

~

~~ / /

~ _.,. ~

$3,000,000

~~,,,.

~

If' -------$1,000,000 f - - - - - - - - - - - - , - - - - - , - - - - , - - - - - , - - - - - - ! $0

$200

$400

$600

$800

$1,000

$1,200

$1,400

Cost of delivered chemical ($/ tonn e liquid product or $/1000 Sm3 for pure oxygen)

Figure 2. Chart showing Net Present Va lue and sensitivity to delivered cost of bulk chemicals For the four options of chemica l dosing in the Tu gun-Elanora rising main case study defined as Scenario 2.

Table 7. Results of financial analysis For SCENARIO 2.

Chemical Dosed

Oxygen injection Magnesium hydroxide dosing Ferrous chloride dosing Colciurn nitrote

Capital Cost

Operating Costs (per year)

Net Present Value (at 7% discount rate)

$293,000 $196,000 $357,000 $196,000

$16 1,700 $418,800 $247,900 $492,300

$2,440,000 $5,841,000 $3,634,000 $6,859,000

Discussion The results of the finan cial analysis are surprising since conventional "wisdom" at the scare of che research project was chat one or more of che alternative liquid chemicals for odour control would be more cost-competitive than injecting gaseous oxygen. Moreover, oxygen inj ection was generally perceived by Gold Coast Water to be of dubious cost-effectiveness and presemed certain occupational health and safety challenges. However, the model developed as pare of the research project suggested chat the positioning of che oxygen injectors relative to the system retention time and odour sensitive discharge location is critical to che effectiveness of oxygen.

The same is true of nitrate dosing.

T he financial analysis shows chat actual (or assumed) chemical consumption will be critical in choosing the most cost-effective chemical. Although oxygen and nitrate may have lower costs (in Scenario 1), both have the d isadvantage chat they are rapidly consumed by hetero crophic bacterial activity in the sewage and/or biofilms on the sewer pipe inner walls. For example the oxygen (or nitrate) uptake races per unit biofilm area (pipe wall) determined from laboratory studies and used in the model are approximately 16 g0 2/m 2.d (12 gN/

Journal of the Australian Water Association

m 2 .d for nitrate) at temperatures of approximately 23 (± 1) 0 C. Therefore, under quiescem conditions (when rising mai n p ump station pumps are not running) when oxygen (or n itrate) cannot be injected, the rapid consumption of oxygen or nitrate results in growth of hecerocrophic bacteria and loss of biodegradable COD in che sewage. In the long term ch is can have negative consequences, which to some ex ten c reduces the effectiveness of dosing oxygen (or nitrate), namely: • Sulfide generation (due to SRB activity) recommences in any case under anaerobic conditions when oxygen (or nitrate) is depleted.

• More growth of heterocrophic biofilm can accelerate oxygen (or nitrate) uptake races, resulting in more rapid return co anaerobic conditions and sulfide generation. Therefore oxygen dosed coo far back in che cacchmem is largely ineffectual in controlling sulfide concentrations at the end ofche pipeline (e.g. at the WWTP) and ironically may even be cou nter-productive.

• Loss of biodegradable COD may result in loss of nitrogen (and/or phosphorus) removal pocemial at the WWTP if rhe process relies on b iological nucriem removal.

technical features [i] ~fereed paper

odour manag The advantage of dosing iron sales or magnesi um hydroxide is char they give a "storage" effect for controlling sulfide. In che case of iron sales, a precipitate is formed when added to water (or wastewater). In the absence of sulfide, the precipitate will be some form of iron-hydroxy-phosphate (Fe~ P~OH). Research as pare of chis project has shown chat when sulfide is subsequencly fo rmed (or added), sulfide displaces phosphate (or hydroxide) from the Fe~P~OH precipitate since che iron-sulfide precipitates are thermodynamically more stable. Effectively, the Fe ~P~OH precipitate acts as "reservoir" of iron char can be transported down the sewer and remains present co control dissolved sulfide if anaerobic conditio ns develop. If the Fe~P~OH precipitate is nor completely used up, ic will serve a useful fu nction in providing supp lementary chemical P removal in che WWTP. Similarly, magnesium hydroxide "locks" dissolved sulfide in the non-volatile ionic forms (mainly HS¡ with traces of S2¡) at a pH in rhe range ~8.0 co 9.0 irrespective of whether conditions are aerobic, anoxic or anaerobic. Moreover, rhe chemical dosed will deliver useful alkalinity to rhe WWTP. Many WWTPs require alkal inity supplementation in any case to achieve stable biological nitrogen removal, especially where supplementary chemical P removal (typically with alum or iron salts) is practiced. Magnesium hydroxide dosing in the catchment will therefore at least partially offset supplemen cary alkali dosing ac the WWT P. One of che key uncertainties in rhe costeffectiveness of oxygen is che efficiency of its injection. Inefficient (coarse bubble) injection direccly into che rising main pipeline can result in significant loss of undissolved gas from air release valves downstream (or manholes and vent pipes where rhe system incorporates gravity sections). In chis study we were unable to di reedy measure the efficiency of oxygen injection bur the models suggested char it lies in the range approximately 20 co 50%.

For the cost comparisons presented in above (Scenarios 1 & 2), an injection efficiency of 40% was adopted , based on modelling a full-scale rising main pipeline where residual dissolved oxygen was measured in situ. If che efficiency is lowered to 20%, the outcome of the finan cial analysis is nor changed. However, che margin in whole of life cost (N PV) difference between oxygen injection and the next cheapest option (nitrate dosing in Scenario 1) is reduced from $1.8 million (abour 40% cheaper) co approximately $0.3

million (6% cheaper). In Scenario 2, ferrous chloride dosing might be costcomparable to oxygen (or marginally cheaper on an NPV basis) if oxygen injection efficiency is as low as 22%. More careful investigation of rhe efficiency of oxygen injection (e.g. using updated and better diffuser systems) seems warranted.

The position of oxygen (o r nitrate) injection relative to che target point for odour control (e.g. WWTP) is critical and needs to be based on dynamic model sim ulation.

Conclusion Dependi ng on che emphasis on cost minimisation by Gold Coast Water, a fullscale trial of one or more of the alternative chemicals to oxyge n is warranted. The fina ncial analysis shows char oxygen is likely to be che most cosc-effecrive if ch e dosing efficiency and location are optimised. On balance, ferro us chloride may be che next best alternative co oxygen, given char ic may have indirect cost offsets against savings on chemicals for supplementary P removal in the WWTP and is nor as labile as oxygen.

Laboratory data from chis research project suggest char rhe iron dose requi red may be lower than suggested by suppliers or manuals of practice (" industry experience").

To rest chis premise and truly minimise cost at full-scale, it will be important co optim ise not only dosing location but also iron dosing control based on dynamic model predictions. The objective will be co ensure a sufficiencly high dose of chemical for diurnal periods (o r "slugs of wastewater pumped into the rising main) when retention rimes and su lfide production are expected to be high. To th is extent, simulation using dynamic models is an invaluable tool.

Finally ic should be noted char chis case study was focussed on a rising main sewer system. Sulfide productions races in gravity sewers fo llow rhe same mechanism bur the net production rares will differ for partially fi lled pipes, particularly if significant gas transfe r (i ncluding oxygen) occurs. Oxygen dosing may be less appropriate for gravity sewers, since these are not pressurised bur vented co the atmosphere.

Oxygen so lubility limitations as well as losses of oxygen and stripping of gases to atmosphere are therefore likely to be problematic. Nevertheless, with che exception of oxygen, on a relative basis the findings made in chis paper regarding chemical dosing for hydrogen sulfide co ntrol are expected to be si milar fo r gravity mains.

Acknowledgments This project was funded by an ARC Linkage Grant, with Gold Coast Water and Syd ney Water as industry partners.

The Authors

Keshab Sharma, Zhiguo Yuan and Jurg Keller are all ar che Advan ced Water Management Centre (AWMC), University of Queensland, Brisbane. Shaun Corrie and Kelly O'Halloran are wich Gold Coast Water while David de Haas (emai l david.dehaas@ghd.com.au) is with GHD consulting engineers in Brisbane and parttime ar the AWMC.

References De Haas OW, Hertle C K, Hamlyn-Harris D and Walpole R. (2001) The performance of l'\vo oxidation ditch sewage treatment plants in South-East Queensland. Proceedings of the 2nd World Water Congress of the International Water Association, Berlin, Germany, I 5-19 October 200 I. De H aas OW, Sharma KR and Yuan Z. (2007) Application of a new model for predicting hydrogen sulfide formation in sewer systems.

Proceedings Australian Water Association Ozwater 2007 Conference, Sydney, Australia, 5-7 March 2007 .

Hvirved-Jacobsen T. (2002) . Sewer Processes: Microbial and C hem ical Process Engineering of Sewer Networks, CRC Press, USA H vitved-Jacobsen T, Vollertsen J and Tanaka N. ( 1998). Wastewater quality changes during transport in sewers - an integrated aerobic and anaerobic model concept for carbon and sulfu r microbial transformations. Water Sci. Tech. 38 (10), 257-264 Sharma KR, de H aas OW and Yuan Z (2008) .

Water 35 No 2.

Sharma KR, Yuan Z, de Haas OW, Hamilron G, Corrie Sand Keller J. (2008) Dynamics and dynamic modelling of H 2S production in sewer systems. Accepted for publication in

Water Research.

US EPA (1985). Design manual - Odor and Corrosion Control Sanitary Sewerage Systems and Treatment Planes. US Environmental Protection Agency. EPN 625/1 -85/0 I 8, Cincinnati, Ohio, USA, October 1985, Chapter 3 .

WEF (2004). Control of Odors & Emissions from

Wastewater Treatment Plants - MOP 2.

Water Environment Federation, Alexandria, Virginia, USA, p277.

Yuan Z , Sharma KR, de Haas OW, Hamilton G, Corrie Sand Keller J. (2006) Dynamics and dynamic modeling of H 2S production in sewer systems. Proceedings 2nd International

!WA Conference on Sewer Operation and Maintenance. Vienna, Austria, 26-28 October 2006, p99- I 06.

Journal of the Australian Water Association

Water

MARCH 2008 143

technical features [i]

fereed paper

ODOUR EXPOSURE CRITERIA AND ODOUR MODELLING IN WESTERN AUSTRALIA I Wallis, K Cadee Abstract

Halls Head WWTP - Existing - Oesludglng

The Water Corporation has undertaken extensive odour modelling and community surveys at ten wastewater treatment plants of various sizes in Western Australia to identify an odour criterion that applies at all the plants. An odour criterion of 5 OU (99.9% and 1-hour averaging) has been found to best represent the boundary between acceptable conditions and annoyance about odour for urban, residential areas in Western Australia. An odour classificatio n scheme is p roposed and it that shows that apparen tly different odour criteria from various jurisdictions throughout the world reflect a similar differentiation between acceptable odour and odour an noyance.

&â&#x20AC;˘00000

6399500

.;~

l

6399000

C)

C:

Introduction

:E 't: 0

T he Water Corporation operates 101 wastewater treatment plants in Western Australia, ranging from small plants in country towns treating less than 1 ML/d co major metropolitan plants at Beenyup and Woodman Point treating 120 ML/d. Western Australia is experiencing rapid growth and d evelopment and there is pressure to treat increasing flows at many plants while, at the same time, residences are being developed ever closer to treatment plants.

Inevitably it is difficult to build larger treatment plants (often with more complex wastewater treatment processes and increased recovery of effluent and biosolids for recycling) within a smaller buffer. There is also conflict between the desire of residents (or potential residents) living adjacen t to treatment plants for an environment free from odour nuisance and the needs of the Water Corporation to use available fund s on expanding the treatment capacity and recycling potential of the planes and imp roving energy efficiency (odour control consumes energy).

Over the lase seven years, Water Corporation has faced and resolved chis conflict at many treatment planes of various sizes in Western Australia. In doing so, extensive use has been made of odour modelling and community surveys to identify appropriate criteria to determine acceptable and unacceptable levels of odour.

Our fi ndings are summarised in chis paper.

Odour Perception and Odour Measurement Residents who live near a wastewater treatment plant are continuously monitoring odour levels, even if unconsciously. After they have been annoyed by a series of odour events chat they consider unpleasant or unaccep table, their monitoring of odour can become more co nscious and meticulous.

The odour that they detect is a complex mixture of many odorous compounds. Perception of odour may be influenced by its concentration , intensity, character and hedonic tone (Deffra,

Odour criteria should reflect a distribution of odour concentrations.

144 MARCH 2008

Water

Journal of the Australian Water Association

z 6396500

63ll6000

~=

6397500 ..)....;_.__.s:._.....

376000 376500

377000

377500

370000

Cone. (Odour_Units): 99.9% avg. â&#x20AC;˘ 3 odour conrplamts dunng dosludglng In 2005

Figure 1. Predicted Odour Contours for Halls Head WWTP. Note: red con tour is considered upper level of acceptability.

2006). W hile standard methods exist to define each of these factors, the concentration of odour is most commonly used to measure odours from wastewater treatment plants.

Odour concentration is expressed in odou r units (OU), which is the n umber of times chat a parcel of air must be diluted with odour free air so chat the diluted odour is just at the threshold of detection. Note chat AS 4323.2 defines the unit 'ou' in lower case, but 'OU' in upper case is easier to read. The threshold of detection is che odour level at which 50 % of a panel of trained odour "assessors" can just detect the odour in a laboratory setting (in accordance with AS/NZS 4323.3:2001). Thus 5 OU means chat an air sample must be diluted five-fold with odour free air to reduce the odour level to the threshold of detection.

Just because an odour is detectable d oes not mean that it will cause annoyance or may be classified as offensive or a nuisance. These terms are commonly used in Aces and Regulations and there is well established case law relating to these matters in Australia and che UK. The "FIDOL" factors of frequency (how often an individual is exposed to odours), intensity (odour concentrations), duration, offensiveness (hedonic tone/character) and location (type of land

tecnn1ca1 reatures

odour manag

,fereed paper

use and nature of human activity) are useful in assessing the impact of odours (Deffra, 2006).

There is a significant difference between the concentration of odour that is detectable in rhe laboratory si tuarion and the concentration that cause annoyance or nuisance. A commonly quoted rule-of-thumb (Valentin and North, 1980) is rhar an odour concentration of five times the detection threshold (effectively 5 OU) has the potential to be considered offensive.

This is consistent with the findings of Amore (1985) that H 2S becomes noticeable in environmental settings at about fi ve times irs threshold concentration.

Acceptable and Unacceptable Odour The simplest method of defining unacceptable odour conditio ns is to require that odours should never be detectable at any odour receptor. Such criteria are theoretically based, and com monly involve an odour concentration close to o r at rhe odour threshold of 1 OU, at a high frequency (99.9% of the time) and a short duratio n (three minutes). The standard for odour exposure in Victoria (Vic EPA, 2005) is an example of a criterion that uses this approach.

The disadvantage of rhis si mple approach is that it is unnecessarily stringent, leading to very large buffer zones around wastewater treatment plants (and other odorous premises) or excessive expenditure (and energy use) on odour control. For many metropolitan plants, where rhe buffer zone is fi xed by surrounding development, ir may nor be feasib le to meet such stringent criteria without completely re-building rhe treatment plant with covers on all processes plus a building to enclose rhe covered tanks and to provide the secondary containment needed to allow regular maintenance of rhe plant.

An alternative approach is to establish rhe odour conditions that give rise to annoyance and nuisance and to establish buffer zones and design odour control facilities on this basis. This approach has been adopted by most States in Australia, and in the UK and the Netherlands (UKWIR, 2004). The procedure involves validation of the predictions of odour dispersion modelling against odo ur complaints.

United Kingdom In the UK, the most commonly used odour criteria for annoyance and nuisance is 5 OU/m3, 98% frequency and 1-hour averaging rime, which was rhe value derived from the Newbiggen-by-rhe-Sea public enquiry (Department of Environment, 1993). While comprehensive validation has not been published, che apparent success of chis criterion at Newbiggen-by-the-Sea and other locations in the UK has given weight to the use of chis criterion fo r regulatory purposes.

It is, however, recognised in the UK that there are situations where the population may be more sensitive to odours, particularly due to prolonged exposure (UKWIR, 2004). The most recent guidance from the UK Environment Agency (Environment Agency, 2002a and 20026) for odours from wastewater treatment plants set an indicative criterion of 1.5 OU/m3 98% frequency and 1-hour averaging time, to avoid annoyance.

There are signifi cant differences between rhe responses of individuals to odours. Based on a detailed study of available informatio n including 70 case studies, UKWIR (2004) concluded that below 1.5 OU/m 3 at the 98th percentile of hourly averages there are only sporadic complaints. Consistent complaints did not occur until the 98th percentile of concentration is in che range 1.5 to 5 OU/m3• Above 10 OU/ m3, the number of complaints increases dramatically. Thus, depending on circumstances such as

Broome - Existing WWTP - February 2007

8015000

8014500

8014000

8013500

'iii'

i

8013000

g'

8012500

.s :.E t:: 0

Z

8012000

8011500

8011000

8010500

8010000 416000 416500 417000 417500 418000 418500 419000 419500

Cone. (Odour_Units): 99.9% avg.

• Odour Complaint

Figure 2. Predicted Odour Contours for Broome WWTP. Note: red contour is considered upper level of acceptability.

individual sensitivities, past exposure to odours, community expectations and aspirations, che odour concentration chat gives rise to odour complaints may be in che range 1 to 5 OU/m3, 98% frequency, 1-hour average.

Netherlands The Netherlands Emissio n Guideline fo r Air (InfoMil , 2004) states chat the concentration where rhe number of compl aints increases significancly is 2.5 OU/m3 at rhe 98% frequency and !hour averaging. However, rhe guideline also states that there are situations where concentrations in excess of 5 OU/ m3 at 98% frequency and 1-hou r averaging do not result in odour complaints, while there are situations where concentrations of 1 OU/m 3 at 98% frequency and !-hour averaging can give rise to complaints.

The odour criteria in the Netherlands Guideline are based on the density of population in the surrounding area (essentially a measure of the risk of encou ntering odour-sensitive individuals) and on whether facilities are new or existing.

• For densely populated residential areas, the maximum exposure concentrations are 0.5 OU/ m 3, 98% frequency, ! -hour averaging fo r new facilities and 1.5 OU/m 3, 98% frequency, ! -hour averaging for existing faci lities.

• For sparsely populated areas, the maximum exposure concentrations are 1.0 OU/ m3, 98% frequency, 1-hour averaging for new facilities and 3.5 OU/ m3, 98% frequency, 1-hour averaging for existing facilities.

Where odour concentrations at receptors exceed the max imum odour levels listed above, a series of measures co reduce odour emissions are recommended in the Netherlands Guideline. A 90% reduction in odour emissions can normally be achieved though the application of these controls.

Journal of the Australian Water Association

water

MARCH 2008 145

technical teatures

I [i1

, odour manag

fereed paper

Australian Odour Criteria Percentage of Tim e Odour Concentration Exceeded

,.,.

Odo ur cri teria for urban areas in Australia are generally based on much higher frequencies than in the UK or Netherlands, possibly because of the preference for odour modelling in Australia compared co the European preference for field studies of odours.

99.9% _ __

100

10

~

~

Ir should be noted chat odour emission rares in Australia are commonly measured using an isolation flux hood (generally following the standard flux hood design proposed by the US EPA). In Europe, a variety of sampling methods are used co estimate odour emission rates, including flux hoods, wind runnels and back calculation.

In March 2002, the WA EPA issued a Guidance Statement on the Assessment of Odour Impacts (Guidance Sracement No 47) co defi ne odour objectives, with th e following so-called 'green light' odour criteria: • 2 OU at 99.5 percentile at 3 minute averaging period; and • 4 OU at 99.9 percentile at 3 minute averaging period".

In Victoria, odour criteria are based on either the SEPP limit of 1 OU (Vic EPA, 2005) or a risk assessment (essentially following the validation of odo ur modelling with odour complaints as described in this paper) and the Victorian EPA has accepted rhe following cri teria with supporting risk assessments: • 4 OU at 99.9 percen tile at 3 minute averaging period.

NSW (NSW EPA, 2001a, b) uses a very short averaging period of 1 second and the odour criteria are based on rhe population affected: • 2 OU ar 99 percentile at 1 sec averagi ng period (>2000 persons) co • 6 OU at 99 percentile at 1 sec averaging period (10 persons) Queensland (EPA, 2004) has adopted a peak to mean factor of 2: 1 for area sources such as treatment plants and che odour criterion for area sources (such as WWTPs) is: • 2.5 OU at 99.5 percentile at I -hour averaging period.

SA (EPA, 2003) also has criteria based on the population affected: • 2 OU at 99.9 percentile at 3-min averaging period (>2000 persons) to • 10 OU at 99.9 percentile at 3-min averaging period (12 persons)

146

MARCH 2008

Water

f !

---,,~:...-,::=---+---!-.JW!'-'.-- - -~ 1)

_ _ _ _ _.J o., 1000

100

10

Hours por Yoar Odour Conconrtatlon Exceeded

Figure 3. Differentiation Between Acceptable Odour and Annoying Odour.

Averaging Periods and Percentiles Variations in odour concentrations over time at a receptor are unique to each location and local wind conditions. The averaging periods and percentiles are however related by the properties of the local wind patterns. A power law relationship may be developed to convert concentrations from one averaging period to another.

Figure 3 illustrates the odour concentrations for a I-hour averaging period, for a range of percenti les (from 90% to 99.99%). The green line denotes the upper level of rhe range of acceptable odo ur concentrations and connects reasonably closely the following points: • 1.5 OU at 98 percentile (UK, Netherlands); • 2.0 OU at 99 percentile (Mangere WWTP, Auckland, NZ Resource Consent); • 2.5 OU at 99.5 percentile (Queensland); and • 5 OU at 99.9 percentile (Water Corporation).

At odour concentrations above these acceptable limits there is likely to be annoyance and, above three rimes these limits, there is likely to be severe annoyance to the more sensitive members of the community.

On the ocher hand, at odour concentrations below about one-third of these limits, odour is likely to be undetectable.

By including averaging time as a third dimension, three dimensional surfaces can be generated to represent acceptable odour conditions. It is apparent, therefore, chat the various odour criteria are not as different as they appear, as they rend co be

Journal of the Australian Water Association

within in a three-dimensional zone which represents acceptable odour conditions.

Odour criteria can therefore be developed at any averaging period or percentile and converted to equ ivalent concentrations at ocher averaging periods and percentiles.

While an averaging peri od of 1-hour and the 98th percentile have been adopted in the UK and the Netherlands, the authors prefer to use an averaging period of 1-hour and the 99.9% frequency in Western Australia for the reasons explained below.

Experience in Western Australia The WA EPA Draft Guidance Statement on Assessment of Odour Impacts (WA EPA, 2000) advised that: "Th e EPA deems chat the appropriate averaging time fo r odour impact assessment is one hour, and the corresponding percentile compliance is 99.9%".

The Water Corporation ad opted chis criterion and has used it as the principal odour limit for modelling odour in WA, although the 99.5% limit also is modelled to check compliance with the 'green light' criteria.

An averaging period of one hour is representative of che conditions char will give rise to odour comp laints rather than sho rter time periods where odours may be detected but not give rise to complaints.

Modelling in Western Australia indicates chat the predicted maximum odour concentrations are only slightly higher that at che 99.9 percentile levels. Thus the 99.9 percentile criterion provides a high level of protection against shore events of elevated odour (which could be unacceptable co some people) and can be explained to the community in these terms.

technical features

odour manag Fo r treatment planes near the coast of WA, it has b een fo un d that the 99.9 p ercentile criterion resu lts in a contour d iagram that extends downwind in the directio n of the predominant night d rainage wind, which corresp onds to the usual pattern of co mmun ity complaints about o dour (m ostly with detectio n at n ight).

Calibration of Odour Model The resu lts o f od our modelling obviously refl ect the inpms to the m odel. T he ranking o f inputs to the m od el, in order o f greatest to least infl uence on the predictions, is as fo llows:

II.

Table 1. Pro posed O do ur Classification Scheme for Urban Residentia l Ar e as in W estern A ustralia.

Classification for 99. 9% frequency 1-hour averaging,

Description of Odour Situation

N on-detectable

Odours are non-detectable by most, if not at all, individuals and acceptance of ambient air quality by the community is very high.

(< 1.5 OU) Acceptable

O dours, if detected at all, ore so faint and infrequent , that they give rise at most to a few spasmodic complaints. The level of acceptance of the am b ient air quality by the commun ity is high.

(l.5<OU<5) Annoyance

Moderate to strong odours are detected sufficiently often by a proportion of the community to give rise to regular odour compla ints . Many mem bers of the commun ity will con sider the ambient air quality as unacceptable.

(5 <OU< 15)

Severe Annoyance

Strong odours ore regularly detected by a large proportion of the commun ity and give rise to many frequen t odour complaints.

M ost members of the community will be annoyed and agitated about the ambient air quality.

(> 15 OU)

• Odou r em ission rate; • Community response to od ou r; • Wind patterns; • T ype of odour model; • Surface roughness; and • Terrain.

Odour Emission Rate Odour emission rate is the fac tor which has the largest effect o n pred ictions. It is very d ifficult to measure odou r em iss ions fro m the various u nits o f a treatment p lan t accurately, p articu larly for fugitive em ission s or em issio ns during maintenance events when covers must be removed. From stud ies at Subiaco and Beenyup Wastewater T reatment Plants, it has been estab lished that the standard d eviation of the individual odo ur emissio n measurements is app roximately 70 per cen t o f the m ean, whether the od ou r measurem ents are mad e in sam ples collected using an odour flu x hood o r obtained from a ducr d rawing air fro m beneath covered ranks.

!fereed paper

Variations in od our em issio n rare also refl ect large differences between d iffe rent sampling techniq ues, the variation between laboratories and the natural variab ility from ho ur- co-hour and d ay-co-day. T he odou r measu remen t method in AS 4 323 .3 is based on the E uropean CEN m ethodology with the d ifference that in Australia the results from the first of three sample runs must be d isregard ed (this is op tional in the CEN method ology) .

I n WA, both fl ux hood s an d wind mnnels have been used for sampling; there is a substantial d ifference between the odour emiss io n rates measured using th e two m eth ods and the Corporatio n has now srandard ised on using a fl ux hood . To p rovide a reasonable but conservative m easure of odo ur emissions, it is the practice of the Co rporatio n to use the upper 75th percentile of the measu red emission rates as input in the model. It has

been foun d that fo r most odour sources, th e 7 5% emissio n rate is about 30% above the mean value derived from multip le tests of odour emission rate (CEE, 2006 ).

Community Response to Odour T he sensitivity of od ou r detection and annoyance varies from perso n to person .

Frequently one finds that a p erson who is very annoyed by od ou r has a spo u se or neigh bour who d oes no t no tice the odo ur.

T o provide a reasona bly accurate measure of communi ty response, it is ou r p ractice to use both odour co mplaint reco rds and telep ho ne surveys to establish th e extent o f odou r ann oyance and the extent of detectable od ou r.

Wind Measurements Peak odo ur concentrations ocrnr mostly at night du ri ng period s with light w ind s and low mixing rates. An anemo mete r in a

Odouridder® & Oda Vent®

Biofiltration Systems •

odour.s.... c01urolled naturally

Phone: (08) 9418 7594 Fax: (08) 9434 6819 E-mail : info@odatech.com .au (Odtlech 1s under l icense 101eement w11h ARC.ASA Pty L1d)

• • • • • • •

Biologica l treatment of hydrogen sulphide (H2S) and volatile organic compounds Innovative patented design Proven high H2S removal efficiency >99% Modular and transportable construction Low capital and life cycle costs Long media life (3-5years). No hazardous waste Environmentally sustainable technology Complies with WSA12 1 Industry Standard for Biofillers for Odour Control

Applications. Control and treatment of foul odours from sewage collection and transfer facilities, septage receival facilities, abattoirs, piggeries and meat rendering industries

Journal of the Australian Water Association

Water

M ARCH 2008 147

technical features )fereed paper

o_ d our manag valley, even a small valley, will record many more hours of su ch conditions in a year than an anemometer on a nearby h ill. To provide representative wind conditions, it is the Corporation's practice co use winds measured on che sire of the creacmenc plane.

Odour Model The odour model most commonly used in Australia is Ausplume. This model is distributed by the Victorian EPA and it is mandated as the regulatory model for predicting odours in chat Scace. The Ausplume model is based on a similar model issued by che US EPA, but with improvements.

An alternative model, Calpuff, is used occasionally co check the Ausplume results.

Ausplume is more conservative than Calpuff as it incorporates lower races of dispersion than Calpuff and the straight line of the plume simulated in Ausplume extends furth er from the plan t th an the irregular travel path of che puffs simulated in Calpuff.

Thus Ausplume generally predicts higher odour concentrations than Calpuff.

Nonetheless, if calibrated co community complaints and responses about odour, both models are satisfactory for modelling the movement and dispersion of odours and , in our experience, give similar predictions in terms of che area of buffer zone required.

Other Parameters Greater surface roughness (due co buildings and fences and even limestone pinnacles) increases dispersion. Hence lower odour concentrations will o ccur at sites where che air traverses a rough landscape (houses or forest) and higher odour concentrations will occur at sires where the air muse traverse a smooch surface (e.g. a lake) .

Topography has a complex influence on wind patterns, particularly at night, when cooler air flows down hills inco valleys and then travels out co che o cean (the night d rainage breezes) . However, most odour sources at a creacmenc plane are at or near ground level, and so topography has a smaller effect than th e ocher factors listed above.

Importance of Verification An odour model for an existing treatment plane muse predict rhe distribution of community concerns about odour, as indicated by odour complaints and responses co surveys about annoyance and detection of odours.

T he modelling package which has been used extensively in WA (Ausplume/on-sice winds/99.9% frequency/I-hour averaging) is reasonably successful at marching community responses. Examples of recent

148 MARCH 2008 Water

odour modelling are in Figures 1 and 2 for che treatment planes at Broome and Halls Head; in these figures, the predicted 5 O U contour is shown in red and the odour complaints are shown as blue docs. From comparisons between modelled contours and community responses at 10 treatment planes in WA, it has been established chat the predicted 5 OU (99.9% and 1-hour averaging) contour generally encompasses che area of community concern abou t odours.

Acceptance of Buffer Zones in WA Up co 1993, generic buffer zones were specified in planning guidelines based on che size of treatment planes and che nature of che treatment processes. The buffer distances sec by che guidelines ranged for 300 m for small planes co over 1 km for very large planes.

Since 1993, buffer zones have been accepted only on a case by case basis, with odour modelling following che procedures sec out in chis paper.

Conclusions Recognising the inherent variability associated with che response co odo urs, ic is co nsidered chat odour criteria should reflect a d istribution of odour concentrations as well as a range of different responses, rather than a single value char represents a sudden change from an acceptable to unacceptable odour situation.

An odour criterion of 5 OU (99.9% and 1-hour averaging) has been found co best represen t che boundary between acceptable conditions and annoyance about odour for urban, residential areas in Western Australia.

Table 1 presents a classification system using che bands defined in Figure 3 and the proposed odour criteria of 5 OU (99.9 % and 1-hour averaging) as the upper level of acceptable conditions for urban residential areas in Western Australia impacted by wastewater odours. This odour criterion has been shown to be appropriate for modelling with Ausplume and a comprehensive odour emissions inventory chat p rovides a conservative (but reasonable) upper bound estimate of odour emissions from the treatment plan c.

Odour criteria for many ocher jurisdictions in Australia and throughout the world may be positioned within che bands in Figure 3 as they reflect a similar differen tiation between acceptable odour and odour annoyance.

The Authors

Ian Wallis is Principal Environmen tal Engineer, Consulcing Environmental

Journal of the Australian Water Association

Engineers, specialising in air quality, environmental studies and strategic planning. Address is PO Box 201, Richmond V ictoria 3121, email wallis@cee.com.au; Keith Codee is General Manager Water Tech nologies Division, Water Corporation, Leederville, Western Australia 6007, email keich.cadee@wacerco rporation.com.au

References Amore, J .E. (1985). T he Perception of H ydrogen Sulfide Odor in Relacion co Setting an Ambient Standard. Prepared for California Air Resources Board.

AS/NZS 4323 .3: 200 1. Stationary Source Emissions - Determination of odour concenrrarion by dynamic olfaccomerry.

CEE (2006) Beenyup Wastewater Treatment Planr - Results of Odour Monitoring and Modelling Program. Report co Water Corporation.

Deffra (2006). Code of Practice on Odour Nuisance from Sewage Treatment Works.

Department for Environment Food and Rural Affairs, UK. www.defra.gov.uk.

Department of Environment ( 1993). Report by the Inspeccor on a Public Enquiry inro rhe Appeal by Northumbrian Water Limited for Additional Sewage Treat ment Facilities on Land Adjacent co Spiral Burn, Newbiggensby-rhe-Sea, Northumberland. DoE APP/F 2930/4/92/206240 .

Environment Agency (2002a) . Draft Horizonral Guidance for Odour: Part 1 - Regulation and Permitting.

Environment Agency (20026). Draft Horizontal Guidance for Odour: Part 2 - Assessment and Control. l nfoMil Netherlands Emission Guidelines for Air. (NeR), English Edition, www.infomil.nl.

NSW EPA (2001a). D raft Policy: Assessment and Management of Odour from Stationary Sources in NSW. NSW EPA (2001 6). Approved Methods and Guidance for the Modelling and Assessment of Air Pollurants in NSW. Queensland EPA (2004). Odour Impact Assessmenr from Developments.

Environmental Protection Agency, Brisbane, www.epa.qld.gov.au.

SA EPA (2003). EPA Guidelines: Air Qualiry Impact Assessment Using D esign Ground Level Pollutant Concentrations (DGLC's), Sout h Australian Environmental Protection Aurhoriry, EPA 386/03 .

UKW1R (2004). Odour Standards for rhe Wastewater Industry. Report Ref. No 04/WW/13/6. UK Water Industry Research Limited.

Valentin, F.H.H. and N orth, A.A., (eds) (1980) .

Odour Control - a concise guide. Warren Springs Laboracory. Department of Environment UK. VIC EPA (2005) . Scare Environment Protection Policy (Air Qualiry Management) , Viccorian Environmental Protection Authoriry.

WA EPA (2000). Guidance Statement on the Assessment ofOdour Impacts (Guidance Scarement N o 47).

technical features

odour management

Abstract A $55 million upgrade of the Merrimac Wastewater Treatment Plant has been completed by the Merrimac WWTP Waterfutu re Al liance to cater for the substantial growth p redicted for the Gold Coast populatio n. T h is upgrad e involved increasing the treatment capacity by around 60 percent to 57.5 ML/d.

Implicit in this upgrade was the effective management o f odou r emissions from the new facility. I t co mprises the covering of the main bioreactors and other selected eq uipment, the installatio n of ductwork for extraction of fou l air from these tanks, an expanded inlet works and biosolids processi ng facil ities and fi nally a new 52,0 00 m 3/ h r odour contro l faci lity. The selectio n of areas to be controlled for odou r was based on extensive emissions sampling of existing facil ities and associated dispersion modell ing. Modelling was used to gauge the cost effectiveness of the various odour control scenarios in eliminating nuisance odours to the surrounding community.

The odour control facility selected for such a large plant was uniq ue in that it represented an environmentally sustainable

approach to odour control. T his was achieved through the use of bacteria for the primary treatment stage rather than hazardous chemicals, as has been the case for most other large wastewater treatmen t plants around Australia. The facility consists of fou r bioscrubbers followed by four activated carbon fi lters and is currently the largest biological o do ur con trol facility for a wastewater treatment plant in Australia.

This paper discusses the in tegrated design app roach to the odou r management works and the construction , o peration and perfor mance testing of the facility, which has successfully achieved the project objectives on time and u nder budget fo r the benefit of Go ld Coast Water and the community. It also details the design philosophy of the project that featu red not only enviro nmental, bu t practical sustainability of this essential community asset.

Four bioscrubbe1¡s followed by /om¡ activated carbon filters represent minimum whole-of-life cost.

Introduction The Gold Coast Region with in So uth East Queensland has been experiencing almost constant growth of around 4% since the end of World War 2; growth that is expected to contin ue ro an expected maximum o f 1.2 million people b y 2056.

Availability of water supply is one of the key constraints on the ultimate population level. To address ch is fu ture growth, Gold Coast Water (GCW) has developed a master plan chat caters for the expected population growth co the 2056 planni ng horizon.

The Merrimac Wastewater Treatment Plant plays a vital role in cateri ng for che sewage treatmen t needs of approximately 25% of the Gold Coast population wit h a treatment capacity pre-2006 of 36 ML/d (120,00 0 EP). Therefore, upgrade of this plant was essential in addressing this projected growth and involved increasing the treatment capacity, through the const ruction of new wastewater treatment fac ilit ies, by around 60 per cent ro 57.5 ML/d (equivalent to 190,000 EP) rogether with associated odour cont rol works at a cost of $55 mi ll ion. Due co site limitations,

Journal of the Australian Water Association

water

MARCH 2008 14 9

technical features

the maximum capacity possible is approximately 80 ML/d ADWF, with expansion beyond this requiring either substantial technological change or the sourci ng of a new site .

A further $27 million upgrade is currently being undertaken for the plant involving the construction of a seco nd recycled water storage lagoon (45ML), a new wet weather recycled water pump station, installation of additional pumps and other minor refurbishment works.

History of Odour Nuisance The existing plant (pre-2006) had a history of odour nuisance, compounded in recent years by encroaching urban growth and odour sensitive areas such as the adjoining golf course (Lakelands) and two wedding reception areas in close proximity to rhe plant. Buffer zones had been reduced over many years, such that th e EPA was becoming increasingly involved in dealing with odour complaints in areas surrounding the plant. Consequently, a feature of the design brief for the plant upgrade was the objective to effectively manage sewage origi nated odours from the sire. A benchmark target was set at less than 2.5 odour units (as measured by dynamic olfactometry), at rhe adjacent Lakelands Golf Club. Ir was rhe first sensitive receptor beyond rhe plant boundary that other rhan passing people could be impacted.

During rhe planning of this project, it was recognised chat the plant was approaching its design throughput capacity, and that minimum treatment standards could potentially be compromised during wet weather fl ows. Therefore, rhe project was fast tracked using a competitive rendering system.

Alliance Formation and Structure GCW had always intended the works to be carried out by an Alliance of designers, contractors and specialist equipment supply companies and chose to deliver this outcome using a competitive 'Target Outturn Cost' (TOC) . Expressions of interest were called for Alliance partners to bid their capabilities for the chance of being selected for one of two final positions. Two consortia were selected and each was paid to provide a design and TOC. GCW retained copyright of all designs from this process, so chat beneficial ideas from the unsuccessful team could still be incorporated into the final design if warranted.

150

MARCH 2008

Water

Merrimac WWTP Odour Control Facility.

This process cu lminated in the selection of John Holland Water QHW) and Montgomery Watson Harza (MWH) to join with GCW as the Merrimac WaterFuture Alliance (MWFA) .

Subsequently Richard Flanagan & Company (RFC), RPC Technologies and Aromatrix Australia were engaged as suballiance partners through a similar assessment process.

In order to achieve effective management of the project, the Alliance was structured in the following manner: â&#x20AC;˘ Design review groups - which reviewed and developed designs to address integration issues and ensure project objectives were achieved; â&#x20AC;˘ Alliance Project Management Team (APMT), which handled the day to day operation of the Alliance and the project; â&#x20AC;˘ Alliance Leadership Group (ALG) which provided strategic direction and financial control and adjudicated on the rare escalation of issues from the APMT. All designs were rigorously reviewed at weekly technical review meetings attended by all Alliance members and in particular GCW operators and maintenance staff.

This process resulted in many design improvements with full compliance with GCW and regulator standards. It also enabled close cooperation between Alliance members and proved to be a very proactive means of resolving problems. It also led to many innovations an d design 'breakthroughs' not normally experienced in jobs of this size and nature.

Integrated Design Approach During the design planning stage, atmospheric dispersion modelling (using

Journal of the Australian Water Association

the Ausplume dispersion model), was undertaken to assess various odour mitigation scenarios. However, one of the first major hurdles to overcome was calibration of chis model to previous odour complaints. The closest meteorological station was Coolangatta Airport, approximately 15kms to rhe south of rhe site. While it was thought that chis data would give a fa ir and accurate approximation of local atmospheric conditions, the resultant model was not able to mimic rhe location of the majority of recorded odour complaints. It also indicated ocher populated areas where complaints should have been received but weren' t. The solution to resolving chis issue came from the plane operators who had been keeping an 'unofficial' daily plant log of weather conditions over many years. T he use of this data within the model produced almost instant correlation with surveyed complaints, and became the basis for all subsequent planning and design decisions, including the many "what if' scenarios.

Once the model was calibrated and sensitive receptors identified, internal odour sources within the plant were located and q uantified. An extensive emissions testing survey of odour sources was undertaken and used within the model to determine treatment priorities. This survey concentrated on the emissions of odour (AS4323.3) and specific odorous compounds such as hydrogen sulphide, mercaptans, dimethyl sulphide and volatile organic compounds. The principal odour sources in order of severity were the Preliminary Treatment Area (PTA) (incorporating inlet screens, grit removal, screenings and grit cleaning and compaction faci lity) , the anoxic and

technical teatures

odour management anaerobic zones of the bioreactors, the waste sludge belt filter press building and the dewarered sludge storage hopper.

Once this survey was completed, treatment options for each part of the site were considered, together with other sire-specific conditions. For example, the zincalume roof and bearers inside the belt filter press building were showing signs of premamre corrosion due to the warm moist co nditions developed during press operations as well as the accumulation of sewer gas within stagnant pockets within the building. The solution was to specify an extraction rate that would provide for effective air movement with in all areas of the bui lding thus preventing stagnant pockets from form ing and reducing co ndensation. Modelling showed that discharge of this ventilation air through a 15m high stack at sufficient velocity would result in gro und level concentrations below the benchmark target and therefore eliminated the need fo r fu rther treatment.

All other odour sources were covered and extracted to an Odour Control Facility (OCF) at a rare of 52,000 m3/ hr (with the requ irement for the OCF to allow upgrade to a maximum capacity of 72,000 m3/ hr) which was designed to achieve almost 100% capture efficiency, and which is equivalent to 15 air changes per hour based on the enclosed headspace vol ume.

The relatively large volume of air to be treated so mewhat narrowed the odour treatment choices available with regard to whole of li fe costs (WOLC) and footprint.

Biofilters (e.g. compost and so il bed filters) alone were ruled out due to the extensive area requi rements, although both biofil ters and activated carbon adsorption systems by themselves exhibited high WOLC (low capital but high media replacement costs) .

Chemical scrubbing, which is typically implemented fo r plants of this size, not only exhibited high WOLC, bur also had the possibility of added risk and adverse environmental impact fro m the use of hazardous chemicals. T he most cost effective and environmentally sustainable option proved to be che use of bioscrubbers (otherwise known as biorricl<ling filters refer article titles "Biorrickling Filters Cur the Cost of Odour Control" in March 2006 Water journal, for a derailed description of this technology). as the main treatment stage followed by activated carbon pol ishing. Although capital costs are higher than some of the other options, operation and maintenance costs are much lower (F igure 1) .

Ventilation Stack (16m height)

Duty I Duty I Standby

Extraction Fans

t

Gas Inlet

52,000 m3/hr

Figure 1. Process Flow Diagram.

Additional Design Parameters and Limitations

still allowing reaso nable sire access to service veh icles, cankers and B-doubles for sludge disposal. B-doubles, or double articulated sem i trailers as they are also known, are permitted on the eastern seaboard as a means of reducing transport costs and improving capacity for each truck movement. However, rhe main drawback is that they require wider passages through rhe plant, and more extensive turning circles, all of which are additional facto rs infl uencing the amount of process space available within the plant.

The Merrimac WWTP is a full y operational facility and since it was nea ring its maximum loading capacity at the commencement of this project, the design and consrrncrion of the plant upgrade was not permitted to impact on the capacity and treatment effectiveness in any way.

Planning for cur-ins for example had to be planned for low flow periods such rhar any flow backlog could be cleared before rhe next di urnal peak. This often mea nt crews working through the night to achieve process interruptions wi th 4.30am being the rime of lowest flows. Because of this, plant operato rs had to be kept informed of all developments and design decisions. This was simplified by appointi ng rhe plant supervisor to rhe APMT as well as the client's representative (Asset Creation Specialist) to advise as req ui red on standards and limitations as well as environmental, maintenance and specialist asset management issues.

Construction and Commissioning

From an environmental perspective, rhe plant borders an area of public open space char is a significant wetland precinct and forms part of rhe major Merrimac floodp lain. As a consequence, strict attention was given to acid soils management, as well as site restriction on any land fi ll or encroachment on rhe flood plain rhar could affect potential flood levels. This restricted activity to the rising portion of rhe sire had placed some significant limitations on where various process components could be located, while

A further upgrade (Phase 3) is currently being undertaken involving the construction of a second recycled water storage lagoon (45ML), a new wet weather recycled water pump station, installation of additional pumps and other minor refurbishment works.

Constructio n of rhe upgrade of rhe Merri mac Wastewater Treatment Plant occurred in rwo phases. Phase 1 works involved increasing the capacity of rhe Merrimac WWTP from 35ML/day to 57.SML/day through the modification to rhe existing inlet works and construction of new wastewater treatment bioreactors, rwo new clarifiers, a new biosolids handling facility, a new disinfection system and plant recycled water systems. Phase 2 works involve construction of the odour control facili ty (OCF).

Phases 1 and 2 have been completed on time and without any Lost Time Injuries (LTis) on rhe 370,000 plus hours worked.

Journal of the Australian Water Association

Water

MARCH 2008 151

technical features

Table 1. Performance Trials Results.

Parameter

Min

Ave

Max

No. of

Std.

Removal

(ppm)l

lppm)l

(ppm)l

Samples

Dev.

Efficiency (%1

4.1 3.0

5.8 5.2 0.21 0.60 0.27 64167

8.5 8.5 0.78 1.60 1.40 101000

42 42 42 42 42 42

1.27 1.53 0.14 0.67 0.33 16482

N/A N/A N/A N/A N/A N/A

0.0005 <0.010 3 <0.003 3 <0.0l3 <0.062 3 68

0.003 <0.010 3 <0.0033 <0.013 0. 150 163

42

0.001

42 42 42 42 42

0.000 0.000 0.000 0.030 42

99.992 >99.8 ]2, 3 >98.6 2,3 >98 .3 2,3 >77.02,3

Specification Requirement Removal at Peak Load (%)

Max discharge concentration (ppm) 1

99.75 99.75 99.9 99.5 N/A N/A

0.1 0.1 0.003 0.03 N/A 1000

Inlet to odour control facility

H2S - On-site analyser H2S -Tubes Mercaptans- Tubes OMS-Tubes VOC-PIO Odour (OU/m 3)

0.10 0.20 0.00 40650

Stack outlet

H2S - On-site analyser H2S -Tubes Mercaptans- Tubes OMS-Tubes VOC-PIO Odour (OU/m 3)

0.000 <0.0103 <0.003 3 <0.0l3 <0.05 3 17

99.892

Notes: 1 Concentrations in ppm unless specified 2 Removal efficiency based on bioscrubber inlet and stack outlet concentrations 3 Precise values not achieved due to detection limits with measurement techniques used The covering of an operating bioreactor and associated ducts presented a number of challenges to the team. A number of construction techniques were used including mounting temporary handrails on the covers during the installation period to prevent falls as well as fu ll time spotters throughout the job. Limiting the number of lifts and placement of covers was achieved by joining a number of covered section panels together on the ground and lifting these into position in a single lifting step.

The commissioning of the OCF and associated covers and ducts were undertaken in a phased approach to ensure chat there would be minimal fugitive odour release at any time. T he site utilised a temporary odour treatment system to treat odours from the PTA. It was important to minimise any downtime of odour treatment so Aromatrix's standard bacterial inoculation procedure for the bioscrubbers was essential to ensure chat when the changeover occurred, it would be able immediately to treat odorous air to an acceptable level.

The seeding of the bioscrubbers involved o btaining recirculation fluid from another bioscrubber facility installed by Aromatrix for Gold Coast Water. An acclimatisation period of six weeks was adopted to ensure che bacterial population was fully developed prior to performance testing. The carbon filters ensured chat the required odour discharge concentrations were compliant at the time of changeover whilst the 152 MARCH 2008

Water

bioscrubber was acclimatising to the odour load.

The changeover from the temporary odour treatment system to the new OCF involved removal of temporary ductwork and the connection of the new permanent ductwork to the PTA. An extensive ductwork system was designed to extract odorous air from the wastewater treatment bioreactors 4 and 5, and che sludge hopper, in addition to the PTA. However due to timing of the installation of covers and ducts, and the progressive odour loading of the OCF, initial flows were purely from the PTA. Once inscallacion of the covers and ductwork was complete and the bioscrubber performance stabilised, these were progressively brought on line.

Performance Testing Performance resting of the facility was a critical aspect of the project. An extensive sampling program was specified requiring duplicate gaseous sampling and analysis over each of21 days for hydrogen sulphide, mercapcans, volatile organic compounds (VOC) and odour dilution units. Samples were taken from the inlet of che OCF, between the bioscrubbers and che carbon fil cers and at the discharge to the ventilation stack. Less frequent sampling of dimethyl sulphide (DMS) was also undertaken at each of these locations.

Samples were taken at different times of the day throughout the 21 day period to ensure coverage of d ifferent diurnal cond itions.

Journal of the Australian Water Association

D raeger cubes were used for measurement of hydrogen sulphide (together with a continuous monitoring analyser), mercaptans and dimethyl sulphide. voes were measured by a MiniRAE 2000 portable VOC Photoionisacion Detector.

W ith regard to odour concentrations, gas samples were analysed on the day of sampling in accordance with AS4323.3.

The duration between sampling and analysis did not exceed 9 hours and were generally sampled within 4 hours.

From the results of the performance trial, the OCF achieved a hydrogen sulphide (H 2S) removal efficiency based on manual sampling methods and the continuous monitoring analyser of >99.8 1% and 99.99% respectively. Tha average stack outlet concentration was 0.0005 ppm as measured by the analyser.

The removal efficiency of mercapcans across the OCF was measured to be >98 .6% with concentrations discharged from the stack always <0. 003 ppm. For DMS, the removal efficiency across the OCF was >98.3% while all outlet concencracions from the stack measured at <0.01 ppm. For voes, the removal efficiency across the OCF was >77% with an average outlet concentration from che stack of <0.06 ppm.

Odour removal efficiency across the OCF was 99.89%. The inlet concentrations ranged between 40,650 OU/m3 to 101 ,000 OU/m3 while the stack outlet concentrations ranged between 17-1 63 OU/m 3 with an average outlet concentration of 68 OU/m3.

Based on the above results, the OCF has achieved the design performance requirements with I 00% compliance for H 2S and odour. For mercaptans and DMS, sampling methods were not sufficiently accurate to quantify removal efficiencies specified, although all measured outlet concentrations were less than design levels.

Key Features One of the benefits of the OCF is its compact modular des ign having been co nstructed to allow easy upgrade to a future maximum capacity of 72,000 m3/hr (additional area has been currently reserved for this upgrade) through che addition of bioscrubbers, carbon fil ters and fans. The other benefit relates to the ease of media replacement for any of the vessels. Activated carbon is very effective at adsorb ing contaminants and has been designed for a minimum media life of 12 months. By having four units operating in parallel for the bioscrubbers and carbon filters, media replacement can be carried out relatively easily by caking one unit off line at a time whilst keeping the overall facility operating at the required performance. Another design feature was the dual bed arrangement of the carbon filters which is effectively two filters in one resulting in a substantially reduced footprint compared to traditional single deep bed fil ters.

T he bioscrubbers were designed to operate on recycled water (Class B) in addition to potable water. Using recycled water brought rhe obvious benefits of conserving rhe co mmunity's valuable potable water supplies, however it also co ntained enough nutrients for the bacteria within the bioscrubber to function effectively with out the need for external nutrient dosing which is required when potable water is used. Although a dedicated nutrient dosing facility has been incorporated, it is not currently used as pare of normal operation.

A furth er consideration of the modular design, and one of the design features that helped guarantee a low WOLC for this plant, was the consideration of future maintenance and renewal activities within the OCF. Considerable discussion centred on optimisi ng design such char almost all of the major process units and ductwork can be removed from site and replaced without having to unbolt anything more than just the adjacent process component. All walkways were constructed of marine grade al uminium and featu red bolted sections char are easily demountable if a vessel beneath needs to be excracred for maintenance or renewal, although most have clear vertical access and can be removed through the use of cranes.

One of rhe major reasons for adopting marine grade aluminium for stairs and landings, apart from the obvious corrosion resistance, was the "man-handle-ability" of the removable sections. Attention to chis derail also means that expected rimes for major component replacement is considerably reduced, potentially saving thousands of dollars in environmental costs and fines. T he plant has also been set up such that any process unit can be isolated and/or bypassed while the plant is still on line ensuring compliance with rhe plant's operating license.

procedures implemented resulted in no lost time injuries over the 370,000 manhours worked.

Conclusions The OCF implemented as part of the upgraded Merrimac WWTP represents an environmentally sustainable approach to odour control. This has been achieved through the use of bacteria for the primary odour treatment stage rather than hazardous chemicals, as has been the case for most other large wastewater treatment plants around Australia.

Through the use of an integrated design approach by all members of che Alliance, and the implementation of a proven technology, the OCF has successfully achieved the project objectives under budget for che benefit of Gold Coast Water and the co mmunity.

The Authors Gary Finke is Managing Director of Aromatrix Australia Pry Led specialising in the design and supply of odour treatment systems. Email gary.finke@aromatrix.com.au.

Peter R Oliver was Mechanical T echn ical Specialist in Gold Coast Water's Asset C reation Department th roughou t this project. He has recently left Gold Coast Water after 18 years and has now joined Aromarrix Australia. Email perer.oliver@aro macrix.com.au.

Mark Thomas is a Commissioning Manager with John Holland Water Pry Ltd. Email MA.Thomas@jhg.com.au.

Dr Ian Evanson is Principal Process Engineer (Technical Leader, Odour Management) with MWH Australia Pry Ltd. Email Ian.Evanson@au.mwhglobal.com.

-0

a:

= Odour

C, C, C, C,

Control z Systems ;::: = Odour en Measurement i! and Analysis t-

C,

..I C, ..I

C,

t-

a:

= C,

>-

• Provider of innovative solutions to air quality and odour related problems within industry and the community.

• Our technology is based on over 20 years of research and experience around the world.

-+---'

_J

>-..

-+---' CL

cu cu .._ -+---' (/)

::::, <(

X ·;::: -+---'

cu

• Manufacturers E 0 of AroBIOS"" .._ biotrickling <( filters, AroFIL™ biofilters, AroCARB™ activated carbon filters, AroCHEM™ chemical scrubbers, and Odormat™ olfactometers.

• Services include : Odour control strategies, sewer modelling, atmospheric dispersion modelling.

Significant attention had to be given to the disposal of waste fluid (blowdown) fro m the bioscrubbers. Since this waste stream has a low pH, it couldn't just be discharged to the nearest waste stream within the plant as it could potentially interrupt the wastewater creatment process. To address this issue, a new pump station was built with a dedicated delivery pipe to the centre of rhe secondary anoxic stage of rhe bioreactor. This location presented rhe most cost effective discharge point for suitable flow mixing and dilution so as nor to adversely impact the treatment process.

Lastly, occupational health and safety (OH&S) was an important issue for all Alliance partners during the project. The OH&S Journal of the Australian Water Association

Water

MARCH 2008 153

technical features ifereed paper

ATTITUDES TO THE NATURAL RESOURCES MANAGEMENT LEVY IN ADELAIDE Z Wu, J McKay, E Hemphill Abstract The responses of the Adelaide urban communi ty to paying a levy fo r natural resources management (NRM) were investigated. The present NRM levy has been implemented since 2006 and is based on property values. The survey indicated chat the Adelaide urban community did not feel well informed about the NRM levy bur few complained about the amount as at present it is a very small fractio n.

Respondents were however more willing to accept a NRM levy when based o n the volume of water co nsumed rather rhan on p roperty value.

Introduction The emergence of the concept of sustainability has provided a new way of app roaching economic growth and natural resources management. H opwood et al.

(2005) believe all proponents of sustainable development suggest a need to change the present fashion of consumption, although there are major debates abo ut che actions necessary fo r rhe co ncept to be achieved.

The application of u ser charge systems to natural resources management requi res those who access natural resources to contribute to maintaining chose attrib utes or cover costs of degradation to encourage economically and environmentally sustainable use of the resources.

Fiscal stress on local governments has been motivating the implementation o f levying or charging systems. Responding to chis, policy-makers, public managers, and researchers seek to understand community attitudes to the public services they receive and to assess their willingness to pay for the services.

Local Natural Resource Management Policy The South Australian Natural Resources Management Act (NRM Act) 2004 came in to operation in 2005. U nder the Act, eight regional NRM boards were created across the State.

Before l July 2006, Councils collected catch ment levies in the former catchmen t

154 MARCH 2008 Water

water managemen r board areas and paid these to the relevant regional Catch ment Water Management boards. Likewise Councils conrributed from general rare revenue for animal and plant control work and made these payments to relevant regional Animal and Plant Control boards.

Since I July 2006 the catchmenr levies and rhe contriburions have been consolidated into a regional NRM levy which is presenred to rhe public as a separate line on their council rate notice, ro help cake care of local natural resources, such as water, soil , th e marine environment, native plants and animals.

Then; are two D ivisions of NRM levies.

Local councils, for the property-value based levy (D ivision 1), and the Department of Water Land and Biodiversity through Water Licensing for the water based levy (D ivision 2), collect the levy from property owners and water licencees before forwarding it to the relevant NRM Board.

The present study o nly concentrates on the Division l NRM levy.

Description of Study Site The survey site, urban Adelaide, locates within and hoses most of the populatio n of the Adelaide & Mount Lofty Ranges NRM Region. The Regio n covers a land area of 535,508 hectares and supports a population in excess of 1.2 million with only about 90,000 living in the rural areas (Adelaide & Mount Lofty Ranges NRM Board 2006a).

The NRM Act 2004 enables the Board to specify the amount to be collected by constituent councils from the commun ity towards the cost of performing its functio ns. The Board has consulted with con sti tuent cou ncils and recommends chat the levy be based on the value of land or a fixed charge of the same amount. For 2006/07, the average levy paid by Adelaide urban communities within the Regio n

"The problem with water is not water but society. " Lundqvist et al. (2001)

Journal of the Australian Water Association

ranges from $ 16 to $32 per household (Adelaide & Mount Lofty Ranges NRM Board 2006a).

The Questionnaire G iven th e advantages of fast speed, low cost, ease of entry of online data collection (Michaelidou and Dibb 20 06), the survey adopted a web-based questionnaire. The survey respondents were chosen randomly from an email list purch ased from a marketing company.

The questionnaire incl uded an introducrion which informed the respondents the reason for the study as well as an incentive of a draw for one of ten gift vouchers (each valued at $50) . The ema il invited recipients to click on che hypertext link, which presented the individual with the webbased questionnaire posted on the website of the University of South Australia. Once complered, responses were transferred ano nymously inro Microsoft Excel spreadsh eers.

The final survey co ntained 59 quesrions and rook 15 minutes to complete. T he survey was presented as a one-page document, th rough wh ich respondents were able to scroll up and down. This minimised the waiting time required to 'cum' to the next page and made the questionnaire convenient to handle, easy and quick to complete.

T he questionnaire was launched on a weekday morning with 4000 emails sent out. Overall, the response rate climbed to 19 .28 per cent, and nearly three-quarters arrived on the day of transmission which suggests chat respondents decided fa irly quickly whether or not they want to participate. Ten days later, 770 responses had b een received and no more entries came in. H owever, it should be noted chat the type of online questionnai re used (an email with URL embedded) and the nature of the particu lar sampling frame meant chat the response rate is n ot directly comparable with stud ies using d ifferent recrui tment methods (e.g. newsgroups posting, web page questionnaires) as well as different samples (Dibb et al. 2001) (e.g. students, academic

refereed paper

etc.). This is also in line with Schillewaerr et al. (1998) who found that different recruitment methods tend to attract different response rares and respondent profiles.

survey respondents (55.5%) indicated they were unsure.

Table 1. Information so urce of participants' a ware ness of the existence of NRM board.

Responses Number Percent

Information source

Demographic Profile of the Sample

90 65 56 42 22 13

Local government Newspaper

31.3 22.6 19.4 14.6 7.6 4.5

Apart f,-om a separate line located on council rate, continuing the work of increasing community awareness through a number of effective information and education approaches is needed.

Brochures from NRM board This study focused on Adelaide Community awareness of the NRM TV urban dwellers and sought to board Radio interview a sample of properry The community lacks confidence in owners over 18 years of age. The Internet processes in wh ich public authorities sample is restricted to property Total 288 100.0 undertake policies alone, hence the owners and hence it is difficu lt to need fo r effective interaction must be make a direct comparison between Results and Discussion recognised (Crawford 2002) . the sample demographic profile and the Adelaide urban population characteristics. A Participants' perception of the In our study, only a quarter of survey comparison wi th the 2006 Census of replacement of Water Catchment Levy by respondents stated that they knew they had Population figu res indicates that the sample NRM levy a local NRM board . The result was similar includes a large number of female co a previous finding in May 2006 As mentioned above, the current NRM levy respondents (sample 72.6%, census 5 1.2%), (Adelaide & Mount Lofty Ranges NRM is a replacement of previous Water a much higher median aged (25-54 years Board 20066). Sources of information are Catchment Management levy and An imal old) respondents (sa mple 79.4%, census listed in Table 1. and Plant control contribution . Such a 4 1.5%), a lower percentage of employed change brought a question to its levy payers: This find ing suggests that a more effective workers (sample 70.1 %, census 88.8%), Was the replacement a good decision or promotion work is needed co raise public and a similarly large number of respondents not? 30.5% of respondents thought it was profile about the NRM levy board. with median household income (sa mple good and 14.0% of th em didn 't think it Participants' perception of the function of 22.8%, census 19.1%). was a good decision but over half of the NRM board

Manage natural resources locally Manage water resources locally Don't know or unsure

!

The key fun ction of the NRM Board is "co develop plans that will assist in the better management and conservation of the region's precious natural resources today and fo r the longer term" (Adelaide & Mount Lofty Ranges NRM Board a).

1-----.- -.--...,....-,----,----r--~

1-----,- -.-- ...,....- ..1-----.--.---,.

C 0

.,.,

Q,

look after local environrrent

a: Keep community inforrred and educated on NRM Monitor waste Ensur e a sustainable use of natural resources

0

5

10

15

20

25

30

35

40

Percentage

Figure 1. What do you think the NRM board should do?

General government

I

SA Water

I

I

I

I

I

State government QI

Ill C 0

Don't know

C.

NRM Board

Ill

~

The results implied that the community were more concerned about water resources when talking ofnatural resources and they would Like to be involved in the NRM in their Local area.

I

Ill

I

I

'

I I

I

Everyone

I

I

I

W ho should be responsible for maintaining water quality in South Australia?

I

Federal government Local government

c::::J

I

0

5

10

When respondents were asked co describe their perceptions of what the NRM Board should do, 37% of them specified the Board's function as managing natural resources locally and 2 1% of them thought the NRM board should manage water resources locally (Figure 1). A large number of respondents (16%) stated that they did not know or were unsure about th is. This suggested that the comm uni ty did not feel they had received sufficient information about the board. 7% of respondents indicated the Board should be responsible for keeping community informed and educated on NRM.

15

20

Percent Figure 2. Who should be responsible for SA's water quality?

25

30

fn the past, the quality of the water in South Australia (SA) has frequently been questioned. When asked who should be responsible for maintaining water quality in SA, "Governments" were given priority,

Journal of the Australian Water Association

Water

MARC H

2008 155

technical features .fereed paper

d rawing the figures and finally, the authors thank the survey respondents fo r their rime and participation.

followed by SA Water. Only less than 6% of respondents assumed NRM Board should look after the water quality for SA (Figure 2).

Participants' perception of the way in which the NRM levy should be calculated

The Authors Zhifang Wu is a PhD student at

th e U n iversicy of South Australia.

Email: zhifang. wu@un isa.edu.au.

Few respondents complained Jennifer McKay is a professor of about the amount they paid for Business Law and Director of che the N RM levy but most of them Centre for Comparative Water cA flat rate • Area size of the land d isagreed on the current levy Policies and Laws, University of o Corrbination of property value and water used c D::in't know or unsure base method. To investigate the c How rreny people in a household • Household incorre South Australia. Email: perceptions about the methods II Other c A'operty value J ennifer.mckay@unisa.edu .au. used to im pose the levy, Cl Volurre of water used • Tied w Ith user type Elizabeth Hemphill is a lecturer respondents were asked to in Marketing and D irector of answer an open-ended q uestion: Figure 3. How do you think the NRM levies should be Postgrad uate Programs in 'H ow do you th ink th e Natural calculated?

Marketing, Un iversity of South Resources Management levies Australia.

Email: should be calculated?' Figure 3 Elizabech.hemphill@un isa.edu.au . likely to be considered as an excise. The sets our the results. The survey fu rther only solutio n of chis is to amend section 90 investigated various scenarios between References or alternatively, to permit the Scates to levy percentage of property value and percentage excise taxes as ap proved by a resolu tion of Adelaide & Mount Lofty Ranges Natural of water bill to calculate the levy. T hese Resources Management Board 2006a, Initial both Houses of Federal Parliament. resu lts also illustrate a strong preference of Natural Resources Management Plan for the community for a levy base on the volume The method used to impose the levy is driven Adelaide & Mount Lofty Ranges Natural of water consum ptio n, which also p rovides by the restriction in section 95 ofthe NRM Resources Management Region 2006-2008, incentive for responsible and sustainable Act and is in stark contrast to the levy-paying viewed 18 August, 2006, http ://www. water management. public who prefer a rate struck on the water amlrnrm.sa.gov.au/PlansStrategies/Regional 56.7

Curren tly the NRM levy (Divisio n 1) has been imposed o n rwo bases: the capital value of rateable p roperties or a fixed charge o f the same amount. H owever, the N RM Act (section 95) sets out a number o f alternative ways in wh ich the levies can be determined: (i) the valu e of rateable land; or (ii) a fixed charge of the same amount on all rateable land; or (iii) a fixed charge of an amount that depends on the purpose for which rateable land is used; or (iv) the area of rateable land; or (v) the purpose fo r which rateable land is used and the area of the land; or (vi) the location of rateable land .

The list above shows that the Act does not relate the NRM charge to the volume of water consumption. Changing the levy base method is not easy because this in part derives from the Federal Constitution and the restriction on Stace powers co levy taxation in section 90. The States cannot impose excise duties because of the first limb of section 90 which scares chat: "on the imposition of uniform duties of custom the power of che [Commonwealth) Parliament to impose duties of custom and excise, and to grant bounties on the production or export of goods, shall become exclusive". A levy on water use is

156

MARCH 2008

11.4

consumed. Perhaps this is a cogent argument for the reform ofsection 90 ofthe Constitution.

NRMPlan/AbouttheRegionalPlan.aspx.

Adelaide & Mount Lofty Ranges Natural Resources Management Board 20066, Public

Awareness ofBoard Existence and Functions,

Conclusions T he p resent study was designed to investigate public attitudes towards a natural resources management policy which is implemented in che Stace of Sou th Australia. First, there is very little negativity toward the NRM levy in Adelaide however there is a huge information void. The results of ch is study indicate chat community needs more in formation and education on natural resource or water resource management.

Home owner respondents clearly preferred che levy to be based on water consu mption not on property value. This would be illegal under section 90 of the Australian Co nstitution and the solution is to amend section 90 or to permit che States co levy excise taxes as approved by a resolution of both House of Federal Parliament.

Acknowledgments The authors express their thanks to the University of South Australia for supporting this work; the authors thank Ms Carol Hannaford from Adelaide & Mount Lofty Ranges NRM Board for providing information for chis work; the autho rs also thank Ganesh B. Keremane for helping

Water Journal of the Australian Water Association

April 2006 McGregor Tan Household Omnibus Survey, McG regor Tan Research.

Crawford, P 2002, 'Water Law and Policy: Back to Basics', in proceedings ofthe 4th Australian Water Law and Policy Conference, Sydney, 24-25 O ctober 2002.

Dibb, S, Rushmer, A and Stern, P 2001, 'New survey medium: Collecting marketing data with e-mail and the World Wide Web',

journal of Targeting, Measurement and Analysis for Marketing, Vol. 10, No. 1, pp.

17-25.

Hopwood, B, Mellor, M and O ' Brien, G 2005, 'Sustainable development: mapping d ifferent Approaches', Sustainable Development, vol.1 3, pp.38-52.

Lundqvist, J, Turton, A and Narain, S 200 I , 'Social, institutional and regulatory issues', in

Frontiers in Urban water Management Deadlock or Hope, eds. C Maksimovic & JA Tejada-Guibert, !WA Publishing, London, pp.343-398.

Michaelidou, N and Dibb, S 2006, 'Using email questionnaire for research: Good practice in tackling non-response', journal of Targeting, Measurement &Analysis for Marketing, Vol. 14, Issue 4, pp.289-296.

Schillewaerc, N, Langerak, F and Duhamel, T 1998, 'Non-probab ility sampling for WWW surveys: A comparison', journal ofthe Market Research Society, Vol. 40, No. 4, pp.

307-322.

STEP SCREENS DESIGN, INSTALLATION AND OPERATIONAL FACTORS Z Slavnic Abstract There appears to be a great deal of confusio n in the industry, particularly among the young engineers, about performance of the step screen s. If not designed, installed and/or operated p roperly, seep screens may fail to provide satisfactory screenings capture race. This paper attempts to address key potential failure mechanisms o f che seep screens which result in inadequate performance.

Introduction T h e author accended an AWA, NSW branch, conference at Coffs H arb our lase year. A number of questions were raised about the seep screens and their performance. The seep screens are relatively new generation of screens and provide much better service chan mechanically raked screens in terms of screenings collection efficiency, maintenan ce, etc.

H owever, there are a number of operatio nal facto rs chat affect performance of che seep screens, with which engineers may not be chat fami liar, particularly chose who have little experience in commissioning and/or operation. The paper will cry to address chis issue by outlining potential fai lure mechanism of the seep screens nor only from the operational, but also from the design and installation viewpoints.

Public Education Raw sewage, apart from dissolved and suspended pollutants, contains large fl oating materials such as rags, cotton buds, plastic bags, etc, and various organic matters (food scraps, leaves, pieces of bark, timber, etc, and of course faecal matter).

These materials, collectively termed screenings when collected by screens, if nor removed from raw sewage, not only interfere with the equipment used at treatment planes, bur also create odour problems. The author firmly believes char che first seep in the removal of many of these materials from sewage is public education. Namely, it is a regular p ractice at many households chat materials such as cotton buds, condoms, pads, wou nd

movable lamellas

stationary lamellas

~

a)

b)

c)

e)

d)

Figure 1. Principle of operation of step screen s. a) Stationary and movable lamellas are aligned in parked position. The captured object is at height A. b) Movable lamellas commence a step and move up. c) The object is lifted above the next step. d) The object is lowered on the stationary lamellas [height B) w hile movable lamellas con tinue their circular motion. e) Movable lamellas completed a revolutio n and stop in parked position, fully aligned wi th stationary lamellas, and ready for the next stepping cycle.

d ressings, gauze bandages, bags, etc, are th rown down the toilers. Many water authorities have advertising campaigns abou t water saving, water reuse, etc, but the author is unaware of a large scale campaign to educate public on what should go down the drain, or better, down the toiler. Since considerable resources are required by water au thorities to deal with the screenings material, they have vested interests to embark on such a campaign at a scale necessary to make significant changes in how we behave at home. Perhaps, we should go a step further, i.e. not only campaign about water pollutio n, water savings and reuse, but encompass ocher areas that also affect lives of us all, such as greenhouse gas emissions and global warming, deforestatio n, energy saving, renewable energy, ere, inco a comprehensive program to reach ou r children in primary schools, but ch ar is beyond the scope o f chis paper.

A pragmatic review ofa relatively new mechanism.

U ncil such a rime, the screenings will have to be removed from raw sewage at inlet works u si ng screening equipment such as seep screens.

Step Screens Operation A seep screen consists of two sets of stepped bars or lamellas made of stainless steel. The size of apertures typically used in sewage treatment ranges from 3-6 mm. Boch secs of the lamellas are linked, bur one set is movable, while the ocher is stationary, as it is fixed to the screen frame. The movable set is connected to a transmission system.

W ith each revolution of its d rive, the movable sec moves in an almost circular pattern, and with every revolution captured screenings is lifted to the seep above (refer Figure 1). Therefore, each revolution cransporcs the screenings seep by seep until the screenings reaches the top of the screen where ic is d ischarged through a chute directly into a bin o r a conveying system for further processing, e.g. washing, compaction, ecc.

Although step screens are commonly designed co handle peak wee weather flows,

Journal of the Australian Water Association

water

MARCH 2008 157

technical features

Figure 3. ' Screeni ngs Sausage'.

Figure 2. Step Screen in O peratio n: a) Side rubber seal s installed properly and in good working order. b) A decent screenings mat formed on the step screen.

overflow facilit ies sho uld be p rovided co d irect excess flows into a bypass channel.

This channel is typically equipped with a coarse mechanical screen which requires

manual removal of screeni ngs. Being a coarse screen (aper tures about 20 mm), it is much less efficient in cap turing screenings.

Design Step screens are installed in rectangular concrete channels. Proper operation and efficiency of step screens is a function of design of inlet channels, therefore d ue

C&SBRAND AUSTRALIAN FILTER COAL FOR DEEP BED COARSE DUAL MEDIA FILTRATION "More UFRVs for your money, and better quality water"

C&S BRAND GRANULAR & POWDERED ACTNATED CARBONS JAMES CUMMING & SONS PTY LTD

Email:

QUALITY ENDORSED COMPANY AS/NZS ISO 9001 STANDARDS AUSTRALIA

jamescumming@jamescumming.com.au

Licence no: 1628

319 Parramatta Rd AUBURN NSW 2144 Phone: (02) 9748 2309 Fax: (02) 9648 4887

158 MARCH 2008 Water

Journal of the Australian Water Association

co nsideration needs to be paid to the followi ng: â&#x20AC;˘ Sewage is most co mmonly pumped into the inlet works of a treatment plane. The zone in front of the step screen muse be of adequate size to dissipate the inlet energy of the pumped sewage and ensure smooth flows to the screen. Failure to observe chis would result in too great app roach velocities which may push screenings between the lamellas and, due to a high head loss, may un necessarily initiate a stepping cycle. â&#x20AC;˘ Apart from being non-turbulent, flows to the step screens need co be as much as possible perpendicular to the screen.

Otherwise, vorcexing of the flows may occur resulting in removal of che already captured screenings from che lamellas and promoting settlement of screen ings in front of the screen. T his in turn may initiate formation of 'screenings sausage,' i.e. screenings rolling up in front of the screen, as discussed below.

Installation As mentioned above, step screens are installed in rectangular channels made of concrete with a smooth fi nish. The

technical features

fo llowing needs to be considered during installation of rhe step screens: • The concrete chan nel walls must be vertical with smooth finish to ensure there are no gaps through which the screenings might escape when the screen is installed .

• The floor must be even and the screen needs to sit properly across the channel , so the screening cannot escape underneath the screen.

• Angle of inscallacio n muse be as per manu factu rers' recommendations. This angle is typically between 40° co 60°. Too great angles promote rolling down of screenings with the conseq uent fo rmatio n of the 'screenings sausage.'

• Side rubber skirts must be properly installed to ensure effective seal against the channel walls (refer Figure 2a).

Operation T he seep screens are comm only o perated in one of the three automatic control m odes, char is: • Upstream Level: A steppi ng cycle initiated when the level in front of the screen reaches a set poi nt.

• D iffe rential Level: A steppi ng cycle initiated when a difference between the upstream and downstream levels reaches a set poin t.

• Timer: A step ping cycle ini tiated based on ri me, typically measu red in m inutes.

screenings sausage is formed, it is impossible co build up the screen ings mat, as most of the screenings ends in the sausage, which in rim e grows in its diameter. Therefore, screenings sausages, when formed, greacly red uce screening efficiency, hence muse be removed as early as possible to enable fo rmation of the screenings mac.

And lase bur nor lease, for the seep screen to perform satisfactorily, the movable lam ellas need to park in the right positio n; char is co stop in the position when they are aligned with the station ary lamellas. It has been observed in practice char the movable lamellas may stop fu lly aligned with the stationary, bur then grad ually drop down wh ile waiting for the next stepping cycle.

T his in effect increases rhe gap between the stationary lamellas, hence larger floating materials, which would normally be captured, may slip through the step screen.

Conclusion There are a number of factors chat affect performance of the seep screens,

-s -

= --

encompassing d esign, construction, inscallarion and operational and maintenance facto rs. Therefore, nor all screenings would be captured by the seep screens, no matter what the size of aperture, if d ue attention is not paid to the above.

The escaped screenings would interfere with operation and performance of rhe downstream equipment such as pumps, centrifuges, etc. Ir is therefore recommended char macerators are installed to prevent operatio nal problems chat escaped screen ings may create. T h is wi ll make such materials more pu mpable, hence protect pu mps from chocking, and also prevent their seeding, for example in anaerobic digesters.

The Author Dr Zoran Slavnic (PhD , MBT, MEngSc, BMechEng) has over 25 years experience in design, construction , comm issio ning, O&M and asset management in the water ind ustry. He is Commissioning M anager with Laing O'Rourke (Australia) . Email: ZSlavnic@westcam den .chbmwacer.com.au

2=-- by-J·as

QUALITY ASSURED ENGINEERED PRODUCTS

When it comes to operation of the screens, it is important co keep in mind that the seep screen should be operated so as co ensure formation of the screen ings mat ( refer Figure 26) , as the mat aces as a fi lrer capturing materials char would normally pass between the lamellas, hence the screeni ngs mat increases the co llection effi ciency. T oo frequent initiat ion of rhe steppi ng cycle wo uld reduce build u p of the screenings and prevent the m at fo r mation , which in turn red uces efficiency of the seep screens.

The fo rmation of a screenings sausage (refer Figure 3), as already mentioned, may occur d u e co vorcexing of the in let flows or coo great angle of installation. A piece of tim ber, p lastic boccie or si m ilar o bject, b rought in w ith storm water, may be d ifficu lt co life. Such objects may keep fall ing back, eventually in itiating formation of the screeni ngs sausage. I n addition, accu mu latio n of far on the lamellas may also m ake scree nings falling back and result in format ion of the sausage. When a

Designer and manufacturer of hig h efficiency, low speed floating and fixed surface aerators from 3kW to 220 kW w ith on unmatched 5 year, unlimited hours guarantee.

By-Jos offers flexi ble fi nanci ng and delivery solutions including rental, purchase and fully maintained operating leases . Ring now for a current stock list. O ther products in our range include settling tonks (12 designs), packaged sewage and w ater treatment plants, reuse filters and clorifiers to Closs B and Closs A standard.

For more information, contact: By-Jas Engineering Pty Ltd PO BOX 424, HASTINGS VIC 3915 Tel: (03) 5979 1096 Fax: (03) 5979 1524

Journal of the Australian Water Association

Water

MARCH 2008 159

LUMINESCENT DISSOLVED OXYGEN SENSOR The LOO sensor uses a sensor cap chat is coated with a luminescent material. Blue light from an LED strikes the luminescent chemical on the sensor and the luminescent chemical instantly becomes excited. As the excited chemical relaxes, it releases red light. T he higher the oxygen concentration, the less red light given off by the sensor cap. The red light is detected by a photo diode and the time it cakes fo r the chemical to return to a relaxed state is measured.

The oxygen concentration is inversely proportional to the time it takes for the luminescent material on the sensor cap to return to a relaxed state. Between flashes from the blue LED, a red LED of known intensity is flashed. The red LED aces as an internal standard fo r reference comparison to che red light given off by the luminescent sensor cap. This comparison allows the sensor readings to remain stable for long periods of time.

Water Business aims to keep readers alert to business news and new product releases within the water sector. Media releases should be emailed to Brian Raulc ac brian.raulc@halledic.com.au or Tel (03) 8534 5014.

AWA wishes co advise readers char Water Business information is supplied by third parries and as such, AWA is not responsible for the accuracy, or otherwise, of the information submitted.

Wastewater Scheme in New Zealand. Earth Tech designed the complete wastewater scheme for Mangawhai Township, a popular beach resort located in a sensitive estuary and marine environment north of Auckland.

The company will also be responsible for the construction, commissioning and operation of the scheme on behalf of the Kaipara D istrict Council for up to 10 years.

The scope of operations includes collection, pumping, treatment and disposal of all wastewater and biosolids. T he EcoCare Scheme is scheduled for completion in July 2009.

• DO was the limiting factor in deployment length, so now site visits are reduced • No membranes to replace • Maintenance of the sensor is limited to wiping it clean • No special skills are required to change a sensor cap ... no stretching membranes, no worrying about air bubbles, no waiting for membranes to relax • No risk of membrane failu res such as tearing • No consumable anodes, membranes, or electrolyte to be maintained or replaced with every calibration

The major elements of the Mangawhai EcoCare Wastewater Scheme include 21 kilometres of sewers, 15 pumping stations, six kilometres of rising main, a small footp rint wastewater treatment plane, an 11 kilometre reclaimed water transfer pipeline and a reclaimed water storage dam and an irrigation system.

• Improved accuracy and stability gives users the best DO data Unlike polarographic dissolved oxygen sensor technologies, the Hach LOO sensor does not consume oxygen. It does not require frequent calibration or cleaning, resulting in longer sensor life and more stable and accurate readings. In addition, because the sensor is not consuming oxygen, a circulator is no longer required co maintain flow around the LO O sensor.

There are no electrodes or electrolyte to consume. There is no membrane to puncture or tear. Frequent calibration is not required as the instrument continually references itself co a red LED of known intensity. The Hach/Hydrolab LOO® sensor is approved by the EPA. The new Series 5 consists of the Hydrolab DS5X (DacaSonde 5X), DS5 (DataSonde 5), and MS5 (MiniSonde 5). H ach LOO is available only on H ydrolab Series 5 instruments.

• The sensor includes an internal standard chat provides a consistent reference after every measurement to maintain stable readings th roughout the deployment • Hach LOO is completely integrated into the Series 5 sondes. T herefore, the sensor can be brushed by the DS5X to remove any active fouli ng (e.g. barnacles) that consume oxygen and would otherwise misrepresent the condition of the water.

The key benefits of the Hach LOO sensor are as follows:

For more information contact Aqua/ab Scientific, website www.aqualab.com.au

• Calibrations last longer than ever before without drift, so deployment length is improved

NEW WASTEWATER PROJECT IN NEW ZEALAND

• No anode or electrolyte to consume means extremely stable measurements

Earth Tech has started construction of the NZ$53 million Mangawhai EcoCare

160 MARCH 2008 Water Journal of the Australian Water Association

Peter Everisc, Earth T ech's Group General Manager, explained that the project involved a two-year period of intensive planning and community consultation co o btain resource consents required under the New Zealand Resource Management Act. "The EcoCare Scheme plays a vital role in m aintai ning and sustaining the long-term environmental and economic health of Mangawhai," Everist said. "This is a very sensitive coastal environment and it was important for Earth Tech co work with the council and community to ensure the success of the project."

Beca Career H ollings and Ferner Led are the technical advisers to Kaipara District Council for the project. Commercial and transaction advice has been provided by specialise advisers EPS Consultants (Internacional) Pcy Ltd.