Water Journal March 2006

Page 1


FEATURES • CJ J ]~ lJ J J_E DJCJ l CJ 0 JSJ\l J~ i CJ J JTf rJ l ~J J J 0

• J\;ifJ~itl~J-~u J~ -1-2 S}~u J <Jl CJ 8 Y • CJD CJ Ul ~ h t;-\J l;-\ 0 -~_17} ]~fil T

Journal of the Australian Water Association

Constructing a dugwell in Muradnagor see page 60

bbe Fish Toximeter - see page 93

Water being collected from an early tubular RO membrane - see page 99

OPINION AND INDUSTRY NEWS OPINION Young Water Professionals taking off and taking over People bottleneck needs our attention Water policy in the 21st Century Atime for reflection

DDay, President, AWA CDavis, CEO, AWA The Hon. MTurnbull, MP Prof DBursill


s 6


AWA NEWS Includes special interest groups: Water Education Network, Young Water Professionals, International Water Association


CROSSCURRENT Industry News: National, State & Territory, International & Personalia


MEMBERSHIP NEWS Overview of AWA and membership benefits AWA new members

28 34



SPECIAL FEATURE: ENVIRO 2006 Conference and Exhibition


Other upcoming seminars and events

44 48

CONFERENCE REPORT: Young Water Professionals Ayoung water professional in the South Australian Water Corporation Her experience of their graduate program






PNadebaum, ABaker


CChow, RFabris, KWilkinson, FFitzgerald, MDrikas


indicates the paper has been refereed)

INTERNATIONAL ACTIVITIES ~ Rural Water Supplies in Bangladesh: An AusAID Project

The emphasis is on reliable supplies, not treatment WATER TREATMENT ~ Characterising NOM to Assess Treatability Two rapid analytical techniques for NOM OUR COVER

Membrane bioreactor technology is advancing rapidly. The first Australian plant has operated for three years at Magnetic Island, Old, for 0.3 Ml/d (page 106). Recently the Victor Harbor plant for 3.4 Ml/d has been successfully commissioned (page 110 ). Photo courtesy of Tenix P/L.

2 MARCH 2006


Volume 33 No 2 March 2006

Joe White Maltings MBR WWTP Bioreactor · see page 1 10

Force choice olfactometer with two sniffing ports · see page 14 l

Herrenknecht tunnel boring machine . se e pa ge 144

ON-LINE BIOLOGICAL MONITORING Quality of Water Resources: Biomonitoring Solutions AFrench research station has compared five types since 1999 Early Warning Monitoring Technologies Capital, set-up, operation and maintenance costs

AAndreoli, S Fass


NO'Connor, ADavison, DDeere, DBaker


EM VHoek


MEMBRANE TECHNOLOGY RO Technology: UCLA's Contributions to Research and Education Fifty years of development and still more to come Magnetic Island Water Reclamation Plant The results of three years operation summarised Australia's Two Largest MBR Water Recycling Plants Design and commissioning data for Victor Harbour and Forrestfield

DYoung, CHertle 106 MNewland 11 0

Memcor® Membrane Bioreactor Products - From Research to Commercialisation Effective membrane scouring is the key Why Small Towns are Choosing Membranes They are suited to remote locations and changing water conditions

FZha 11 5

A Patterson 12 2

ODOUR MANAGEMENT ·~ Odour Control for Sewers and Treatment Plants: Selection Criteria A critical reviewof available technology Bio-trickling Filters Cut the Cost of Odour Control Design and operation factors critically discussed Quantifying Odours from Food Industry Wastewaters Industry wastewater odours are different

J Watson, Rvan Oorschot 128 GFinke 133 RStuetz, FFrechen 14 1

SEWERAGE · Going, Gone, Trenchless: the Hallam Valley Main Sewer Extension Triple bottom line ;ustification for trenchless construction

CCorr 144

COMMUNITY CONSULTATION [ Acceptance of Water Recycling in Australia: National Baseline Data A degree of confidence in some personal uses

J SMarks, BMartin, MZadoroznyj 1 5 1






MARCH 2006 3

YOUNG WATER PROFESSIONALS TAKING OFF AND TAKING OVER As che Baby Boomer Generation passes into retirement, some o rganisations are challenged with a loss of 50% of their workforce over the next five years - o ne in two of rhe most experienced people; fu ll of corpo rate, and in some cases highly specialised, knowledge. In m y column last month I raised the challenge of d eveloping the water sector's skilled resources for a sustainable future which, over the next ten years, will see rhe most rapid change in the history of urban water supply and sanitation. W e are challenged with developing sustainable cities, managing and restoring stressed surface and ground water systems, responding co increased climate variability and climate change, d evelopi ng new technology co support integrated water cycle solutions and end use efficiency; whilst ensuring public healch is protected. We need to look co our Young Water Professionals; retaining involvement of rhe Baby Boo mer Generation, and creating exciting and rewarding career options for che Y Generation.

n etworking activity), with many attending their first conference; o r d elivering their first presentation. More than half rhe participants were women; which bodes well for our future. The 2nd YWP Conference will be at the University of Queensland in February 200 8. Once word of the 1st YWP spread s, the dare will be put into many diaries. On a broader stage, a range of Young Professionals groups has come together co assemble a YP scream in the Enviro 06 Conference in Melbourne May 9 . The key theme is professionally challenging topic of siring a nuclear waste repository. There will b e eminent water people (including AWWA President Andy Richardson) speaking coo, so it won 't all revolve around nuclear waste. (see page 39 for more about Enviro) came together like clockwork. Emeritus Professor Nancy Millis, who is very young at h eart, set the scen e for the conference with an insightful key note address, leaving the YWPs w ith gems as rake-home messages.

I have been gobsmacked by the achievements of the Young Water Professionals' Network sec up in May 2005. In less than 12 months, with Chris Corr as chair, the N etwork has gone from strength co strength, in coordinating activities and services across the branch es, and driving change in A WA co better service the needs and interest of younger professionals across all sectors of the water industry. We are also fortunate co h ave Associate Professor Richard Sruerz, C hair of the International Water Association's Young Water Professionals Committee, and member of the IWA Australia Committee, back in Australia, based at che University of New South Wales (UNSW).

Later in this edition there are reporcs on the program which included a machine gun blast of 5-minure oral presentations on the poster papers, which was a successful d eparture from normal practice. Professor Jurg Keller conducted the high energy session , with 13 p resentations in 65 minutes! (see page 48 for YWP Conference Report)

In under 8 months rhe first ever Young Water Profess ionals' Conference was presented by IWA Australia/ AWA, over three days (1 5 - 17 February) at the University of New So uth Wales in Sydney. There are five people who must rake the lion's share of accolades for the conference: co-chairs Dr M ichael Storey (CSIRO Urban Water), and Dr Pierre Le-Clech (UNSW), Dr Shoshana Fogelman (Griffith U niversity) - see report o n page 14, Dr Sandra H all (University of Queensland) and C hris Corr (GHD ). Supported by the other six members of rhe Conference Committee, they made a formidable team; the attention ro derail was excellent and the whole event


4 MARCH 2006


The half-day fo cus on mentoring and networking session included many invited leaders of rhe water sector co assist by telling their stories and participating mentoring activities. From rhe minute the event began it was abuzz with excitement (especially rhe exhilarating - and ear-shattering - speed

FUTURE FEATURES MAY · Demand management - smart meters , innovation in stormwater, benchmarking

JUNE · En viro 2006 report, water education r epor t and papers , desa lination projects

AUGUST · Distribution i ssues, qual ity and integ rity; pumping & pipel i nes, agricu ltural uses

On a broader stage still, the IWA (International Water Association} Young R esearchers' Conference is slated for May in Singapore - where an array of impressive, invited young researchers will meet, deliver papers and interact. Ar least half a dozen Australian presenters will b e given some financial support co get across co Singapore and rake part, so I expect Australia will be well in evidence over there. (See IWA page 46 for more derails) . All these activities bode well for the future of the water industry - clearly there is a cohort of very motivated and interested people coming through the ranks. While the issue of finding enough people remains a serious challenge (see Chris Davis's column in this issue) the quality and energy of chis group is heartening. It took AWA some rime co react co the needs of this group of younger members; bur I am confident that, now we've done so, it paves the way for not only more specialised YWP activities, but also for a generally more inviting climate in A WA in general. M entoring, strongly supported by the Baby Boomer Generation of our industry is flouri shing; since ad vancing in an industry with such long traditions can be a daunting prospect fo r younger practitioners generally, and women in particular. I realise that setting up mentoring relationships is easier said than done; it relies on chemistry as well as an enabling environment. We have, I hope, created an enabling environ ment and I hope now the chemistry rakes over and creates some bonds that will help many young people enhance their careers.

Darryl Day

Providing Solutions to Odour and Corrosion Problems Does your sewer network or treatment plant suffer from hydrogen sulphide odour and related asset corrosion problems? Do you experience low wastewater pH within the sewer network or treatment plant? Are you experiencing any compliance/performance/cost issues? Do you have a requirement to reuse treated wastewater for irrigation? Orica Watercare offer a comprehensive range of proven technologies for odour and corrosion control. Technologies avai lable include both liquid and gas phase treatment and a novel technology t o prevent corrosion of munici pa l infrastructure. With our range of technologies, applications expertise and engineering support, Orica Watercare can tailor and implement an effective solution to complex odour and related corrosion problems

corrosion control as well as contributing valuable alkalinity. Sulfalockn" is particularly well suited to applications where : • sewer retention times are high • sewage pH is low • dissolved sulphide levels are hi gh • sewer corrosion and odour is severe • a non-dangerous liquid is required • a non sodium alka li is required

Odourlock™ An iron sa lt used to reduce odour an d corrosion in sewage networks and t reatment plant s. Added directly to sewage, Odourlock"" reacts with dissolved odour causing su lphides to form a insoluble precipitate. Odou rlock" " is particu larly well su ited to applicati ons w here: • sewage net works are large and have long ret ention times • sewage pH is neutral or higher • sewer corrosion and odour is severe • t here is a requirement to remove su lphides from solution Odourlockr" and Sulfal ock™ are t echnologies that can be used alone or together to maximise performance.

Corrosion Control Sulfalock HiGel™ HiGel"' is a specially formu lated product designed for spray application onto concrete assets su bject to acid corrosion.

So if you have answered "yes" to any of the above questions, contact Orica Waterca re to arrange for a review of t he issues and a costed treatment proposal. South Australia Western Australia Queensland New South Wales Victoria Tasmania

8150 5914 9314 4653 3867 9519 9352 2107 9283 6283 6331 9799

Gas Phase Odour Control Odorgarde Odorgard is a catalytically enha nced hypochlorite scrubbing/packed tower gas phase treatment technology. The catalytic process offers many benefits over conventional scrubbers including: • el imination of chlorination side reactions/chlorine gas emission • reduced space, capital and chemical requi rements

Liquid Phase Odour Control Sulfalock™ A stable suspension of magnesium hydroxide. Added directly to sewage, Sulfalockrr" increases sewage pH and the solubility of sulphides gases preventing their release. Sulfa l ock™ can provide the dual benefit of a odour and

Sprayed ont o manho les, pump stations and sewer lines, Sulfalock HiGel™ forms a durable alkali ne coating that neutralises acidic surfaces and reta rds further acid production. HiGel™ can eliminate acid corrosion greatly increasi ng asset life. HiG el™ offers many benefits over conventional corrosion control technologies: • extremely effective corrosion protection • straightforwa rd application with minimal surface preparation • applied to live sewers - no diversion required • cost effective

Chemical Storage & Dosing CDU™ The CDU™ gives a secure, totally self conta ined and fully automated bulk chemical installation t hat complies with all legislative requirements for bulk chemical storage. Avai lable in 2000, 5000 and 9000 L capacities, the Orica CDU™ can be supplied as part of a comprehensive service agreement in wh ich Orica Watercare provides chemicals, storage and dosing equipment and scheduled maintenance.

RURAL WATER SUPPLIES IN BANGLADESH: AN AUSAID PROJECT P Nadebaum, A Baker Abstract Australia, through AusAID, has funded the setting up of a Bangladesh-Australian Centre for Arsenic M itigation in D haka, and programs fo r safe water, safe food, and health. This p roject was led in Bangladesh by the D haka Community H ospital. The main focus of the program was to avoid the use of shallow groundwater contaminated with arsenic and to p rovide alternative safe water su pplies to villagers. The p roject included the ap plication of new methods of managing water supplies wh ich are being developed in Australia and by the World Health Organisation, and in p articular the use of the principles of risk management and H azard Analysis and Critical Con trol Poin ts. The project commenced in March 2002, and was largely completed and in the fi nal stages o f reporting at the end of 2005.

COMPONEN T I Bangladesh -Australia Centre for Arsenic Mitigation

Development of overall risk-based management system for water supply

COMPONENT4 Community Health


COMPONENT2 Snfe \Voter


Review of existing options for water s upply

Trials of options

Field sampling and analysis program

Bioavailability test work and exposure


Field survey and development of education program

Development of Rural Water Supply Guidance Manual

Introduction It is estimated that more than 30 million people of Bangladesh have water supplies that have arsenic concentrations more than 50 ug/L, and considerably exceed the Australian guideline of7 ug/L. The arsenic results from naturally occurring arsenic in the de!taic sediments, which have been intersected by shallow cube wells installed for drinking water supply purposes. Approximately 10 million shallow tube wells have been installed in Bangladesh over the years, and approximately 20% of these wel ls have concentrations of arsenic over 50 ug/L. Many villagers in Bangladesh are being affected by arsenicosis, with kerotosis and other effects such as cancer becoming more prevalent. Australia, through AusAID, has fu nded three projects co assist Bangladesh with chis problem. This paper provides information on one of these projects, which has involved setting up a Bangladesh - Australian Centre fo r Arsenic Mitigation in Dhaka and programs for safe water, safe food, and health.

60 MARCH 2006 water

Figure 1. Overa ll Framework fo r Pro ject.

The Organisations and Personnel Involved

Garman, and Phillip Crisp of the University of New South Wales.

The p roject has involved a number of organ isations and many individuals. These include: • T he Dhaka Community Hospira! (DCH), an organ isation that has had a key role in identifying the issue and championing assistance to villagers throughou t Bangladesh. DCH's involvement was led by Director Q uamruzzaman.

• T he U n iversity of Sou th Australia, responsible for the Safe Food Program, particularly Professor Ravi Naidu and Gary Owens.

The focus for arsenic mitigation shifts to safe alternative supplies. • GHD Pry Ltd, the Australian managing contractor, particularly Dr Alison Baker, Dr Peter Nadebaum, and Simon Deeble. • The CRC for Waste Management and Pollution Control, responsible fo r the Safe Water Program, particularly Dr David

Overall Project Objectives T he goal of the project was to provide a substantial contributio n to the prevention of arsenic poison ing in the communities of Bangladesh th rough an integrated and sustainable set of programs to p revent exposure to arsenic from ground water and food. T he project is consistent with the policy of the Government of Bangladesh to focus o n alternative safe water sup plies, cacher than treatment o f contaminated water. The project supports five key areas that AusAID has targeted for Australian assistance to Bangladesh in dealing with the arsen ic problem. T hese are: • Provision of safe clean water; • Relevant app lied arsenic research and technology development that is cost

refereed paper

The No. 1 in Odour Control "Every Great Accomplishment Is At First Impossible"

Please visit us at booth# F298 at Enviro '06 - May 9 -11 - Melbourne Exhibition & Convention Centre

Bioway Technologies Australia Pty. Ltd. Sydney, NSW, 2000 Phone: +65 6224 5852 Fax: +65 6224 7822 Email: info@bioway.com www.bioway.com


BIOWft.Â¥ '"

Working with the community and local tradesmen to build a Pond Sand Filter. effective, ecologically sustainable and acceptable by the Bangladesh communities; • Health/epidemiology; • Com m un ity activities for safe drinking water; and • C apacity building. The p roject commenced in March 2002, and was largely complete and in the fina l stages of reporting in late 2005 . T he various componen ts of the project are shown in Figure 1.

Bangladesh - Australia Centre For Arsenic Mitigation A key aspect of the project was to set up a Bangladesh - Australian Centre fo r Arsenic Mitigation in Dhaka. The Centre was establ ished within the D haka Community Hospital (DCH), with the objective that the Centre and its programs would co ntinue beyond the life of the current fu ndi ng from AusAID. Th e Centre had an important fu nction of being a centre of excellence in the field of commun ity water supply, providing information and advice on safe water, food, and health. This information is provided at various levels, ranging from simple information suitable for villagers and nongovernmental organisations, to technical and scientific in fo rmation for government agencies such as the D epartment of P ublic Health Engineering in Dhaka, and other technical persons working in the area. DCH is rep resented on the Bangladesh Government Arsenic Technical Committee, and this assisted in ensuring that the project findi ngs are d isseminated to approp riate agencies in Bangladesh. T he information on arsenic mitigation is being p rovided to persons not only in


MARCH 2006


Bangladesh, but also in other countries where arsenic is a problem, including Vietnam, Nepal, India, Pakistan, Cambodia, C h ina and Taiwan. This dissemi nation of information is being achieved through a variety of existing networks, includ ing linkages with th e W orld H ealth O rganisation and other arsenic research centres throughout the world. In addition to the provision of advice, the Centre also has laboratory faci lities that can assist in u ndertaking field programs and in the analysis of various types of samples (water, food, biological and soil) for arsenic, and in the preparation of samples for advanced analyses (such as arsenic speciation).

Safe Water Program The use of grou ndwater for drinking and cooki ng in villages is the most significant cause of excessive exposure to arsen ic, and immediate action was required to reduce exposure by avoiding use of con tam inated groundwater or by treatment of the groundwater. Options such as rainwater tanks, sh allow dug wells, pond and river water fil ters and deep wells can be free of arsenic, although for some of these options treatment is needed to ensure that the water will be microb iologically safe. T ube well water containing arsenic can be created to remove the arsenic, and there are many methods available for this. I nforma tion on these options is being included in the safe water program . However, assuring that these treatment methods will work reliably in the village environment is a challenging requiremen t, and the Government of Bangladesh has a policy that supports the

refereed paper

lMENT ery drop is precious e treating it right ............... ...... 0~0 0


·-§ :


Rejly on:

Wallace & Tier

Water 1reat•e


Chlorinators, Metering Pumps, Dry Fee'8 Disinfection, Analysers, Controllers, Gas Dete Dioxide Generators, On-Site Sodium Hypo ounted Water Treatment Systems and Odour

Visit us on the Siemens booth no: F312


international provision of alternative water supplies that are free of contamination, and requires a high level of assurance if point-of-use groundwater treatment systems are ro be offered ro villagers. The program included mapping by Geographical Positioning Systems (GPS) to determine the location of cube wells and sampli ng of these cube wells. Villagers with contaminated cube well supplies were consulted to determine the most appropriate and acceptable water supply option These incl uded rainwater ranks, shallow dug wells (which were known to be generally free of arsenic), and pond and river sand filters. In addition, a new largethroughput iron-based treatment system was proposed for creating relatively large volumes of irrigation water. A number of the options being trialled were not new to the com munities (especially the dug well), and the objective was to draw on rhe experience and knowledge of the communities in these systems and where possible improve on them. Alternative systems that provided fo r reticulation on a local basis were attractive; however, they introduced rhe potential for contamination ingress through leakage and carefu l design and management was required for such systems. Ensuring that any system was pro perly maintained and operated in the long term is essential, and a major pare of the program has been the esrablishmen r of a management framewo rk and an appropriate body of information and training methods. Ir was particularly recognised char rube wells, while incurring risks associated with arsenic, have avoided many of the problems associated with microbiological contamination. Ocher options have a greater potential for microbiological contamination, and ir was a very important aspect of the program to ensure char whatever option was adopted, char ir could be maintained free of contamination.

Safe Food Program Approximately 80% of rhe water used for irrigation of crops in Bangladesh is from shallow rube wells, and rhe irrigation water can contain arsenic. Certain crops can selectively rake up arsenic present in soil and irrigation water, possibly exceeding permissible levels fo r food. The use of arsenic-contaminated water for general use in the villages can also give rise to elevated arsenic in surface soils and dust within houses. The safe food program had the objective of measuring these concentrations, to assist in understanding the significance of exposure to arsenic via water, food, soil and dust, and to identify the strategies chat

64 MARCH 2006


The community involved in constructing a dugwell in Muradnagor.

should be adopted at the village level to red uce the risk associated with chis exposure. The safe food program involved raking more than 4000 samples of water, plant, food , soil and dust from 30 villages in Bangladesh, and determining from these samples the concentrations of arsenic rhar occurred in the various village situations. Training in laboratory methods was provided to OCH staff, including training in quality managemenr systems. Samples were prepared for analysis in Bangladesh, and the samples were then shipped to Australia and analysed in Australia. Established methods have been used for rhe analysis of arsenic in food and soil, including speciacion to determine the form of arsenic. The bioavailability of arsenic was an important factor to rake into account when assessing the significance of arsenic in food. To assess chis, feeding studies were carried out in Adelaide using approximately 50 young swine. Plants commonly cultivated in Bangladesh (silverbeec, radish, amaranrhus) were grown hydroponically in Australia in a solution artificially spiked with arsenic. After growth planes were assessed for total arsenic which was found to accumulate in the order silverbeec >> radish Âť amaranchus with the distribution being mainly in the planr roots (roots Âť stem > leaf). l e was found that the arsenic in the edible portions of the plants could exceed the Australian food guidelines of 1 mg As kg- 1 fresh wt. Rice, the main food staple in Bangladesh, was also cooked in artificially spiked water and subsequently assessed for total arsenic. This study indicated char significant

quanriries could be sorbed into the rice via cooking with contami nated water. The resulting rice and plant material were fed to young swine and the arsenic content monitored in faeces, urine and in blood. Knowing the total dose administered in the foodstuff allowed rhe bioaccessibiliry to be estimated. Methods were developed for rhe analysis of blood, urine, faeces and animal tissue, using microwave-assisted digestion. Test work has shown that the methods are robust and achieved good recovery. The resu lts of this work showed char bioaccessibiliry was dependent on the plane species. The average bioaccessibility was determi ned to be 28% for silverbeer, and 85% fo r rice.

Community Health Program The objective of the com munity health program was to develop and implement a Community Health Education System with selected local communities. A particular focus was to educate the community on arsenic poisoning, so char they could identify whether arsenic was a problem and determi ne how they could minimise this impact. The program provides information on the health issues and included discussion and evaluation of these issues in an overall lifestyle context. As part of the program, surveys were carried out of villagers in the various villages in which the water and food studies were carried our, particularly the role of women in obtaining and using water, and how they may best be part of the solution. Information was also obtained on hygiene and rhe practical use of water within the village, so char other sources of contamination (e.g. microbiological) can be avoided as far as possible.

refereed paper

international Fortunately, DCH had established a Rural Health Scheme in Bangladesh char could be extended co include a Safe Water component, and chis offered a solution co these difficult requirements.

Requirements for Developing a Sustainable Water Supply An essential aspect of a solution to the arsenic problem in Bangladesh was co ensure chat any solution would be accepted and owned by the villagers, and chat they would maintain it so chat a safe water supply is provided for in the long term. T his is not a simple matter, and there are many examples of where a water supply has been provided co a community, but the community has not taken ownership of the system and ir has failed. In Bangladesh there are many millions of villagers involved, and the government systems and infrasrrucmre are not well established . For most smaller villages it is not possible co rely on government organisations co provide for and manage water supplies, and it was necessary co adopt an approach char the community could own and be committed to maintaining rhe water supply. Notwithstanding chis, any solution muse have general applicability and be able to be incorporated into any local government or central government organised approach co the arsenic problem, as may exist in a particular area.

Extending the DCH Rural Health Scheme to Include Safe Water

A dugwell with a reticulation system being constructed in Nabinagor.

Another very important issue for the project was the extent of Australian funds available. T hese were limited and, for the proj ect to achieve a significant contribution, it was essential char any approach be such char villagers would be able co afford the solution, and would want to implement ir.

CGS Short Courses & Conferences Australia - New Zealand - Malaysia 2006

Centre ror Oroundwater Studies

. 011 B roehures at: ..www..eround water.com.au/.sh art-courses. htm see Course Reg1strat1 Tuesday 18- Friday 21 April 2006

l" Malaysia NAPLs and Groundwater

MALAYSIA Wed IO - Fri 12 May 2006 Rockhampton QLD

Kuala Lumpur

14t" Getting To Know Groundwater and Surfacewater 3 I" Australian Groundwater School

Mon IO - Sat 15 July 2006 University of WA WA

5th NAPLs and Groundwater

Tue 18 - Fri 21 July 2006 UNSW Sydney NSW

2nd Groundwater Model Calibration 32n" Australian Groundwater School

Wed 30 Aug - Fri I Sept 2006 Brisbane


Mon 25 - Sat 30 Sept 2006 Brisbane


5 Australian Groundwater Modelling School:

Tues 14 - Fri 17 Nov 2006 Perth



Concevts; Avvlication Visual MODFLOW;PEST

1st New Zealand Getting To Know Groundwater and Surfacewater 6'" Australian Groundwater Modelling School:

Wed 15 - Fri 17 Nov 2006 University of Canterbury Christchurch Tues 21 - Fri 24 Nov 2006 Sydney


Concepts; Avvlication Visual MODFLOW;PEST

I" New Zealand Groundwater Modelling School: Concevts; Avolication Visual MODFLOW;PEST

l" Irrigated Soils Management 2'0 Australasian Hydrogeology Research Conference

9th Australasian Environmental Isotope Conference WORLDS FRESH WATER (fluid only)

Tues 28 Nov - Fri I Dec 2006 Universitv of Auckland NZ To be advised Wed 13 - Fri 15 Dec 2006 (Concurrent C onfs) Glenelg, Adelaide SA

CONTACT DETAILS: Trevor Pillar, Business Manager, Centre for Groundwater Studies.


Phone:61 8 8201 5632 Fax: 61 8 8201 5635 Email: cgs@groundwater.com.au Web: www.groundwater.com.au SURFACEWATER

66 MARCH 2006


Endorsed by: National Groundwater Committee






T he DCH Rural Health Scheme is based on raking a communal approach co health care, and draws on the established village support structures in which all in a village will contribute to the extent possible co send one of the village children co university, or for a wedding, or for building a village building. Thus there is an established basis for the community co contribute co and support the Safe Water Program. The key aspects of the Safe Water Program included the following: • The village establishes a water management committee, an elected group who collectively co-ordinates the supply of safe drinking water co those households who have agreed, through consultation, to contribute money co a commun ity-based water supply scheme. A person from DCH or from a nearby village with che scheme already in place meets with the village and explain the scheme. It can cake many village meetings before a committee is form ed. A similar approach is taken in the case of the Rural Health Scheme, with a Family H ealth Committee being form ed. • T he Committee generally includes: a chairperson, a maintenance coordinator, a monitoring coordinator, an education coordinator, a membership coordinator, a person responsible fo r the collection of monthly contributions from member households, and other persons as appropriate (e.g. sponsors). The Rural H ealth Scheme Committee was similar but it appoints and coordinates Family Health Workers from the village, organises the collection of monies and education, and also manages the interface between DCH and the village (e.g. calling on a doctor or specialise such as a dentist from DCH to visit the village and provide services). • The responsibilities of the Committee include distributing water supply cards to member households, collecting an agreed monthly amount of money from member households, employing a community-based "water worker" who was responsible for the day-to-day operation and maintenance of the water supply system, arranging for water sampli ng and analysis co be undertaken, communicating water quality results co member households, directing corrective actions in the event of water quality results

refereed paper

When it's your good name on the line ... It's great to have two good names behind you! When push comes to shove, it all comes down to data. Data that is accurate, complete and credible. Fail on any count and it's goodbye to your good name - or worse still ~our career! The good news is that you are not alone - there's 2 great names ready to help protect yours: Unidata and MEA. Unidata has been building world class data loggers right here in Australia since 1984. Under the guidance of their new owner, NIWA, the company has embarked on an ambitious R&D programme, the first result of which is an upgraded Starlogger range. Inside the case you will find a new heart, one which beats as reliably as the old but which, with 4 times the memory of its predecessor, gives you 4 times the storage time for that critical data. Keep watching for details on more new and upgraded products from the company. Like Unidata, MEA too has a 20 year plus history. While Unidata's core skills are in product development and manufacturing, MEA's are in systems integration - building and maintaining systems for clients across the full spectrum of environmental monitoring. As the newly appointed Australian distributor for Unidata, MEA is here to help you to get your job done. Whether it's advice on what can be measured and how, parts for you to build it yourself, supply of a turn-key system or- technical support and calibration, you'll find all the answers at MEA.

Data Loggers, Sensors & Systems for Environmental Monitoring To find out more about how these two great names can help safeguard yours, give us a call or visit our web sites: Unidata Measurement Engineering Australia 40 Ladner St O'CONNOR WA 6163 41 Vine St MAGILL SA 5072 08 9331 8600 Phone 08 8332 9044 08 9331 8611 Fax 08 8332 9577 Web www.unidata.com.au www.mea.com .au Email i5a1es@unidata.com .au sales@mea.com .au

nidata •••••••••


international exceeding a critical control limit, educating the community on good practices chat would protect che water supply system, and responding to enquiries from mem ber households. In che Rural Health Program the Family H ealth Workers regularly visit each family and carry out health checks and provide information on protection of health. • DCH provides information to che village on water supply options and their requirements fo r setting chem up and their operation, and attends meetings of the village committee as necessary. OCH runs a training program which water workers attend. In the case of the Rural Healch Program, for example, 70 training camps are held each year, with some 20000-3000 Family H ealth Workers attending. • In the Rural H ealth Program, a DCH doctor or consultant visits the village periodically, provides services as required, and checks on the operation of che water scheme. An important aspect of this program is the development and provision of guidance on the options that are available for safe water supplies which the villagers could understand and select appropriately. T his guidance includes info rmation on the

requirements for ongoing maintenance and testing. For chis purpose, the project has prepared a Water Supply Manual in a form chat is suitable fo r use by DCH trainers and the villagers. It includes, fo r example, information on: • Establishing a Community Water Supply (steps, requirements, responsibilities, support) • Selection of a Preferred Option (decision basis) • Requirements (short outline of overall req uiremen cs/ cons idera tions) • Siting • Building/commissioning • Sanitation • Operation and maintenance • Monitoring • What to do if things go wrong • Training and awareness • Audit Separate sections have been prepared on a variety of options: e.g. dug well, mbe well, pond sand filter, rainwater tank, head tank and distribu tion system, and use of water within the house. Information on each option includes:

• Applicability • Siring • Building and co mmissioning • Sanitation • Operation and maintenance • Monitoring • What to do if things go wrong (trouble shooting)

Risk Management and HACCP Assuring Water Quality T here has been considerable effort in Australia to develop an improved systems approach to che management of water quality, resulcing in new Australian Drinking Water Guidelines published by the Australian National Healch and Medical Research Council in 2004. T hese guidelines draw on the principles of Hazard Analysis and Critical Control Points (HACCP) used internationally fo r food safety, and ocher management systems such as ISO900 l (Quality Management) and ISO 14001 (Environmental Management). A similar approach has been taken by the World Health Organisation (WHO, 2003).

Its Guidelines for Drinking Water Quality (3 rd Edition) also cake a risk-based approach and incorporate HACCP.

---- -=-s by-J·as CHALLENGER = :> ._,



A complete range of manual and automated valve solutions for the water, sewerage and irrigation industries.

w WaterMark Licence 2105'4

ENVIRO 06 ..... STANDB039

Designer and manufacturer of high effi ciency, low speed floating and fixed surface aerators from 3kW to 220 kW with an unmatched 5 year, unlimited hours guarantee. By-Jas offers flexible financing and delivery solutions including rental, purchase and fully maintained operating leases. Ring now for a current stock list. Other products in our range include settling tanks (12 designs), packaged sewage and water treatment plants, reuse fi lters and clarifiers to Class B and Class A standard.

For more information, contact: By-Jas Engineering Pty Ltd www.cmwvalves.com.au


68 MARCH 2006

FREE CALL 1800 120 751


PO BOX 424, HASTINGS VIC 3915 Tel: (03) 5979 1096 Fax: (03) 5979 1524

refere ed paper

international The project has drawn upon these methods, boch wich respect co: • Formulating a sound overall risk-based management approach co village water supplies; and • D eveloping sound technical guidance chat will ensure char op tions chat are adop ted by villagers will provide for a safe water supply, and wi ll minimise as far as possible che risks inherent in village water supplies. An analysis of how che Safe Water Scheme meets che Requirements of che Australian Framewo rk for D rink ing Water Quality is shown in Table 1. The Water Supply Manual involved a systematic identification of che hazards, as listed below. • The source or cause of a hazard; • The risk associated wich chis if no controls are applied (considering che likelihood and severity of the hazard); • If the risk is high or very high, chen che question is asked: Are the co ntrols specifically designed co eliminate or reduce the likely occurrence of a hazard co an

acceptable level? If the answer is yes, chen this is a Critical Control Point. • If the answer is no, chen the question is asked: Could co ntamination with identified hazard(s) occur in excess of acceptable level(s) or could these increase co unacceptable levels? If the answer is no, then chis is not a Critical Control Point.

• If the answer is yes, then the question is asked: Will a subsequent seep eliminate identified hazard (s) or reduce the likely occurrence co an acceptable level? If the answer is yes, chen chose controls will be the C ritical Control Point. If the answer is no, then there is a need for an improved level of control. This analysis follows the principles of HACCP, and identifi es both the C ritical Control Points (i.e. chose aspects of the system rhac it is cri tical co properly control) and whether, after control, the risk is red uced co an acceptable level. Examples of Critical Control Points chat were identified for a dug well system involving pumping from the dug well to an elevated head tank and reticulating water co

a number of houses, include the controls necessary co ensure that: • Latrines are not located close co the d ug well (e.g. by providing an exclusion zo ne); • Animals are no t housed or penned close co the dug well (e.g. by p roviding an exclusion zone); • Pollu ced water is not able co enter the dug well from the surface (e.g. by providing a concrete apron around the well , provision of an exclusion zo ne, excluding activities such as washing close co che well, and sealing of the joints of the well rings); • Insects (e.g. mosquicos) will nor breed in water (e.g. by providing a screened well cap); • Manual methods of water removal are not adopted (if a pump breaks down) that could give rise co pollution of the water (e.g. by developing a contingency plan for fixing the pump if ic should break down); • Over- use of water does not occur, such chat ic could give rise co contaminated water being drawn inco che well; and

Table 1. Meeting the requirements of the Austra lian Framework for Drinking Water Quality. Provision of safe water that meets WHO Guidelines Government of Bangladesh policy on water supply

Policy/ commitment Requirements

& guidelines

Assessment - understonding of risk issues/ problems

WHO Guidelines for Water Quality As< 50 ug/L Systematic review of options for supplying safe water, including: • the identificotion of water quality hazards, assessment of risks, and development of methods for risk minimisation and control; and • the identificotion and assessment of management and operational factors that pose a risk to the operation of the water supply system.


Development of a Community Safe Water Progrom based on the DCH Rural Health Scheme. This scheme is to be able to be implemented by villages with limited external assistance, or by Local Government Authorities as appropriate.

Progroms, procedures, work instructions, manuals

Development of Water Supply Manual that provides information on : • the management and operation of the program (e.g. for use by DCH and local government); and • technical requirements (e.g. the selection of options, and the implementation, maintenance and operation of each option) . The Water Supply Manual is to be suitable for use by villagers, troiners, a nd technical persons (e.g. Local Government engineers) .


The Water Supply Manual to include requirements far checking systems and monitoring water quality, including: • checks to be carried out prior to construction; • checks to be carried out after construction and prior to use; and • monitoring during operation.

Incident and emergency response

Procedures included in the Water Supply Manual on what needs to be done if things go wrong (e.g. pollution is detected during commissioning or during normal operotion). Tro ining of water workers and DCH staff.

Research and development

Monitoring of research and development and the results of other investigative work relevant to the project. The Centre forms the fo cus for this work, as information relevant to arsenic mitigation and water supplies has been assembled in the Centre and will continue to be expanded and upgraded.

Employee awareness and troining

Development of DCH troining program for water workers, and for trainers of villagers. The provide important information and aids for training.

Community involvement and education

Community will be involved through the community-based progrom, and through the training progrom. DCH will run technical seminars that will disseminate information more widely, including to NGOs and donors.

Government liaison

DCH is a member of the Government of Bangladesh Arsenic Technical Committee. Linkages to be developed with other government departments, including Local Government Division and the Department of Public Health Engineering .

Performance evaluation and continual improvement

Periodic review of the performance of the scheme by DCH.


MARCH 2006


Water Supply Manual will

refereed paper

• The provision of additional water does nor give rise to increased quantities of wastewater that would pose a risk to health. This process of analysis led to the requirement to validate whatever control approach is proposed. For example, if an exclusion zone is proposed, then the question needs to be answered as to how large the exclusion zone needs to be to ensure that water quality will be protected. The advantage of this process was that it provided a systematic method of identifying how risks might arise, and what strategy should be adopted to properly deal with the risk. This assessment has been carried out for each of che water supply options being included in che Water Supply Manual.

Environmental Management While the primary objective of the program was to avoid or minimise exposure to arsenic, it was essential that che strategies adopted should nor impact on the environment. AusAID has guidelines fo r the design and assessment of che environmental aspects of projects, and che program complies with these guidelines. This included: • An initial assessment of che environmental aspects and impacts of the project, at each field site, using a checklist, including factors such as the increased quantity of wastewater chat would result from an improved water supply (particularly if water is reticulated to a larger number of supply points within a village), erosion in the areas of water supply, disposal of treatment sludges (if any), and handling and disposal of chemicals associated with analysis and water treatment. • For each of the impacts identified, development of an appropriate management strategy and environmental management plan. • For each field site, establish ment of a monitori ng program to ensure that environmental management was being implemented effectively, and water quality was being maintained. • Reporting of environmental management activities in the project reports to AusAID.

Conclusions The project is unique in that it has taken a holistic approach to the problem of arsenic in water supplies and, in addition to co ntrolling arsenic exposure through direct ingestion of groundwater, it has also assembled information that would enable the significance of ocher exposure routes through food, soil and dust to be evaluated and controlled as necessary. It has also focussed on establishing a village-centred infrascrucrnre co ensure long-term operation of an appropriate Safe Water System. The approach taken in th is project can be applied generally for the management of small water supplies, and for managing regional water quality problems such as arsenic in mining areas.

Acknowledgments The authors of chis summary gracefully acknowledge: • The support and funding of the project provided by AusAID; • Permission to publish this paper by AusAID; and • The support provided by many individuals within DCH and the organisations involved in the development and carrying out the project.

The Authors Peter Nadebaum (Corresponding author, email peter_nadebaum@GHD.com.au) is Senior Principal - Environment and Alison Baker is Manager, International Development Assistance, GHD Pry Ltd.

refereed paper

Polibrid 705 Solventless Elastomeric Polyurethane Provides the next generation of High Performance Protection for Steel and Concrete in Water / Waste Water Applications

3,000 facilities in the USA have used Polibrid over 20 years to protect their assets in the areas of: • Containment , Waterproofing, Corrosion Control, Abrasion Resistant Lining, Roofing Membrane & Bunding • Polibrid meets the standard for Waste Water and Potable Water in USA, UK, Germany, Brazil & Israel • Zero VOC emissions; Unlimited maximum b uild ; Highly flexible • Now approved to Australian Standards AS4020 for Potable Water • Suitable for new construction or maintenance of deteriorated assets • Applied by trained and approved applicators International Paints, Global Products ..... .... Local Service


Web: www.international-pc.com Email: ipcmarketing@akzonobel.com .au Toll Free:131 474

CHARACTERISING NOM TO ASSESS TREATABILITY C Chow, R Fabris, K Wilkinson, F Fitzgerald, M Drikas Abstract In recent years, considerable research effort has been expended to understand the impact of various forms of nacural organic matter (NOM) on d rinking water treatment processes. To characterise NO M, resin fractionation and high performance size exclusion chromatography (H PSEC) have proven to be useful tools, much simpler than organic chemistry methods They have become more accepted by treatment operators as metho ds for understanding treatment processes, in particular, the optimisation of pilot work prior co commencing new rrearmenc schemes. Boch techniques have been employed co determine the influence of environmental conditions and biodegradation on the character of rhe NOM, rhe efficiency of water treatment processes, rhe id enrifi cation of NOM fractions recalcirranr co conventional treatment and invesrigarion of rhe formation o f d isinfection by-products.

Introduction NOM is a complex matrix of heterogenous organic material which comes fro m decaying terrestrial and aquatic organisms. The composition/character ofNOM can also be affected by seasonal variation. Irs presence in source water can be problematic for rhe production of high quality drinking water and ir is always considered as a key facto r in the d etermination of both coagulant and disinfectant doses. Furthermore, NOM can react with disinfectants to produce disinfection byproducts (DBPs) and also can ace as a carbon food sou rce fo r bacterial growth in distribution systems (Edwards, 1997; Hwang et al., 2000). Our research has focussed on understanding how NOM impacts on treatment processes. This enables optimisation o f water rrearment processes and minimisation of any deleterious changes that may occu r th rough the d istribution system. For a better understanding of the types of organic compounds p resent before and after rhe treatmen t processes, a number of characterisation techniques have been developed worldwide. Some of these are very simple such as colour and UV

74 MARCH 2006


abso rbance (UVabs or UV254nml, wh ilst others such as d issolved organic carbon (DOC), molecular weight d eterminatio n and fractio nation provide more information but require more complex analytical procedures. Some more advanced methods, such as pyrolysis-gas chromatography-mass spectrometry (Py-GC-MS), nuclear magnetic reso nance spectroscopy (NMR) and diffuse refl ectance infra red Fourier transform spectroscopy (DRIFT), require the support of sophisticated analytical instrumentation. However valuable, rhe results are not in a fo rm char can be interpreted easily by rhe treatment plane operators. A suitable analytical cool co assist operators should be both simple and informative. This paper concentrates on two relatively simple tech niques, high performance size exclusion chromatography (HPSEC) and rapid resin fractio nation. Both techniques have found increasing application fo r NOM characterisation , due to their rapidity, reproducib ility and minimal sample preparation (Vuorio et al., 1998), and are useful tools to evaluate treatment processes.

Two rapid analytical techniques for NOM can predict coagulation performance. Experimental Water sources Raw and treated water samples from d ifferent locations th roughout Australia, including New South W ales (NSW), Northern T erricory (NT), Queensland (QLD ), Sourh Australia (SA), Victoria (VIC) and Western Australia (WA), were collected fo r chis scudy. They were chosen co represenr the wide variation in water quality found in Australia and includ ed both surface and ground waters . In addition, some synthetic waters created in the laboratory usin g a jar test co simulate convenrional water rrearment or ocher rrearmen r processes were also used.

River water so urces were represented by rhe Darwi n (NT) River Dam, supplied by the Katherine River, and Morgan (SA) supplied by a combination of catch ment and River Murray water. The reservoir water sources chosen included Little Para (SA), Myponga (SA), Hope Valley (SA) and Moorabool (VIC) Reservoirs. T he Jandakot Mound (WA) was chosen as a groundwater source. It is a shallow unco nfi ned groundwater system, situated near Perth and d irectly replenished by rainfall. Copi Hollow, located in rural NSW, was chosen as an example of a seasonal extreme and does not necessarily represent typical water quality througho ut the year. D ue co its long residence t ime in open scorage, the NOM in the warer was considered co be highly biodegraded and concenrrared th rough evaporation. T his water exhibited a particularly high DOC (36.0 mg/L) and low specifi c UV absorbance (SUVA) of 1.25 m-lmg-lL ar the rime of the investigation.

Dissolved organic carbon analysis DOC concentrations were determined using a total organic carbon analyser (Model 820, Sievers Instruments Inc., USA) as d escribed in Standard M ethods (APHA et al., 1998). UV absorbance at 254nm is often used as a surrogate for DOC (Edzwald , 1993). The absorbance at 254 nm was measured using a UV/VIS spectrophotometer (Model 918 , GBC Scientific Equipment Ltd., Australia) with a 1 cm quartz cell. The specific UV abso rbance (SUVA), UV/DOC x 100, can be used co determine the character of the organics with respect to d egree of conjugation and aromaticity.

Rapid fractionation using resins The rapid fractionation technique is reported in derail in C how et al. (2004) T he tech nique is focussed on short turn around time and small sample volu me, necessary when jar tests are being monitored. It results in four fracrions: very hydrophobic acids (VHA), slightly hydrophobic acids (SHA), hydrophilic charged acids (CHA) and hydrophilic neutral (NEU). T hree 20 cm (length) x 13 mm internal diameter (I D ) glass columns

refereed paper

The new DR5000 UV-VIS Spectrophotometer Simply more ...

... 240+ Test Parameters ... Pre-packaged Reagents ... Intuitive Operation

The Complete Instrument & Reagent Analysis System • Automatic method detection with TNT plus ' " - Bar code reader yields simpler, faster results • Built-in accuracy with automatic averaging - Eliminates errors from blanks & cuvettes • Touch Screen Operation - Easy to use, fast set up • USB Data Communications Port - Easy data export and upgrade integrity

containing DAX-8, XAD-4 and IRA-958 ion exchange resins. (supplied by Supelco (Belefonte, PA, U.S.A.). were sec up in series, as shown in Figure 1.

- 5 - 1 < (360ool.) pH-2

FlowRolt • 1,5.,i.;,.,





• 1.Slllain (0.1B_,)

• 1.50ll.1"'in (0.1B-)


Btd• 14.1..t.


VHA is adsorbed by DAX-8, SHA is adsorbed by XAD-4, CHA is adsorbed by IRA-958 and NEU is nor adsorbed on any of che ion exchange resins. The DOC concentration of each fraction is calculated by che subtraction of che DOC concentration before and after contact with the resin.



With good coordination of the DOC analysis, rhe complete fractionation result can be obtained in 7 hours. The analytical time can be further reduced to less than two hours if only rhe VHA (usually rhe dominant fraction) is to be determined.

High Performance Size Exclusion Chromatography (HPSEC)

Figure 1. A schematic of the rap id fractionation scheme.

HPSEC separates NOM co nsriruenrs based on a differential permeation process, acco rding to molecular weight (size) . NOM is adso rbed on a porous resin and then differentially eluted. Large molecules are unable to penetrate the pores and so elute more qui ckly than small molecules, which are adsorbed and therefore elute more slowly. T he resultant chromatogram is based on measurement of UV adsorbance ar 260 nm.

Applications of rapid fractionation (a) Characterisation of NOM in source water

We used a Waters 2690 separation module and Waters 996 photodiode array detector operating at 260 nm. Separation was performed on a Shodex KW 802.5 column (S hoko Co. Led., Japan) using a 0.1 M phosphate buffer solution (pH 6.80, ionic strength adjusted to 1.0 M with sodium chl oride). T he flow rate was I mL/min and the injection volume was 100 µL. The system was calibrated using polystyrene sulfonate (PSS) standards (Polysciences In c., USA) of mo lecular weights 35,000, 18,000, 8,000 and 4,600 Daltons (Da). This procedure was based on the method described by Chin eta/. (1994).

Although rhe eastern Australian (NSW and VIC) supplies were also surface water samples, they are generally fed from rivers originating from alpin e snow melts togerher with protected catchments and therefore exhi bit low DOC. The lower DOC supplies appear to be generally associated with lesser

Results and Discussion Total dissolved organic carbon (DOC) T he total DOC concentrations of samples, both raw and treated, from around Australia were quite diverse - ranging from low in eastern states (NSW and VIC) and NT (where treatment is nor usually optimised to remove DOC) to very high in SA and WA. where authorities have optimised treatment to obtain significant removal as shown in Figure 2.

76 MARCH 2006


Most data have been obtained from SA where che surface waters are generally higher in DOC (Figure 3). The WA supplies are also high in DOC which is so mewhat surprising as most are groundwater supplies. In both cases chose sa mples high in DOC were found to exhibit high VHA content and low NEU.

VHA and higher NEU content. Samples from NT also had a very low VHA and high NEU. (b) Effect of organic fraction on coagulation Previous research has shown that the removal ofNOM by metal coagulants is influenced by the proportion of humic/no nhumic fractions, rhe hydrophob ic/hydrophilic character and rhe molecular weight (MW) of NOM constituents. Therefo re, the characterisation of NOM is extremely important to rhe water indusrry and considerable research has been undertaken worldwide to establish links between NOM character and creacabilicy (Croue et al., I994; Owen et al., 1995; Gjessing eta/., 1998; Chow eta/. , 2000; van Leeuwen et al., 2002).

14 12


[::=J R [::=J



=Raw =Treated


~ Cl










6.9 6.1





Figure 2. Dissolved orga nic carbon (DOC) levels in Australia - Snap shot from early


refereed paper

The new DR5000 UV-VIS Spectrophotometer Simply more ...

... 240+ Test Parameters ... Pre-packaged Reagents ... Intuitive Operation

The Complete Instrument & Reagent Analysis System • Auto matic method detection with TNTplus r ... - Bar co de reader yields sim pler, faster results • Bui lt-in accuracy with automatic averag ing - Eliminates errors from blanks & cuvettes • Touch Screen Operation - Easy to use, fast set up • USB Data Commu nications Port - Easy data export and upgrade integrity

water treatment The organic fractions remaining after treatment at various doses indicate chat increased alum doses result in increased removal of rhe VHA and SHA fractions and to a lesser extent the CHA fractio n, but little or no removal of rhe NEU fraction. (Figure 4). Consequently, the character of treated water differs sign ificantly from that of raw water. Details of chis experiment can be found in Chow et al. (2004) . This suggests chat a greater proportion of DOC can be removed by coagulation from water with a higher hydrophobic (VHA/SHA) content. T his was confirmed by an eighteen month case study which moni tored the organic character and applied alum dose in rwo South Australian water treatment plants. During this study Qan 2001 to Jul 2002), the raw water DOC concentration increased, however, the applied alum dose required to maintain constant treated water DOC concentration decreased for both WTPs (one of rhe criteria to select rhe ap plied alum d ose was based on the previous month DOC concentration in the treated water). It was fo und that the increase in organic carbo n was main ly an increase in concentration of the VHA fraction. This fraction typically has the characteristic of being easily removed by alum treatment. This extra organic fractio natio n information can be used as a feed forward control of the coagulant dose to allow more timely adjustment to maintain treated water quality (Chow et al., 2005).

Applications of High Performance Size Exclusion Chromatography (HPSEC) HPSEC has proved to be a useful technique for evaluating various water treatment processes. NOM removal can be determined by comparing NOM profiles before and after treatment (Gjessing et al., 1998; Bairo et al., 1999; Chow et al., 2000; Cook et al., 2001; van Leeuwen et al., 2002) . Additionally, H PSEC determination of rhe molecular weight distribution of NOM has also indicated a significant correlation with ch lorine demand (Vuoria et al., 1998). (a) Organic character in source water Six d ifferent water sources from various locations in Australia were analysed by HPSEC to compare molecular weight profiles of differen r water sources. Aside from the obvious d ifference in NOM concentration, the profiles indicate several interesting d ifferences in NOM character, as shown in Figure 5. Ir is evident char rhe river water sou rces (DOC: Little Para 5.0 mg/L; Morgan 6. 1 mg/L; and Darwin River Dam 3.2 mg/L) contain substantially lower NOM

78 MARCH 2006


DOC: 1 - 3 mg/L VHA: 19-38 % SHA: 14 - 25 % CHA: 6-7% NEU: 33- 50% o

DOC: 1 - 6 mg/L VHA: 35 - 58 % SHA: 12-24 % CHA: 6-8 % NEU:19-26%


DOC: 6 - 13 mg/L VHA: 60-82 % SHA:7-21 % CHA:4 - 6 % NEU:7 - 22%

DOC: 2 - 3 mg/L VHA: 40-54 % SHA: 15-23 % CHA: 5-9 % NEU:20-32%


DOC: 5 - 14 mg/L VHA: 50-70 % SHA: 15-25 % CHA:5-17% NEU: 6-16 %


DOC: 1 - 2 mg/L VHA: 16-40 % SHA: 15 - 31 % CHA: 5-8 % NEU: 20 -56 %

Figure 3. Resin fractionation information from a selectio n of Australia n raw waters. Snap shot in 2003.


7~; ~ SHA 25%

VHA 51 %



CHA 14%



22'¾j o )V HA CHA 13%

SHA 20%

56 .,_,


VHA 50%


SH 20%

§ 4


NEU 29%( 1 \VHA ~ 37% CHA 16%


SHA 18%


~ (.)







Alum Dose (mg/L) Figure 4. The fractionated dissolved organic carbon (DOC) concentration in alum treated water. (b)




Copl Hollow Raw M;ponga Raw


Jandakot Raw Little Para Raw Morgan Raw Darwin River Dam








Jlndakot Raw LittlePtraRaw M>rganRaw Derw In RN-er Cum

~0.004 E





:::, 0.001

100 1000 10000 100000 Apparent Molecular Weight (Da)

0.000 -.===-- ~ ~_.:,,~___..::,,,.__, 100 1000 10000 Apparent Molecular Weight (Da)

Figure 5. Comparison of HPSEC NOM profiles for different water sources (a ) full a nd (b) base enlarged .

refereed paper


PVC-0 .._____8

Manufactured to comply with AS 4441 (lnt}:2003 Minimum lifespan of 100 years Australian innovation

info@vinidex.com.au www.vinidex.com.au

Vi ~ Systems & Solutions

*Comparisons have been based on Vinidex Vinyl Iron pipe and Supermain pipe.


Endorsed Company


eo to0, uo 110

water treatment In each case, alum treatment concentrations than the other Little Para Raw Aug 04 resulted in the general removal water sources (DOC: Myponga 0.010 Little Para Ra~ of the higher molecular weight 12.6 mg/ L; Jandakot 11.5 mg/L; aromatic organic compounds, as and Copi Hollow 36.0 mg/L) o.ooa u evidenced by changes in NOM and their profiles are 0.006 profiles (Figure 7 - region a) and characterised by lower average C 0 UV absorbance values (not molecular weights (Mw: Little ~ 0.004 shown). Interestingly, a Para 1140 Da; Morgan 1010 Da; @ significa nt variation in the and Darwin River Dam 890 Da). 0.00 2 removal of the low MW NOM The NOM profiles for Little Para 0.000 (specifically Mw 500 to 700 Da; and Morgan are similar, as 100000 100 1000 10000 Figure 7 - region b) was expected, since both reservoirs are observed between the four water Apparent Molecular Weight (Da) supplied essentially by the same samples. Approximately 50% of river source. The Copi Hollow the low MW fraction cou ld be water is unusual, with an average Figure 6. Compari son of HPSEC NOM profiles of the Little removed from both Myponga molecular weight (Mw: 1070 Da) Poro reservoir before and after a heavy rai n event ( l in 10 and Hope Valley water, whereas similar to chose waters of low year storm) . alum treatment had licrle impact DOC concentration. on the low MW fraction of Myponga and Jandakoc exhibit either Copi Hollow or similar molecular weight profiles H PSEC has also been used co in vestigate Mooraboo l water. T his suggests chat (Mw: Myponga 1330 Da; and Jandakoc che infl uence of environ mental diffe rent water samples could contain 1460 Da); both sources have high NOM conditions on NOM character. For NOM co nstituents of similar MW, but concentratio ns and comprise higher MW example, Figure 6 shows NOM profi les significantly different chemical fractions relative to the river water sources. of Little Para Reservoir, before (May) and composition. Recalcitrant NOM, chat is, The longer catchment residence times after (Aug) heavy rainfall. The poseNOM remaining after coagulation, is experienced by these waters would not only rainfall profile shows significant increases thought to include low MW conjugated explain the increased NOM conce ntration, in NOM conce ntratio n an d in particular organic compo unds. This fractio n is of but also the differences in NOM character, an increase in higher MW fractions, particular importance to che water due to a greater extent of biodegradation. characteristic of catchme nt so urce water. treatment industry, since it is associated Perhaps che most significant difference Furthermore, an early eluting (h igh MW) with the for mation of disinfectio n bybetween these two water sources is the peak, similar to that of Mypo nga water products and bacterial regrowth within the presence of early eluting fractions, i.e. che (F igure 5) is also observed. T his fu rther distribution system (Edwards, 1997) . peaks above 50,000 Da. The composition sugges ts chat higher MW fractions are to determine the efficiency of In order ofNOM responsible for these high MW derived from vegetatio n decay from the NOM removal and identify recalcitrant fractions is clearly different for each of the catchment. Th is h igh MW peak (above of NOM, sequential alum fractions water sources and is thought to comprise 50,000 Da) will be discussed furthe r in treatment of Hope Valley Reservoir water colloidal material (organometallic the 'Compariso n of differe nt treatment was undertaken In chis study, an optimal complexes). methods' section. alum dose of 30 mg/L was determ ined for (b) Understanding treatment processes Hope Valley water according to jar tescs and the impact of NOM using HPSEC conducted at pH 6, and a five-stage HPSEC was used to investigate the impact (a) treatment was employed. H PSEC analysis of NOM on processes, such as coagulation, was performed on raw water, and created disinfection and biodegradation,by water after l , 3 and 5 successive comparing relative molecular weight applications of 30 mg/L alum doses, as profiles before and after treatment, and shown in Figure 8. Details of the sequential identifying chose fractions most likely to study can be obtained from Chow et al. affect treatabilicy. (1999).




i) Coagulation The effect of coagulation on NOM removal fo r four water sources, Copi Hollow, Myponga, Moorabool and Hope Valley, was investigated by H PSEC. As shown in Table 1, the raw water quality parameters for each water source are quite varied.

T he HPSEC profiles were processed using Peakfit software (Version 4, Syscac Software Inc.), an advanced data treatment process utilising peak fitti ng techniques. Fo ur peaks were identified in the raw water NOM profile and the average molecular weights, Mw, of the resolved peaks were calculated

Table 1. Summary of raw water quality parameters. 100



Aparert Molccuhr Wcw< (Dal


Copi Hollow



Hope Valley 3.9

DOC (mg/L)




Figure 7. Molecular we ight profiles of

Colour (HU)





fou r water sources, Copi Hollow, Myponga, Moorabool, Hope Va lley. R: raw and T: alum trea ted.

UV obsorbonce @ 254 nm lcm·1)



0. 178




2. 58



MARCH 2006


SUVA (m·1mg·1 L)

refereed paper


Look to the future with the Wastewater Professionals When you work professionally with wastewater management, you want to be sure that your systems work really well. And that they work all the time, now and in the future. Grundfos products are made to last. And now, Grundfos offers a full range of wastewater pumps, mixers, and flowmakers specia lly

i m

mo or sizes up to 52


el combines top perform a nee with

complete reliability. So whether you are planning an upgrade or a wtiole new wa stewater system, choose a supplier th at always looks ahead. Go to tfle Wastewater Professionals. Go to Grundfos.

water treatment










"''""'I ill ::_j__---<2.......-<~~

2days 5 days 7 days 9days -12days 14days




,,~. .I





~ ~


0.003 0.002 0.001 0.000 100

!2~-- -JX\""--'"->-- -----





Appa rent Mo1ocu1ar vvo1gm (Da)



Figure 9. Changes to HPSEC profil e of treated water during a BDOC stu dy over 14

: :~ _ __ ,/}C\L.._.L,.....:.,__ _ _ _ __





Apparent Molecular Weight (Da)

Figure 8. HPSEC profiles with peak

ii) Assessing NOM biodegradation

fitting after sequential alum treatment (successive application of 30 mg/L alum doses) of Hope Va lley water.

HPSEC has also been used to investigate rhe changes in NOM composition during studies of biodegradable disso lved organic carbon (BDOC) which supporrs bacterial growth in the distribution system. T he method measures the amount of organ ic matter that is biodegraded by a sand co lonised by bacteria from a water treatment plant filter. Figure 9 shows the changes in molecular weight profile of alum treated Myponga water (laboratory pilot plant study) during biodegradation over 14 days. Initially, a general reduction of the main NOM fract ions was observed; in particular che peaks of 350, 750 and 1000 Da average MW. The reduction in MW fractions between 400 and 1000 Da continued throughout che study, but

to be 500, 800, 1100 and 1400 Da. A co mparison of the HPSEC profiles indicated char rhe rwo peaks of higher MW (Mw: 1100 and 1400 Da) were readily removed by the first alum dose. Although further addition of alum resulted in some reduction of the 800 Da MW peak, the 500 Da peak, and a significant proportion of the 800 Da peak were still present, even after five successive alum treatments. T his suggests that these recalcicrane fractions would not be removed during conventio nal water treatment.

Ho(!e Valley 0.024

0.010 8 C


Mroonga 0.020





@) 0.006




.,C ...





6 0.000







1000 2000 3000 4000 5000

Apparent Molecular Weight (Da)


1000 2000 3000 4000 5000 Apparent Molecular Weight (Da)

Figure 10. Molecular weight distribution of UV absorbing compounds present in raw and treated Hope Valley and Myponga waters. (a) A lum coagulation without pH adjustment; (b) alum coagulation at pH 6; (c) MIEX®; and (d) com bined alum/MIEX® treatment.

82 MARCH 2006


interestingly, che low MW peak (Mw: 350 Da) actually increased afrer 7 days. This suggests a change in composition of the 250 Da NOM fraction; possibly, bioconvers ion of low MW constituents co more stabl e produces of similar MW. H PSEC analysis also indicates che formation of new NOM fractions, of low (<200 D a) and high (>1 050 Da) MW. This broadening of che molecular weight d istribution suggests char biodegradacion processes have resulted in significant changes co the chem ical composition (character) of the water's NOM and can be used as a characterist ic co ident ify where biologically derived changes have occurred.

iii) Comparison of different treatment methods A particularly useful application o f HPSEC is the ability to compare different water treatment methods. Figure 10 shows the molecular weight distribution of Hope Valley and Myponga water created by four different m ethods: (a) alum coagulation without pH adjustment; (b) alum coagula tion at pH 6; (c) M IEX®; and (d) combined alum and MIEX®. Treatment derails can be obtained from Drikas et al. (2 003). A certain degree of overlap of NOM removal is observed between the alum and MIEX® treatment processes. Alum coagulation (a, b ) readily removed high MW UV absorbing compounds, particularly chose of MW >20 00 Da. In contrast, MIEX® treatment (c) removed a much broader NOM MW range, including chose fractions which were not easily removed by alum coagulation (i.e. MW <2000 Da).

refereed paper


F/One Sewer Systems can make tough sites buildable - and, cut your sewering costs up to 50%. With F/One Sewer systems, you can develop parcels where gravity sewers are too expensive - or simply impossible to put in. No massive trenches. The F/One low pressure system uses a small main in a shallow trench that follows the contour of the land.

Easy in -

at half the cost.

You can sewer virtually anywhere. Including sites where old septic systems may be dying and polluting. And, F/One systems are totally reliable - no preventive maintenance, all but invisible.

Let us prove it - free. Send us the topo map for your next challenging project. We'll show you how an F/One system can make your project viable - and save you up to 50% over a conventional gravity system. Call, write, fax, or visit us on line. Environment One Corporation, 49n 6 Bayview Street, Runaway Bay, Queensland, Australia 4216 Voice (07) 5537 8807 Fax (07) 5537 8809 www.eone.com/au A PCC Flow Tech110/ogies, Inc. Company



water treatment A combined alum/MIEX速 treatment process (d) resulted in the production of created water with no high MW UV absorbi ng NOM (>2000 Da) and a significant reduction in low MW UV absorbing NOM (<200 0 Da) . The NOM profiles of both treated waters after combined alum/MIEX速 rrearmenr are very similar, indicating the recalcitrant NOM from both water sources is likely to comprise rhe same chemical constituents. A further interesting point is that the high MW peaks (above 50,000 Da) in Figu res 5 and 6 are generally considered to be colloidal organic materials.However, this peak remains undefin ed in molecular weight because the molecular weigh t exclusion limit of the col umn used in the separation was 50 ,000 Da. From our pasr observation , rhis peak is easily removed by coagulation but nor by rhe MIEX速 process.






so N @)

"'co 0.010


> ::)

0.000 100

84 MARCH 2006







-Treated - Prior to Disinfection After 7 days Chloramination - After 7 days Chlorination

0.006 ,






> ::)

This paper reports a small sub-ser of the research and development work which was focussed on assisting the water industry in understanding NOM and its behaviour in water treatment processes. Rapid resin fractionation and H PSEC have been demonstrated as convenient and useful


Apparent Molecular Weight (Da)

Recalcitrant NOM can react chemically to produce halogenated DBPs and it is widely accep ted rhar low MW NOM is rhe precursor (Hwang et al., 2000). HPSEC analysis was performed on chlorinated and chloraminared Myponga water (both raw and rreared) to invesrigare the impact of disinfection on NOM character.


After 7 days Chloramination After 7 days Chlorination


~ .0

T he result indicated that HPSEC is potentially a useful analytical tool to study organic ch aracter in distribution systems.

Raw - Prior to Disinfection



iv) Impact ofNOM on disinfection

Ir is clear char for borh raw and rreared warer, chlorination has a greater impact o n NOM than chloraminarion, due to the higher oxidation effi ciency of chlo rine. Borh disinfectants resulted in a red uction of UV absorbing compounds across a very b road MW range after 7 days contact rime, with a significantly greater red uction observed in the case of ch lorination, as shown in Figure 11 T he selection of the initial disinfectant dose was based upon rhe criteria char residual had to be main tained over the 7-day period. Ir is also worth noticing char rhe colloidal high MW peak d iscussed in the previous section was unchanged (Figure l la) after disi nfection either by chlorination or chlorami narion. This probably indicates rhar the colloidal organic matter has nor reacted wirh disinfectant and so is unlikely to form D BPs.






0.000 100




Apparent Molecular Weight (Da)

Figure 11 . Comparison of molecular weight (HPSEC) profi les of (a) row and (b) treated Mypongo water and the influence of disinfection (chlorination and chloramination). analyrical tools for the characterisation of NOM in potable water. The examples shown in this paper represent only a fraction of their application; however, they demonstrate rhe importance of organic characterisation and the impact ofNOM on water treatmen t processes. The rapid fractionation technique can be used to determine the concentration of the fou r organic factio ns, VHA, SHA, CHA and NEU ofNOM. Ir is possible to use these relative percen rages to assess treatment performance. In addition, rhe organic fractionation survey aro und Australia has ind icated waters from easte rn Australia (NSW and VIC) and NT have low DOC; however, these low DOC waters are also associated with lower VHA and higher NEU. Frorp this it can be conclud ed char although these waters have low DOC, ir would b e d ifficult to reduce it further. HPSEC is b ecoming a popular tech nique chosen by the warer operators for various investigations. Similar to rapid fractionation, HPSEC scans can be used to distinguish between various water sources, such as river, ground and surface ere. and the knowledge gained can be used to assess and predict coagulation performance. In

ad d ition , HPSEC can also be used as a tool to assess MIEX速, biological rrearmenr and disinfectio n processes. The major limi tation to H PSEC is its reliance on UV detection: not all NOM constituents exhibit UV activity and a p roject withi n the CRC for Water Quality and Treatment is currently involved in develop ing a DOC-specific detector.

Acknowledgments This work included the results from a number of CRC Water Quality and T reatment projects, 321,2 11 , 23 11 , 2403 and 250 1. The experimental work of David Cook and Maxime Favier is gratefully acknowledged.

The Authors Christopher Chow (email Chris.Chow@sawater.com.au) is a senior research scien tist, Rolando Fabris, Kerry Wilkinson and Fiona Fitzgerald are research scientists, and Mary Drikas is research leader water treatment at the CRC for Water Quality and T reatment, Australian Water Quality Centre, Private Mail Bag 3, Salisbury, South Australia, 51 0 8, Australia.

refereed paper

water treatment References APHA, A WWA and W E F 1998 Standard Methods For The Examination of Water and Waste Water, 20th Edition, American Public Health Association, Washington, DC. Bolto B., Abbe-Braum G., Dixon 0., Eldridge R., Frimmel F., H esse S., King S. and Toifl M . ( 1999) Experimental evaluat ion of cationic polyelectrolytes for removing nat ural organic matter fro m water. Wat. Sci. Tech. 40 (9), 71-79. Ch in Y.P. , Aiken G. and O 'Loughlin E. (1994) Molecular weight, polydispersicy and spectroscopic properties of aquat ic humic substances. Environ. Sci. Technol. 28( 1 I), 1853- 1858. Chow C.W.K., van Leeuwen J.A., Drikas M., Fabris R., Spark K.M. and Page D .W. (l 999) The impact of the character of natural organic matter in conventional treatment with alum. Wat. Sci. Tech. 4 0(9) 97- 104. C how C .W.K., van Leeuwen J ., Fabris R. , King S., Withers N., K. Spark and Drikas M. (2000) Enhanced coagulation for removal ofdissolved organic carbon with alum - A fractionation app roach. Wa terTEC H Conference, April 9-12, Australian Wacer Associat ion, Sydney (Australia) . C how C.W. K., Fabris R. and D rikas M. (2004) A rapid fractionation technique to charac-

cerise nat ural organic matter for the optimisation of water treatment processes . J W'ater SRT - Aqua 53 (2): 85-92. Chow C.W.K., Fabris R., Drikas M. and Mike H olmes (2005)in press) A case study of treatment performance and organic character. J Water SRT -Aqua. 54 (6): 385395. Cook 0 ., C how C.W.K. and Orikas M . (200 1) A laboratory study ofconventional alum treatment versus MIEX® treatment for the removal of natural organic matter. 19th Federal Convention, April 1-4, Australian Water Associat ion, Canberra (Australia). CroueJ . P., Marcin B., Degu in A. and Legube B. ( I 994) Isolation and characterisation of dissolved hydrophobic and hydrophilic organic substances of II reservoir water, natural organic matter in drinking water. Proceedings American Water Works Association Conference, Denver (USA), p 73. Drikas M., C how C.W.K. and Cook D. (2003) The impact of recalcitrant organic character on disinfection srabiliry, trihalomethane format ion and bacterial regrowth - An evaluation of magnetic ion exchange resin (MIEX®) and Alum Coagulation.} Water SRT -Aqua 52(7) 475-487. Edzwald, J. K. ( 1993) Coagulation in drinking water treat ment: Particles, organics and coagulants. Wat. Sci. Tech. 27(11 ), 21-35.

Edwards M . (1997) Predict ing DOC removal during enhanced coagulation. /. Am. Water Works Assoc. 89(5), 78-89. Gjessing E.T. , Alberts J.J. , Bruche c A., Egeberg P.K. , Lydersen E., McGown L. B., Mobed J .J., Munster U., Pempkowiak J. , Perdue M ., Ramawerra H., Rybacki 0 ., Takacs M. and Abbe-Braun G. ( 1998) Multi-method characterisat ion of natural organic matter isolated from water: characterisation of reverse osmosis-isolates from water of two semiident ical dystrophic lake basins in Norway. Wat. Res. 32( 10), 31 08-3 124. H wang C.J ., Sclimente M .J. and Krasner S.W. (2000) In Natural Organic Matter and Disinfection By-Products; Barrer S. E. , Krasner S.W. and Amy G.L. Eds.; ACS Symposoium Series 761; American Chemical Society: Washington, DC, pp 173-187. Owen D.M., Amy G .L., C howdbu ry Z. K., Paode R., McC oy G . and Viscosi! K. (l 995) NOM characterisation and trea rabi liry. j. A m. Water Works Assoc. 87( 1), 46-63. van Leeuwen J ., Chow C., Fabris R. , Withers N., Page D. and Orikas M. (2002) Application of a Fractionation Technique for the Better Understanding of the Removal of NOM by Alum Coagulat ion . Wat. Sci. Tech.: Water Supply, 2 (5-6), 427-433. Vuorio E., Vahala R., Rintala J. an d Laukkanen R. (1998) The Evaluation of Drinking Water T reatment Performed with H PSEC. Environment fnternfltio11al, 24 (5-6), 617-623.

••• ••• Australian




refereed paper


MARCH 2006


QUALITY OF WATER RESOURCES: BIOMONITORING SOLUTIONS A Andreoli, S Fass Physlco-chemlcal sensors and analysers

Abstract T he Centre Internacio nal de !'Eau de Nancy, (NANCIE) in north-eastern France, under a European Com mission LIFE 99 ENV/F/ 000492 project, h as reseed for some years the respo nses o f an array of five different biomonitors, arranged in sequence on the same water flow, along with the usual physico-chemical monitors. They are d eveloping computer processing of the information, thus giving the water manager an earlier and more complete picture of the quality of the water.

Keywords: Biosensor, biomonitoring, water resource, early warning systems, Fluotox, Gymnotox, T oxConcrol

ChaMl.1 2 1


Cho1sl1 1 , •Turbtdity


•Temperature • pH • Dissolved oxygen • ConductNity


• Hydrocarbons

~ii .. . . '



:z: I lJl,,

Editor: The following paper in this issue, by

O'Connor et al, discusses the need for realtime earl.y warning systems (EWS) and the limitations ofpurel.y physico-chemical monitors fo r detection of any of the infinite array of contaminants and micro-pollutants which could enter the water source, either by accident, from natural events, or intentionall.y, from vandals or terrorists. The authors of this current paper have also discussed these matters in full, but to avoid undue repetition, the Editor has deleted a considerable part oftheir Introduction, in order to concentrate on their valuable results.

Monitoring and Safety of Water Supplies In general terms, Early Warning Systems (EWS) have to achieve two d ifferent but complementary obj ectives: the firs t is to preserve h uman health by detecting the p resence o f toxic com pounds in the water resources before the d rinking water treatmen t. The second objective is the protection of aquatic organ isms in their natural environment.



Oymnotox Elt!ctrogen~sls in fishes. DrelSNM Monitor

Vatvar acttvily of mussels

i=m Aquatox comrol

SMmm.ng achv,ty of daphnM

• the d istrib utors, so chat they can meet the quality requirements d ue to the regulation in the water they distribute • the water pol ice and water agencies, to identify the origin of any pollution and remedy the problem.

86 MARCH 2006


Bacteria tut Lummescence

of bacteria

Fluotox Algal pholosynthesis

Blologlcal detectors

Figure 1. BEWS station implemented in 1999 in Nancy, France.

The present article o nly focuses on biomo ni toring in the water resources; ocher researchers work on the biomon itori ng implemented in d rinking water distrib ution networks (SAFER project) . T he biomoni toring before treatment enables either to mod ify the treatment itself or to stop the pumping of rhe contaminated resource, whereas the biomon itoring after treatmen t solely aims to p rotect the health of rhe consumers.

A research station in France has been comparing five types of biomonitors since 1999.

Information from EWS is essential for: • the producers, so chat they can optimise the treatment of the water


• Nitrate • Ortho-phosphDIH

• Redox potenli.1l

•Chtorophytl a



Choul1 J ,

The classical monitoring of rhe resource, i.e. physico-chemical monitoring, permits no more than limited application. The in reraction of substances, the growing number of pollutants, the presence of compounds in q uantities below the detection thresholds of chemical analyses -

all these are arg uments in favou r of rhe concept of integratio n of all the effects of physical and chemical parameters. Fo r chis reason, it has now beco me viral to link the physical and chemical d ata to biological infor mation. Several examples of Biological Early Warning Systems (BEWS) are currently in operatio n in Germany, the Netherlands and France (see EW S web sire) . T hese EWS and BEWS have a particular respo nsibility to alert the water managers in rhe event of accidental pollution of the resource. T his type of pollution is d efin ed as "the identification, either by d irect observation or by laboratory tests, of a harmful but non-permanent effect on the surface water or grou ndwater arising either from an unforeseeable o r unintentio nal event, o r from an event caused more or less intentionally" (French interdepa rtmental circular, 4rh July 1972). From chis defin ition, we gather that accidental pollu tion incidents are characterised by a sudd en, un foreseeable and marked qualitative change in the ecosystem (so me estimates put the

daphnia Scsnsdssmus 400x

,, Secondary consumer = fish

producer= alga Figure 2. O rganisms representing 5 links in the food chain.

frequency of such pollution incidents on French hydrographical networks as high as more than a hundred per year). It is therefore necessary co be equipped with adequate cools in order co be able co give the alarm. This is the primary ro le of the monicoring and warning stations. Furthermore, the information must be provided quickly, especially if corrective measures can be taken (such as adaptation of the treatment procedure fo r water destined co be processed for human consumption). When po llu tion occurs, the monitoring stations also provide a means of monicoring the way the situation changes unti l such time as cond itions return co normal.

the project undertaken in Nancy was co adopt more realistic solutions for improved monitoring and ma nagement of water quality. To th is end, we demonstrated that monitoring and warning objectives on hydro-systems can be attai ned by setting up relatively classical physico-chemical sensors and analysers in association with more innovative cools: biosensors (Figure 1) as described in detail in the rest of this article.

fn addition co the detection of accidental pollution incidents, the EWS & BEWS must also keep the managers informed at all times of the qualitative condition of the resource and present trends.

The first systems using living organ isms co gather information on the quality of water actually made their appearance in the late ni neteen-sixties. The water was assessed continuously and automatically, with a view co monitoring industrial effluents. Lacer on, the field of investigation was broadened co include raw water destined for human consumption and surface water in general.

When a biological approach is adopted fo r the monitoring of an aquatic mili eu, it is primarily with the intention of taking advantage of the integrative capaciry of biological organisms in general. Bue thei r capaci ties go much fu rther than chis. Thus, if we consider the sec of characteristics to which these warning systems correspond , the ideal biodecector - if such a th ing exists - would be sensitive, rapid, reliable, automatic, continuous, on-line, easy to use, requiring little maintenance, and providing long life at low cost. Set against these desirable fea tures are constraints and difficul ties which should not be underestimated. This consideration makes it necessary co redefine very precisely what can be legitimately expected of biosurveillance, and if necessary to reco nsider the givens pertaining co the problem .

Despite the great diversity of biological su rveillance systems, various basic constituent elements are found in all of chem: fi rst of all the sensitive element (the biological organism or reagent) which is continuously exposed co a flow of water. It develops a response (= an image of its environment) which is detected and transformed into a signal (most frequently electric) by the transducer(= mode of detection). A processing unit converts this signal inco a fo rm suitable for evaluation,

First of all, it must be emphasised chat it is unreasonable to expect more of a system than what it was designed to provide. Thus biosensors or biological warning systems are adapted to detection of relatively marked, short-term changes in the quality of the milieu. Fu rthermore, the nature of the information provided by these systems is very simple (YES/NO type): that is co say that they express the fact that there is or is not a problem of quality with the water. The purpose of biosensors - and it is

Biodetectors or biosensors

These are defi ned as any system dedicated co the continuous and aucomatic detection, pcocessing and analysis of the biological responses elicited in aquatic organ isms for the surveillance of an aquatic milieu.

T he current situation as regards qualitative moni tori ng of water resources reveals a co nsiderable degree of contrast. We find that so me French urban areas completely or very nearly unprovided with monicoring and warning installations, whereas others are lavishly equi pped, more like fu ll-blown on-line analytical laboratories involving heavy investment and operati ng costs. The situation with which water managers are confronted every day co ncerns both the nature of the information to be obtained and the means co gather and process the data. Before the present project, there were already complaints about current trends cowards the provision of more and more complex analysers. Thus one of the tasks of

by computing or electronic means, for example. By reading the information in accordance with an algorithm, it then becomes possible co disti nguish between a normal situation and the appearance of coxic conditions, for example (when an alert will be triggered).

\ I n dep e ndent laboratory toxicity testing service for the ANZECC Water Quality, sediment quality ocean disposal guidelines . Toxicity testing of • Efflu e nt s • Rece ivin g w a t ers • L eac ha tes • G ro un d wa t e r • Sed i me nt s • Soi l s • D re d ge spo il • C h e mi ca l s • N A T A e n d or se d t o AS I SO 1 7 0 25 • Tox i c ity I de ntif ica t io n Ev a lu a ti o n ( TI E) s tudi es · Ex p e ri ence a n d ex p er ti se • Rapi d turn a ro u nd tim es • Cos t eff ec tiv e t es t s unit 27/2 chaplin d rive lane cove nsw 2066

T>61 2 9420 9480


NATA F>61 2 9420 9484

ecot~x S ERVI C E S



MARCH 2006 87

Table 1. Classification of biosensors by kind of organism used. Organism

Principle used


Respiration (oxygen consumption or CO2 production)


5 mg BOD/I


Microtox or ToxControl

3.5 Dichlorophenol - 0.6

Fluorescence based on photosynthetic activity

DF-algae test, Fluotox, Aquasentinel (Greenbaum

Atrazine - 0.001 Copper - 0.8 Lindane - 0.001 Chlorophyll A· 0.001



Sensitivity (mg/I)

et al., 2004) Euglena

Movement parameters of the motile un icellular flagellate (Tahedl and Hoder, 2001)

Daphnia Mussels


Locomotion behaviour using camera

BBE daphnia taximeter


Phototactic responses

Aquatox control daphnia test

Chrome -1.5

Filtration activity

Mussel Monitor® (Kra mer and Foekma, 2000) Dreissena Monitor (Pothus et al, 2002)

Tributylinoxide - 0.04 Pentachlorophenol (PCP) - 0.01

Exploitation of the valve movement


Ammonium · 37

Death (not in use anymore)



Locomotive behaviour

Tru itosem, BBE Fish taximeter

Lindane 0.013, PCP 0.24

Physiological criteria (such as respiration or heart beat)


N/ A

Electrical activity


Trichloroethylene - 1 Chrome - 1.6 Cyanure - 0.035 Gas oil- 1.9



Selection of an enzyme known to be inh ibited by a specific pollutant



Polymerase chain reaction of a specific gene chosen to detect one or two pollutants (Gresle and Bridoux, 2005)


important co stress chis - is not co define the nature of the problem (or to characterise the type of pollucion), although ch is is sometimes possible in certain specific cases, but co alert if something wrong occurs. The two capabilities, sensitivity and speed of detection, of qualitative changes in their milieu, constitute a major part of what could be designated the level of performa nce of a biosensor. This level will be different, d epending on:

• the test organism (used as a "biological reagent"), • the response observed,

To this end, fi ve rest organisms, representing fi ve different links in the food chain, have been selected (Figure 2).

• the techniques of measurement,

They are:

• the pollutant (type, concentration , mode of action). There is no such thing as the ideal warning system or che universal sentinel organism, so therefore NANCIE put forwa rd a new philosophy: the mulcicrophic rou te.

L in e a r s tepper motor



Chrome VI - 27.8; 2-4 DCP - 2.1


L ight t ight casi ng

..,_ Syringe

l. Bacteria, micro-organisms belonging co rhe trophic level of the decomposers. They can transform dissolved organic or inorganic matter into particulate one, thus making ic available for ocher (micro) organisms 2. Algae, representing the trophic level of the primary producers. These photosynthesising organisms exploit mineral substances, which they transform into organic matter 3 . Daphnia, the first link in the chain of consumers. They are herbivores 4. Mussels, the second link in the food chain. They feed on phytoplankton and bacteria 5. Fishes, the lase link in the trophic chain . Bio logical monitoring is being everincreasingly used . The current literature abounds in biological early warning systems, the diversity of which is attested to not only by the variety of species used bu t also by the number of principles developed (Table 1). O n the market up co 40 biosensors are available nowadays, from prototype co fully commercial.

Figure 3. Schematic view of Tox Control biosensor.

88 MARCH 2006


Apart from che trophic ch ain, there are now enzymatic and even DNA based biosensors (G resle and Bridoux, 2005).

Those new technologies appeared one co two years ago on the market.

behaviour is modified. T he change is aucomacically detected Informa tion sent to and generates an alarm. le needs processi ng system co be maintained once a week. In Nancy, we built the station in 1999 and have run it ever since, The Dreissena-Monitor on water pum ped from the biosensor analyses the valve Blue LED Moselle River. Our biosensors activity of mussels. It utilises two were those of the market at that types of response: the percentage time. Since then most of those of mussels open and the number biosensors have been improved in of valve movements pe r mussel terms of robustness, compactness, per hour. Each mussel is reliability. Supply of the living immovably fi xed co a support Filtering membrane organisms is now more (Figure 6) . A magnet is glued co convenient, and most can just be one of its valves. A sen sor kept in a fridge. New biosensors positioned in close proximity is even uti lise naturally occurrin g water to be analysed used co record the magnetic field wild algae as the test agent which is a fun ction of how far the Figure 4. Sketch of Fluotox system. (McCarcer and Kassinos, 2005). mussel is open. Under normal Biosensors prices range from cond itions, the mussels are open â‚Ź20,000 co â‚Ź30,000. co enable alimentary and The Fluotox biosensor exploits resp iratory exchanges with the aquatic The Biosensors under test phocosynchecic activity of algal cells. The environment. If pollution occurs, rhe T he prin ciples of our five sensors is briefly algae are fi xed on a cell ulose memb rane, animal naturally closes irs valve while rhe add ressed. placed in the curren t of water under test. number of valve movements qui ckly T he ToxControl biosensor (old After luminous excitation, the algae emit increases. Data is exported every 5 minutes fluorescent light, wh ich is captured by a Regensburger Leuchbakteriencest) exploits and rhe unit needs atten tion once a month. the natural luminescent properties of photocell (Figure 4). T he presence of T he fifth biosensor used in the project is bacteria (Vibrio fischeri). le is an automated pollutants (in particu lar weed-killer and rhe Gymnotox. lr uses the electri cal herbicides) inhibits photosynthesis and th us version of laboratory standardised tests used discharges naturally emitted by Apteronotus emi tted fluorescence. The measurem ents co detect toxic (AFNOR T 90-320, ISO a!bifrons (tropical fish). (F igure 7). The 11348). Under conditions of toxicity, an are taken once per minute and enable the frequency of these discharges are naturally fluoresce nt deviation over time co be inh ibition of luminescence is observed . The stable, bu r vary proportionally co the inhibition is rapid, and is proportional co assessed. seriousness of pollution. The electrical the co ncentration of poll utants. The The AquaTox-Control Daphnia Test information emitted by three fish es placed bacterial reagent is automatically biosensor exploits a combi nation of the in separate ranks is collected once a second regenerated by addition of water and locomotor activity of the daphnia and their through stainless steel electrodes. Jc operates sodium chloride (Figure 3) . Two optical heliocropism behaviour. The biosensor without maintenance fo r at lease one senso rs detect the variations of co mprises two tanks (Figu re 5) each month. (Thomas eta!, 1995) luminescence of a sample of the water co be co ntaining 12 daphnia. Low level luminous As stared, chose biosensors are co upled with tested fo r its quality. Measurement are excitation is produced alternately at two physico-chemical sensors and analysers for made befo re and after exposure. The different places in each tank. The moni toring the basic parameters biodececcor considers chat a poll ution is crustaceans prefer co swim coward the light (temperature, dissolved oxygen, pH, present in the supervised med ium, when it source, and their movements are recorded co nductivity and REDOX pocencial), records an inhibition of luminescence by means of sensors. A fresh measurement parameters for the characterisation of higher than 20%. Maintenance is required is made every 15 minutes. Under aquatic media (chlorophyll , turbidi ty), and for about 30 minutes once a week. cond itions of toxi city, che heliocropism substances associated with potential pollutions (ammonium , nitrate, phosphates, hydrocarbons) (Figure 1).

Figure 5. AquaTox-Control Daphnia Test.

Figure 6. Picture showing magnet on valve of mussel a nd sensor (DreissenaMon itor biosensor).


MARCH 2006 89

Figure 8. Data flow cha rt from se nsors to operators. Figure 7. Gymnotox an d Apteronotus


colour code or more detailed message (Figure 8).

Data merging In Nancy, the BEWS feed the remote control station of the drinking water treatment plant on a minu te by minute basis. T he reliability of acquisition and recording is to be faced as well as interpretation by the monitoring staff. As a general rule, rhe raw data obtained from sensors and other on-line systems is examined case by case. The information so obtained is copious and unclear. The only correlations employed are based on the practical vision, more or less intuitive, of the managers faced with the everyday realities of operation. Ir is cumbersome. Therefore data treatment is essential fo r rhe operators to make rhe fines t d ecision at the righ t moment. For instance, data merging, developed in Nancy (Azzam et al, 2003), is an operation of processing all the information and then purring it together. It is a way of giving the manager an earlier and more complete p icture of the quality of the water from a single type or a limited number of type of data. The results are shown on synopsis either by

Similar systems are establish ed at other river systems (e.g. Elbe, Danube, Rhine) . T he alarms may in for m all the staff of the station, so char rhe situation can be checked, and the authorities can be informed within a short time. It is up to the authorities (government o r water police) to find our where the source of the impact is, to make efforts to stop the spill and warn downstream interested affected parties, such as drinking water u tilities (Gu natilaka and Dieh l, 200 0). Thresholds of biosensors are tested by each manufactu rer but more often by the research group who installed them on sire. Data can now be obtained on the more d angerous pollu tan ts for each biosensor.

Challenging the system Forry situations of acute pollution in accordance with a broad spectrum of scenarios were simulated. T hree types of pollution were imp lemented: • simple variations in the physico chemical parameters (conductivity, turbidity, pH, 0 2 - under and over saturation),

Figure 9. Rate of positive reaction of b iosensors regard ing the number of tests done.


MARCH 2006


• simulations of accidental pollutions with undesirab le or toxic substances (nitrate, phosphate, ammonium, atrazine, isoproturon, chrome VI , pyralen, gas oil), • co mbi nation of the two previous situations. The numbers of tests for each type are summarised in Table 2, and typical reactions of each of the five biosensors in Figu re 9. D etai led examination of the results (Figure 9) show chat the five biosensors reacted differen tly co rhe poll ution. For instance Dreissena monitor responds to a wider range of pollutants than Fluotox (respectively 27 /38 versus 19/39). Nevertheless Fluorox reacts to lower herbicides co ncentrations than D reissena mon itor (data nor shown). Together with literature references (Kuester et al, 2004; G unatilaka, 2000; Borcherding and Volper, 1994) we identified the particularities of the sensors and determined the most suitable for the anticipated risks, as shown in Table 3 .

Conclusion Our work has shown char it is feasible to replace comp lex analysers by an association of simp le sensors, which do not need a highly trained workforce. Germany, the Netherlands, England and now France are working hard on th is theme. Several working groups and workshops are co nducted associating manufacturers, water managers, researchers, governments (e.g. the Working Group of the Federal States on Water Problems in Germany (LAWA, 1998), the group animated by the KIWA in the Netherlands). The worki ng theme is actually the problem of stand ardisation of those biosensors. With the expansion of on-line monitoring, the need for standardisation of com mercially available measuring systems is increasing . For online chemical/physical sensors already an

Table 2. Nature of po llutants tested and number of tests. Ag ricultural pollution Industrial pollution Ecolog ical pollution (sodium) Hydrocarbons Phosphate Ammonia Nitrates Under & over-saturation Chrome VI

Table 3. Proposed recomme nda tions of se nsors based on Na ncy experim ents. 12 3

2 4 2 2 2 2 2



Sensors recommended

Agricultural pollution (atrazine, isproturon, nitrates, phosphates)

Fluotox + (Dreissena or Gymnotox) + conductivity + chlorophyll A

Ecological pollution (sodium)

Fluotox + Aquatox + conductivity


Aquatox + (chlorophyll A or potential Redox)


Gymnotox + Dreissena + Bacteria Test


Dreissena + Gymnotox + Bacteria Test


Gymnotox + pH

Ammon ium

Dreisseno + conductivity

Under or over-saturation

Aquatox + pH

Industrial pollution (chrome VI + sulphuric acid )

Dreissena + (Fluotox or Aquatox) + pH + (conductivity or chlorophyll A)

Basic parameters (conductivity, turbid ity, pH)


Chrome VI

Dreissena + Fluotox + conductivity + pH




Aquatox + Bacteria Test

ISO-Standard has been under consideration and a similar has co be anticipated fo r BEWS . Ir is needed by che users and law enforcin g authoriti es, co co mpare che results of measurements from di ffe rent sources carried out by similar instruments bu t prod uced by diffe rent manu fact urers. In order to rake the best decisions, monitoring and control systems

require reliable info rmation about the supervised processes. However, BEWS diffe r from the chemical/ physical senso rs as they inco rporate a living ani mal surrogate as part and parcel of the instrument hardware. (G unatilaka and Diehl, 2000) A dedicated website fo r on-line biomonicori ng www.ews-platfo rm.org has

been establ ished by the Durch co share experiences, data and achieve chose goals. A CD-ROM is available on dema nd from NANCIE. Ir covers a project descri ption, eco nomic results as well as technical ones, presents derails on thres hold or biosensors techniques and, a methodological guide, whi ch hel p co implement such tech nologies.

6) t:APS Australia RDBUSCHI /II fEBl ~~~~~~tS'.! PfyLtd


CAPS are distributors of Robuschi, Gardner Denver and Kawasaki Positive Displacement and Centrifugal Blowers. Reliable, low maintenance, oil free, low and high pressure air blowers for conveying, floatation cells, water treatment, leach tank aeration, bacterial oxidation and any other industrial use are a specialty of CAPS. Whether you require 50m3/hr (30cfm) or 70,000m3/hr (42,000cfm) at pressures up to 2.2bar (35psig), CAPS will build a custom engineered package to suit.

POWER GENERATORS J!/SDMO' CAPS Australia and SOMO are worl<ing together to provide reliable and quality diesel generating sets and controls from 1'rWA - 3000 'rWA. We offer the best packaged solutions in the power generation field for nom1al standby and more demanding applications, with open type, fitted with acoustic canopy and containerised options. We match the right generating, transfer and control technologies with your power needs. Our most recent example is supply of over 30 generating sets of different sizes for Watercorp in WA for water treatment sewage pump applications.

AIR COMPRESSORS <!iJ) ,ngerso/1 Rand The Ingersoll Rand Reciprocating and Rotary Screw (both fixed and variable speed} ranging from 4-350kw are extremely reliable and durable. We also have a large range of Conquest products, air dryers, filters for a total air system and air receivers to satisfy your needs.



Airman Compressors are available in Box Type, Trailer Mounted and Large Mobile. Made by one of the wands largest manufacturers of diesel compressors, Airman is sure to have a compressor to suit your needs.

www.capsaustralia.com.au sales@capsaust.com.au

f REECALL 1800 800 878

CAPS Australia has been supplylng a wide range of industrial products to water and waste water industry, manufacturing, mining and oil and gas clients throughout Australla for over 25 years. We also have the ability to manufacture any standard package into a customised solution to suit your needs. Full servicing is available from specialised technicians to protect and repair your investment. We have 8 branches Australia wide, contact your closest branch on 1800 800 878.



The project LIFE 99 ENV/F/0 00492 "M ultiparameters survey and protection of water quality" has been financ ially supported by the European Commission through the LIFE tool.

Nancy. NANCIE, 149, rue Gabriel Peri BP290, F-545 15 Vandoeuvre les Nancy Cedex, France, Tel: +33 3 83 15 87 87, Fax: +33 3 83 15 87 99, email: and reolia@nancie.asso.fr, URL: http :/ /www.nancie.asso.fr

Eng. Agnes Andreoli is a project manager and Dr Sylvain Fass is Research Scientist at the Centre International de !'Eau de

Encyclopedia ofNanoscience and Nanotechnology, 375-388 Gresle A. and Bridoux G., 2005, GenPlus®: the online monitoring of drinking water line by DNA, L 'eau f'industrie Les nuisances, 283,

References Azzam A., Robert M. And FASS S., 2005, Supervision system to detect pollution alerr of surface water in mulrisensors sracion, in

Science & Environment.

The Authors

Greenbaum E., Rod riguez M . and Sanders C.A., 2004, Biosensors for dereccion of chemical warfare agents, in Dekker

Boccherding J. and Volper M., 1994, T he Dreissena monitor: first results on the application of chis biological early warning system in the continuous monitoring of water qualicy, Water Science Technology, 29, 199-20 I

45-47 Gunatilaka A. and Diehl P., 2000, A brief review of chem ical and biological con tinuous monitoring of rivers in Europe and Asia, Biomonitors and biomarkers as Indicators of Environmental Change, 2, 9-

28 Gunarilaka A., Diehl P. and Puzicha H., 2000, T he evaluation of dynamic daph nia rest after a decade of use, Biomonitors and

biomarkers ofenvironmental change, 2, 29-58 Kramer J.M. and Foekema E .M., 2000, The M usselm onitor® as biological early warning system, Biomonitors

and biomarkers as Indicators of Environmental Change, 2 , 15 7-174 Kuester E., Dorusch F., Vogt C., Weiss H. and Arenburger R., 2004, Online biomonitors used as a rool for toxicity reduction evaluation of insitu groundwater remediation techniques, Biosensors and bioelectronics, 19, 12, 171 1-1722

Eliminates phosphorus from water bodies and breaks the algal cycle: • • • • • • • • •

Reduces phosphorus concentration Caps sediments Prevents remobilisation of phosphorus Stable at varying pH Stable in anoxic conditions Low in toxicity Environmentally non-hazardous Safe to handle Easy to apply

Suitable for use on many types of eutrophied water bodies: • • • •

STP lagoons Lakes, ponds and water reservoirs Golf course water features Farm dams Recreational water courses

Lawa, 1998 - Recommendations o n the deployment of continuous b iomonitors for the monitoring of surface waters, Kulrurbuchverlag Berlin, 1-46 McCarter S. and Kassinos A.C., 2005, Utilization of naturally occurring biosensors ro det ect toxins in our nation's water supply, Industry and

business news, 2 . Pothus P., Palfner L. and Schmitt J.P ., 2002, A new use of Dreissena polymorpha as a preventive m anagement of q ualitacive deterioration of hydrological resources, IWA, Automonet 2002, 429-432 Tahedl H. and Hoder D. P., 2001, Automated b iomonitoring using real rime movement analysis of

Eugfena gracifis, Ecotoxicofogy and Environmental Safety, 48 , I 6 1-169 T homas M ., Chretien 0., Florion A. and Terver D, 1995, Real time detect ion of potassium cyanide pollution in surface waters using electric organ discharges wave emitted by the tropical fish

Apteronotus Albifi·ons, Environmental Technology, 17, 561574

Web site SAFER project: http://www.safereu. com EWS Web site: www.ews-platform.org

92 MARCH 2006 water

EARLY WARNING MONITORING TECHNOLOGIES N O'Connor, A Davison, D Deere, D Baker Abstract Contamination of raw drinking water sources may result from accidental, natural, or intentional (i ncluding terrorist) events. Such events pose th reats to raw waters in water supply catchments, storages and in some bulk water transfer infrastructu re compo nents such as aqueducts. Standard on-l ine water quality sensors (e.g. turbidity, pH, conductivity) play an important role in detecting water quality changes in raw water inflows to water treatment plants. However these devices cannot detect many types of co ntam inants such as fuel, oil, surfacta nts, solvents, biocides, algal toxins and ocher chemicals nor can they detect pathogens. The public health threat posed by contamination events is significant. Given these challenges, a number of water authorities around the world have invested in biologically based early warning monitoring systems. Biological response sensors operate by using an organism or groups of organisms as sentinels to detect changes in water quality (similar to the use of canaries in coalmines in the past). The systems operate by passi ng a continuous scream of the water to be monitored past the test organism and monitoring the behaviour of the organism in real time. The sophistication of these on-line toxicity meters has been facilitated by the development of advanced electro nics and fast computers. While there are many technologies in development, only a handful of devices are currently in commercial operation. In th is paper we review 6 devices. The devices are manufactured in Europe and che USA and involve the use of either fish, Daphnia (a species of zooplankton), algae (phytoplankton) and mussels. All the devices are currently in operation overseas and are commercially available in Australia. A guide to the capital, sec up, operation and maintenance coses for each system is also presented.

Keywords: Biosensors, online monitoring, early warning, risk management, due diligence

Introduction Most raw drinking water sources are susceptible to a wide variety of potential

contaminants due to accidental, natural, or intentional (i ncluding terrorist) events. Such events may pose threats to raw waters across entire water supply catchments, in bulk water storages and in transfer infrastructure components such as aqueducts. In some instances storages and transfer aqueducts containing partially treated or fu lly created water may also be at risk of contamination depending on the level of exposure of rhe system in question and the prevailing securi ty environment. For most raw water systems there is a need to ensu re a rapid response ro spills an d other sudden contamination events to protect consumers, treatment plant functio n, bulk water quality, and to comply with environmental and health regul ations (G ullick et al. 2003). Early warn ing moni toring systems can provide a costeffective solu tion to ensu re that contami nation events are rapidly detected and char timely and effective response measures are implemented. Early warning monitoring systems ca n also boost public confidence in water su pply qu al ity, provide valuable feed back on many low level contamination events, encourage good practice, and in the case of intentio nal or terro rist attack may actually deter potential perpetrators from even trying due to che likelihood of early detection and red uced im pacts.

A guide to the capital, set up, operation and maintenance costs. What are Early Warning Monitoring Systems? In chis section we describe the types of devices char can be used in an early warning monitoring system. The system may consist of a single device or a number of devices operating together, usually in the same facility. In the context of protection of raw drinking waters, we define an early warning monitoring system as a functional system for rapidly detecting a change in water quality that is or could be due to contamination. Furthermore early warning monitoring systems "must reliably identify low

probability/high impact contam ination events (chemica l, microbial, radioactive) in source water or distribution systems in time to allow an effective local respon se" (ILSI 1999) . Most water treatment plants make use of a num ber of on-line instantaneous devices for monitoring raw water quality. Common devices include meters for turbidity, conducriviry, dissolved oxygen, and pH but there is also an increasi ng array of ocher devices capable of monitoring p arameters such as ammo nia, alkalini ty, ch loride, copper and many ocher analyres. T hese devices can provide useful information to assist process control, however for reasons of cost or uti lity on ly a limited range of devices is commonly employed (e.g. turbidity, pH, conductivity). Co nsequen tl y on ly a limited range of parameters can be detected at any moment in rime. Furthermore, most existing on-line monitori ng devices measure standard water quality parameters and are nor d esigned to detect the wide array of toxica ncs char may potentially be present. Any device char can detect or measure a wide range of con taminants may be su itable for use as part of an early warning monitoring system. Early warning devices may be chemically, physically or biologically based devices that simply raise an alarm when some uniden tified con taminant is detected in rhe water, or actually identify and measure the concentration of an array of contaminants. In general biologically based devices are used as simple alarms, whereas ch emically and physically based devices provide some clues to the identity and concentration of the contaminant as well as performing an alarm function. In practice early warning monitoring systems usually consist of early warning devices which measu re a continuous flow of the raw water stream to be monitored. The devices may be housed in an existing water utility building or in a small purpose-built housing close to the water supply system being monitored. The types of early warni ng devices chat are currently commercially available in Australia are described later in chis article.


MARCH 2006


Why Do We Need Early Warning Monitoring Systems? T here is an infinite array of contaminants char could potentially occur in the raw waters (Table I). In some cases due co rhe nature of the event, contaminant concentrations may be highly correlated with a readily monitored parameter such as turbidity. For example Cryptosporidium concentrations may be associated with high turbidities in storm flows in which case the water authority may be able to quickly implement a control measure such as a temporary cessation of water extraction. In many cases however, che contaminant concentration may not be associated with any readily moni tored para merer. For example, an insecticide resulting from an industrial spill may be undetectable in the absence of a broad-spectrum early warn ing dececcion device such as a Daphnia biomonitor (described in the next section). Terrorist type attacks

Accord ing to a recent Australian Security Intelligence Organisation (ASIO) Threat Assessment, terrorise groups worldwide have ready access to information on chem ical and biological, and to some extent, nuclear weapons, via che Internet, publicly available scientific licerature, and scientific conferences. Al-Qa'ida has considered and justified che use of chemical, biological and radiological weapons and has some larenr capabili ty to build and use such weapons (ASIO Threat Assessmenc ciced in Queensland Governm ent 2003).

Types of Early Warning Devices An early warning mo ni toring system muse reliably idencify low probability/high impact concamination events (chemical, microbial, radioactive) in source water or disrribucion systems in rime to allow an effective local response (ILSI 1999). Calles et al. (2005) suggest chat early warning systems should have the fo llowing characteristics: (i) Rapid response time (ii) Fully automated (iii) Screens for a range of concaminanrs (iv) Specific for the contaminants of concern (v) Sufficient sensitivity (vi) Low occurrence of false positives and false negatives (vii) High rare of sampling (viii) Rel iable and rugged (ix) Requires minimal skill and training (x) Affordable cost It is debatable whether any existing early warning technologies possess all these


MARCH 2006


Table 1. Classes of contam inants that may potentially occur in raw w aters. The classes listed are not mutually exclusive . Contaminant Class

Likely sources

Algal toxi ns

Toxic algal blooms in storages due to excessive nutrient loads

Fuels and oils

Passenger vehicles, transport accidents, dumping, contaminated groundwater

Solvents and other Hydroca rbons

Transport accidents, dumping, contaminated groundwater


Transport accidents, dumping, contaminated groundwater


Transport accidents, dumping


Transport accidents, dumping, surface runoff


Transport accidents, dumping, surface runoff, spray drift, contaminated groundwater

Metal solutions, industrial liquors and other inorganic materia ls

Transported accidents, dumping

Organic factory wastes, fa ts, grease, etc

Transport accidents, dumping


Transport accidents, dumping, runoff from adjacent land applications

Sewage (untreated)

Damaged sewers or failing septic systems

Treated sewage effluent

Licensed d ischarges of treated sewage effluent


Transport accidents, runoff from fertiliser applications on adjacent land, discharge from intensive animal industries

Radioactive wastes

Transport accidents, dumping of hospital w astes

Damaged sewers or failing septic systems, di scharge from intensive Microbial pathogens including viruses, bacteria, animal industries, licensed sewage discharges and other licensed helminths, protozoa, waste discharges, transport accidents, dumping fungal spores, prions, and bacterial endotoxins Other toxins

This category refers to toxins used in possible terrorist attacks including ricin, botulinum, sarin, and other inorganic toxicants such as cyanide etc

characteristics. In particular items iii, iv, v, vi, and x relating to the breadth of the contami nant detection spectrum, specificity, sensitivity, reliability of detection and cost have proven the most challenging for system designers. Fo r such reasons some existing early warning monitoring systems consist of a number of early warning devices operated together to provide a more effective warning system. Currently there are is a wide variety of early warning technologies. Many of these technologies are still in the developmental stage and are not yet commercially available nor have they undergone sufficient resting to support their use. Ocher technologies do

nor offer truly continuous measurement but instead rely on analysis of sample batches taken at regular intervals of a few minutes to several hours or more. Furthermore so me devices are really only appropriate for use as part of an array of devices covering different concaminants. These devices fall somewhere between single parameter on-li ne devices (e.g. pH meter) and true broad-spectrum early warning devices. For reasons of space we do not discuss these devices in this review, however, Gullick et al. (2003) provides a useful discussion of such devices. Instead, in chis survey of early warning monitoring systems we have focussed on commercially available devices that continuously monitor water quality. T hese

Table 2 . Distributor and cost details for the Mussel mo nitor®. Distributor:

Cost 12004 prices)

There is no local distributor for this unit. The distributor is: Delta Consult, PO Box 71, 4420 AC Kapelle, the Netherlands Web: www.mosselmonitor.nl/

MOSSELMONITOR®, submersible unit or la boratory version incl. cable reel and power supply: €17,745 Personal computer w ith mon itor: € 1,250 Present IT presentation a nd communication software: €1,250 Instruction in a two-day operation, including Ion site employment (in situ or in a flow-through system); demonstration of the functioning, with real time data collection; and demonstration of data treatment options): €2,325

devi ces can be placed into one of rhe following rwo categories: • Biological response sensors; and • Physico-chemical analysers. These senso rs are designed to detect most chemical conraminanrs in the raw water rather rhan microb ial pathogens. We do nor discuss early warning systems fo r microbial pathogens in rhis paper as ar present, co rhe best our knowledge, such systems have yet ro be developed to rhe level of sophistication and rel iabili ty as rhe sensors described below, although rhe technology is developing ac a rapid rate. A further key poi nt in relation co Figure 1. MosselMonitor®. microbial pathogens is chat water treatment planes are designed to deal with a specific level of mi crobial (e.g. mussels), fis h, zooplankton (e.g. water contaminarion so char the greate r threat to fleas), and phytoplankton (i.e. algae). Four public safety is fro m chemical spills or devices are made by bbe Moldaenke GmBH chemical weapons in terror attacks rather of Germany (the bbe "Taximeters"); one rhan microbial co ntamination. device is made by Del ea Consul t B. V. of Biological Response Sensors T he Netherlands (rhe Mosselmonitor®) and Biological response sensors operate by using one device by Biological Mo nitoring an organism or groups of organisms as Incorporated of Virginia, USA (B iosen tinels to detect changes in water quality. Senso r® Series 7000). Further details of This is similar to the use of cana ries in 19th rhese systems are presented below. Century British coal mines to derecr rhe Mosselmon itor® presence of coal gas befo re it could affect miners. T he systems operate by passing a The Mosselmonitor® (Durch for mussel continuous scream of the water co be moni tor) is made by Delra Co nsult B.V. of monitored past rhe test organism and T he Netherlands. Information on the monitoring the behaviour of rhe orga nism device was obtained from the company's in real time. These systems do not prov ide website and prices and other informat ion information on rhe nature of che water have been supplied by the company (see also Table 2, Figures l and 2. quality change. The organism's respo nses could be due to the presence of The monitor makes use of 8 bivalves (e.g. co ntaminants in the form of disso lved mussels, clams, oysters ere), usually a chemicals, dispersed chemicals (i. e. chemicals chat are miscible in water but not truly soluble) or floating particulate matter or due to changes in physical conditions such as temperature or turbidity. Changes in behaviour can be used to trigger alarms and automated warer samplers which can cake a 2X~2 mussels sample for lacer detailed chemical analysis. Biological response sensors are usually partnered with an array of online physico-chemical analysers such as pH, turbidity, electrical conductivity and dissolved oxygen probes. These additional devices assist in diagnosing the causes of warer quality changes detected by the biological Glut sensors. The types of sensors available include devices based on bivalves

Figure 2. Schematic d iagram of a mussel biological mo nitoring device (reproduced from ILSI 1999).

species of mussel. The device uses rwo electromagnetic sensors to fo llow rhe movement of the shells of each mussel. In normal feeding mode, ch e mussel shells will gape and only close fo r brief periods to excrete. Abnormal behaviour, such as a long p eriod of shell closure, is detected by che monitor. For example, if all mussels are closed for a longer period (e.g. > 4 min), an alarm is generated. Alarms can also be used to induce a uto mated sampling for fu rther proof by chemical analysis. T he manufactu rer scares char the device can be operated unattended fo r weeks 24 hours a day, 7 days a week. Sensitivity of the test organ isms can be controlled and bivalves are replaced every 2-3 mon ths. A co mpanion algal culrure device is used co grow and supply planktonic uni cell ular algae as food for rhe mussels. Dara from the mussel monitor is interpreted in an accompanying software package on a PC. T he PC also a llows remote access so ch at the user can mon itor one or more systems from his or her office. T he device has been tested in Europe usino0 Zebra mussels (Dreissena pofymorpha), a native to Asia which is a serious ecological pest in North America, and two ocher freshwater species. T he Commonwealch Department of Enviro nment and Heritage oversees impo rt appl ications for wildlife and is unlikely to allow importation of rhese species to Australia. For freshwater applications in Australia, a suitable test organism needs to be identified and readily obtainable from the wild , an aquarium supply co mpany, or a University laboratory. Likely Australian species are from the Biva lve family Hyriidae and incl ude Velesunio ambiguus a billabong species and Afythira jacksoni a riverine species, both fo und in the Murray River. Observations by researchers at the Un iversity of Adelaide suggest chat Afythira, which prefers flowing water h abitats may be more sensi tive to dissolved oxygen levels than \\'attr In Ve!esunio.

Bio-Sensor® Series 7000 This device is manufactured by Biological Monitoring Incorporated (BMI) of Virginia, USA (see Figure 3) . Dr David Gruber of BMI provided the fo llowing information and a costi ng for a 7000 series BioSensor®.


MARCH 2006


Table 3. Distributor and cost details for the Bio-Sensor® Series 7000. Distributor:

Cost (2004 pricesl

There is no locol distributor for this product; details for the US distributor are: l 800 Kratt Drive, Suite l 0 l, Blacksburg, VA 24060, USA, Ph: 0011 l (5401953-2821 , Fax: 0011 l (5401951-1481, Web: www.biomon.com

The cost per unit: Approximately US$40,000 (discounts may apply for orders of multiple unitsl. Likely freight costs: Approximately US$ l ,500 per unit

as a series of bar graphs. Individual fis h are automatically and continuously assessed against themselves for statistically sign ificant changes in their behaviour utilising a time series or moving average statistical algorithm. W hen an individ ual fish has been assessed as behaving abnormally, a warning is assessed for char fis h .

Figure 3. Bio-Sensor® Series 7000 from Biologica l Monitoring Incorporated. According co BMI's promotional material, all living organisms generate a bioelecrric field about themselves, the result of their normal neuromuscular activities. In the Bio-Sensor®, rhe electrical fi eld generated by an individual fish is sensed within its monitoring chamber wirh two nonintrusive stain less steel wire mesh electrodes. These bioelecrric signals are amplified, filtered and interfaced co an analog-ro-digiral converter via rhe BioAmp® component of the Bio-Sensor®. These data are subsequently transferred co rhe computer's random access memory (RAM) and processed in real time. Data are also scored on ch e computer's fi xed disk and can be printed out for subsequent environmental auditing purposes. The parameters utilised co define fish behaviour are vencilatory (breaching) frequency and amplicude, and locomotor acriviries. These variables are constantly displayed on the monitor screen during monitoring, either as a representation of an individual fish 's behavioural waveform, or

When che computer system assesses a significant number of fish as behaving abnormally, an alarm routi ne is initiated. Upon rhe generation of an alarm, the mon itor screen changes colour, rhe com purer sounds a cone, and a water sample is automatically collected for add itional chem ical analysis as needed. Various alarm condition status signal outputs are provided for remote communication purposes (e.g. Aucod ialers, SCADA systems, ere) . Additional custom alarm signal output options may be provided. Typically, conductivity, d issolved oxygen, p H and temperature are measured simultaneously with rhe fish Bio-Sensor® output and are automatically and continuously assessed . If these parameters drift outside an acceptable range an alarm is also acrivared. T he actual values, as read from rhe senso rs by the com purer, are displayed on the "General Srarus" monitor screen, the printout, and a data file on the computer's fi xed disk.

For rhe m easurement water samples are continually pumped through the m easuring cell where rhe concentration of algae and their activity are determined fluoromerrically. A fluorescence parameter (the Geney Parameter) is determined and subsequen tly a defined amount of algae is injected into the cell and the measurement repeated. As long as there are no toxic substances present the activity of rhe additional algae stays constant. Windowsbased signal recognition software is used co analyse fluorescence changes and records the results over rime. If the algae are affected (e.g. say due to a herbicide) an alarm is induced.

Daphnia Toximeter

Daphnia are small free-swimming crustaceans that feed as they swim by filtering algal cells from rhe water. They are common in lakes and storages where they often make up a large proportion of the zooplankron. In rhe Dap hnia Toximerer Daphnia swimming and feeding behaviour is continuously analysed using a video camera and analysis of rhe d igital image with signal recognition software (Figures 5 and 6). Toxicity computations an d assessments are based on the measurements of rhe following surrogate behavioural parameters: • Speed observation • Swimming behaviour (alrirude, rums and circling movements)

bbe Toximeters

• Growth of observation

bbe Moldaenke GmBH of Germany manufacture four related devices they have termed "Toximeters". These are che:

• Number of Daphnia Any change in the behaviour of rhe Daphnia is exam ined and analysed and a

bbe Algae Toximerer bbe Daphnia Taximeter bbe Fish Toximerer bbe Daphnia and Fish Toximecer

Algae Toximeter

Figure 4. bbe Algae Toximeter.

96 MARCH 2006


The Algae Toximerer (Figure 4 ) continuously determines rhe toxicity of an inflow scream by determining rhe chlorophyll fluorescence of algae dosed into the inflow stream. Any toxins in the inflow water may incerfere with photosynthesis and subsequently affect chlorophyll fl uorescence. T h e test organisms - the algae - are cultivated in an integrated fermencer and the water sample is mixed with rhe standardised algae.

Figure 5. bbe Daphnia Toximeter.

parameter called the "toxicity" index is continuously calculated. Presence of tox ins is ind icated by significa nt changes in the above behaviours and che sensitivity of the alarm can be adjusted by the user co suit particular circumstances.

• SAC254, SAC280, SAC330 (where SAC254 = Spectral Absorbance at 254 nm) • BOC (bioaccessible organic carbo n); and


Daphnia is a cosmopolitan genus and the most appropriate species fo r use in Australian applications is Daphnia carinata. T his species is used in university ecocox icology laboratories in Australia and starter culcures are available from these sources or specialist aquarium suppliers.

Traces oflinpul5es

Fish Toximeter Figure 6. T he Fish Toximerer (Figure 7) works in an identical fas hi on co the Daphnia T ox imerer however it is based on the behaviour of the test organism Zebra fish (Danio rerio). Zebra fish is a small tro pical fis h from the famil y Cyp rinidae (i. e. the Carp and Goldfish fa mily) and hails from Asia. Zebra fish are available fro m aquarium shops in Australia and thus obtaining rest organisms should not be coo di ffic ult in Australia. Daphnia and Fish Toximeter T his Daphnia and Fish Toximeter incorporates the rwo simul taneous onlin e rest systems described above. According co ch e manufactu rer, bbe Moldaenke, ch is sec up improves the quali ty of the analys is statistically and also increases the sensicivity of che device due co different sensitivities co various toxins of che rest organisms.

Advanced physio-chemical sensors Spectrophotometers As noted above, a range of physicochemical sensors or probes is available fo r standard water quality parameters. These devices will continue co be of importance fo r online water quality monitoring because of their low cost, rel iability, and the significance co overall water quality

Figure 7. bbe Fish Taximeter.

Test Chamber

Schematic diagram of Daphnia toximeter.

management. However, as di scussed , they cannot rest fo r a broad spectru m of co ntaminants. O ne new device which can dececc a broad range of contaminants is a small portable speccrophocomecer manu faccured by DCM Process Co ntrol of Vienna, Austria. T he "S: :can" submersib le speccrophoco mecer is a device chat can be used in field and outdoor analysis. l e operates in the ultraviolet and visible light ra nge (UV/VIS) and provides sensitive determinations of organic ca rbo n compounds, curbidi cy and ni trate in environmental waters. T here are no moving pares so the device can be deployed fo r long periods. Orga nic compounds char ca n be determ ined using rhe speccrophocomecer include: • Dissolved organic carbon (DOC) • T otal organic carbon (TOC) • BOD

• Various single substances incl uding aromatics, phenolics, and hydrocarbons rha c meet particular criteria. T h ese substances can be dete rmined after calibration using companion software to analyse th e UV/VIS abso rption spectru m.

The speccrophoromere r shows some promise although ics sensitivity ro a wide ra nge of coxican rs and con ram i nancs is unknown. T he device will requi re a thorough resting and development regime before it can provide the same degree of co nfid ence as a biological monitori ng device. According co rhe Australian Distributor, the d evice ca n be set up co co mpare spectra against prior moni to ring background levels (e.g. rh e previous 4 hours) so that sudden changes in rhe spectrum due, for example, co a plume of co ntaminants can readily be d etected. T he spectrophotometer ca n be d eployed in a small fi eld housing and can ru n from a battery pack.

Cost Summary A summary of th e senso rs is prov ided below Table 6. Prices were obtained from suppliers in late 2004 and are provided here as an indication only. The distributor should be contacted fo r current p ricing. In general , the fish- and mussel-based biomonico rs rend co be lower in price than the Daphnia- and algae-based biomonicors, while the spectrophotomecer was the lowest cost op tion overall.

Table 4. Distribu tor and cost details for the bbe Taximeters. Distributor:

Cost (2004 prices)

The local distributor for these devices is: Technical & Scientific Equipment Co P/ L, 2/ 5 Arisloc Road, Glen Waverley, VIC 3150 Ph: +61 3 956 1 2030 Fax: +61 3 9561 2040 web: www.techsci.com.au

Algae Taximeter: A$71 ,977.00 Reference measurement for algae loximeter: A$1 ,309.00 Algae Fermenter and Culture Reservoir for Algae Taximeter A$693 .75 Daphnia toximeler: Item No. BG 10000-E (1 Channel), 220 Volt A$66, 175 Daphnia loximeler: Item No. BG 11000-E (2 Channel, 2 Chamber System), 220 volt A$80,032 Accessories Daphnia Toximeter: Item No. BG 12000-E woter conditioner/ healer, A$4,509 .60 Item No. BG 15000-E water conditioner/ healer/ filter, A$6 ,466 Fish Taximeter: Item No. BG 16000-E, A$32,061 Train ing up lo: $18,000 Freight per unit up to: A$2 ,200 Freight per unit up to: A$2 ,200 Daphnia and Fish Taximeter: Item No. BG 17000.E, 220 Volt, A$80,032 Training up to: A$18,000 per unit Freight per unit up lo: A$2,200


MARCH 2006 97


Table 5. Distributor and cost detai ls for the DCM Spectrophotometers.

In our review, we have shown that a range of sophisticated and pragmatic biological monitoring devices is now available. These devices appear to be affordable and will provide an extra level of risk management (and therefore due diligence) to water utilities, especially in the raw water area, a field that has traditionally been less amenable to rea l-rime monitoring compared to the treatment and disrribucion water supply chain components. Moreover, the devices appear to be able to integrate with existing utili ty systems such as SCADA and potentially HACC P, enabling ucilicies to facilitate and augment the monitoring of hazards and hence, the efficacy of control measures, within catchments and storages. The decision to employ an early warning monitoring system should be based on a thorough assessment of the risks to water quality. The risk assessment should also assist in identifying which type of device amongst chose available may be best suited to the water utility's needs. Following on from th is review, the next logical step wi ll be to examine these (and other devices char are close to the commercialisation stage) on a comparative basis and determi ne their in-field value to utilities. The Centre Internacional de l'eau de Nancy, (NANCIE) in north-eastern France is undercaking such research (see the previous paper in chis issue by And reoli and


Cost (2004 prices)

Australian Distributor is: DCM Process Control Australia Phone 07 5444 8469, Fax 07 5444 7812 PO Box 1507, Mooloolaba, QLD 4557, Email: www. dcm processcontrol .com

DCM S::Can A$44,000 includi ng ComStat {can also link ta SCADA) Accessories: Corrosion proofing $2000 Training - Included in equipment supply cost above Freight per unit - Available from Australian Supplier, estimated freight A$200 The device incorporates its own Windows NT-based computer in its associated ComStat pack and therefore no additional PC hardware is required

Fass) and chis information provides a useful resource for Australian water authorities interested in early wa rning monitori ng systems.

The Authors Dr Nick O'Connor is a Principal Consultant and Director of Ecos Environm ental Consulting P/L, specialising in human health and ecological risk assessment for the water and environment industries, Level 1, 339 Whi tehorse Road, Balwyn, Vic 3103, ph (03) 9830 7740, email: naocon nor@ecosec.com.au; Dr Annette Davison and Dr Daniel Deere are the Principals and Directors of Water Futures P/L and specialise in public health risk assess men c and management and interpretation of environmental law for the water industries, 32 Sirius Sc, Dundas Valley, NSW 211 7, ph (02) 9871 1353, email: annene@wacerfurures.nec.au; Ms Danielle Baker is an environmental

scientist wich Ecowise Environmental P/L an environ mental management co mpany which specialises in providi ng best practice solutions to environmental issues, 24a Lemko Place, Penrith NSW 2750, dbaker@ecowise.com.au, ph: (02) 4721 3477.

References Calles, J, Gorder, R.,

Eva ns, M., and Syage, J. 2005. Early warning surveillance of drinking warer by phoroionizacion/mass spect romet ry. journal ofthe American Water Works Association. 97( I ):62-73. Gullick, RW, Grayman, WM., Deininger, RA, and Males, RM . 2003. Design of early warning monitoring systems for source ware rs. journal of the American Water Works Association. 95(11):58-72 !LST I 999 . Early Warn ing Mon itoring To Detect Hazardous Events in Water Supplies . An !LSI Risk Science Instit ute Workshop Report. Thomas M. Brosnan, Editor. Washington DC, USA. D ecember I 999 .

Table 6. Indicative costs of biomonitoring options (Australian dollars, GST exclusive). Costs exclude cost of space renta l for equipment and staff and capital cost of construction of housing facility and purchase of land if necessary. Prices were obtained from suppliers in late 2004 and are provided here as an indication only. The distributor should be contacted for current pricing . Options and Cost


1. Mossel monitor

Personal computer with monitor Present IT presentation and communication software Automated Food Device Item No. BG 15000-E water conditioner/heater /filter Device Reference measurement for algae toximeter Algae Fermenter and Culture Reservoir for Algae Toximeter Training Freight

2. Bio-Sensor Series 7000

3. bbe Algae Taximeter

4. bbe Daphnia Taximeter

5. bbe Fish Taximeter

& Fish Taximeter

6. bbe Daphnia

7. DCM Scan Spectro-photomer


$71 ,977

$6,466 $80,032

$32,06 1



$18,000 $2,200

$18,000 $2,200

$18,000 $2,200

$3000 $200

$2, 155 $2,155 $16,983


$1 ,309

$4,009 $517


$694 $18,000 $2,200

Total AUD$








Indicative annual maintenance cost including labour Rough estimate of one¡off setup costs


$22,340 $5,000

$25,460 $5,000

$19,220 $5,000

$22,340 $5,000


$12,980 $5,000



Mosselmonitor price was received in Euros and converted to AUD$ BM/ price was received in US$ and converted to AUD$, 1 and 2 channel devices were costed, the 2 channel device is included in this price.

98 MARCH 2006


RO TECHNOLOGY: UCLA's CONTRIBUTIONS TO RESEARCH AND EDUCATION EM V Hoek Abstract T remendous improvements in reverse osmosis (RO) technology have been made in the last 50 years. Reverse osmosis processes are now among the most important and versatile water treatment technologies - enabl ing desalination and water recycling as viable components of water resource management plans around the world. Some of the most important innovations and remaining limitations for RO technology are discussed, along with related research and development activities at UCLA, daring from some fifty years ago to the present.

Introduction Explod ing population, degrading local water quali ty, mounting political pressures, and recurring draughts have long forced Sou thern Californian's to seek more sustainable water supply alternatives by investing in new technology development and implementing novel water management practices. Desalination and water recycli ng by reverse osmosis membrane technology have been consistent themes in Southern California water management for over fifty years. Also for over fifty years, experts in the School of Engineering and Applied Science at UCLA have engaged in RO technology resea rch and educated enu merable young scientists and engineers. In chis review article, I touch upon a few poignant even rs in the early history of RO development at UCLA and mention some of rhe key technological innovations char occurred beyond UCLA. Currem education, research, and development efforts at the UCLA Water Technology Research Center are also described.

Early History of RO Technology Development A comprehens ive review of the early history of RO membrane development at UCLA was published previously by Glarer [l ]. I This paper is loosely based on a presentation by Prof. Hoek to RM !T University when he was a Visiting Fellow in 2005.

Figure 1. Water drops being col lected from an earlier tubular RO membrane developed at UCLA, circa 1965. Photo provided by Bud Glater at UCLA.

Fifty years of development of membrane processes... still more to come. summarise a few of the key events herein. In 1945, spurred on by rapid population growth rhe UCLA Department of Engineering was established with three broad research missions: rransporrarion, air pollution, and water. In 1948, Prof. Joseph McCurchan submitted a series of research proposals ro the Office of Sali ne Water a new agency established in rhe U.S. Department of Interior. About rhe same rime, osmotic processes were being studied by Or. Gerald H assler, a member of the research staff, using cellophane membranes. Ir is rhoughr char the term "reverse osmosis" was first coined by Dr. Hassler in an imernal UCLA project report [2] dated August 1956 [3]. During this period, the UCLA Sali ne Water Conversion Group had several projects underway involving reverse osmosis and imerphase separation ("surface skimming") studies.

Beginning about 1955 Professor Charles Reid, at rhe University of Florida, bega n a study emirled, "Osmotic Mem branes for Demi neralisation of Saline Water," which was also supported by The Office of Saline Water. Several publications arising from this work provided experimental details on the performance of various commercial polymeric films available in the 1950s (I]. Techniques were also developed for producing cellulose acetate membranes with good salr rejection, but the poor water fl ux led Reid and co-workers ro give up on the concept of pressure-driven reverse osmosis using commercial membranes. Apparently, the two research groups at the University of Florida and UCLA were compl etely unaware of each other's contributions until 1957, when the United Stares National Academy of Science-National Research Council sponsored a Saline Water Symposium. Professor Reid made a presentation and this conference was attended by Gerald Hassler and Joseph McCurchan.


MARCH 2006


In 1958, reverse osmosis membrane synthesis research began at UCLA under the direction of Prof McCutchan . Early experiments focused on modifying commercial cellulose acetate UF mem branes by heat treatment, bur lacer Loeb and So urirajan decided to hand-case membranes from cellulose acetate polymers. They experimented with various levels of acetylacion , chain lengths, and casting solvents lead ing co the development of the first practical cellulose acetate RO membrane in 1960 (4]. The "LoebSourirajan" membrane was the first majo r technological breakthrough m arking the beginning of modern reverse osmosis technology. Research in membrane casting solution optimisation continued at UCLA leading to the development of a reproducible membrane fabrication technique involving an all organic solution comprised of cellulose acetate, acetone, and formamide [5] . Additional researchers and faculty from the School of Engineering and Applied Science involved in early desalination research included Professor Samuel Yuster, Edward Selover, Serop Majikian, James S. Johnson, F. Milstein, Julius "Bud" Clater, and Mary Justice.

Figure 2. Sid Loeb (UCLA), Joe McCutchan (UCLA), unknown operator and Harry Baldwin (City of Coalinga) standing i n front of the first commercial RO plant cal led "Raintree" in Coalinga Californ ia, 1965. Photo provided by Bud Glater at UCLA.


With our presence in more than 100 countries, CH2M HILL is a globa l leader offering fullservice engineering, consulting, construction and operations management to public and private clients worldwide across numerous industries.

Outperforn1ance Sydney Level 7, 9 Help Street Chatswood NSW 2067 Ph: 02 9950 0200 Fax: 02 9950 0600

In Austra lia, our services are focused on the areas of water, wastewater, recyc ling, environmental, energy and industrial sectors. We craft our solutions with sustainability and integrate business, science, engineering, policy and community relations.


1 '


Our services help our clients protect the environment and realise a greater r:eturn on their investments and we will continue to create innovative solutions wherever our projects take us.

every project, every client, every time. That is our goal.

Melbourne Level 3, 5 Queens Road Melbourne VIC 3004 Ph: 03 9856 4800 Fax: 03 9856 4844

Brisbane Level 1, 33 Park Road Milton QLD 4064 Ph: 07 3302 8700 Fax: 07 3302 8799

Perth PO Box 68 North Beach WA 6920 Ph: 08 9246 7120 Fax: 08 9246 7104

Modern RO Technology Innovations Discovery of the first practical RO membrane at UCLA was fo llowed by novel innovations in hardware and pilot plant design. The most significant innovation involved development of a tubular RO membrane casting tech nique. The tubular cellulose acetate membranes of Loeb, Sourirajan, and McCurchan were wrapped with a non-woven polyester fabric and inserted into perforated steel pressure pipes (see Figure 1). Development of tubular RO membrane technology led to construction of a 5,000 GPO system, dubbed "Raintree" for production of drinking water from brackish water for rhe 5,000 residents of Coalinga, California (see Figure 2). This was the world's fi rst commercial RO plant, beginning operation in 1965 and garnering attention from laboratories and govern ment offices around the world. In lacer years, the UCLA "Pilot Plant Group" built RO plants for seawater desalination in La Jolla, California and for agricultural water reclamation in Firebaugh, California. A comprehensive review of early pilot plant stud ies at UCLA is presented by Glacer (6). In the fo llowing decades, many limi tations of early RO technology were overcome through systematic effons by a host of international researchers and engineers from both academia and industry. An entire issue of Desalination (vol. 32, 1980) was dedicated to developments char occurred over the firs t decade of commercial reverse osmosis desalting, 1968-1978. Key innovations described in rhe special issue of Desalination include: polyamide chin film composite RO membranes; hollow fibre and spiral wound membrane modules; dynamic membranes; electro-dialysis; understanding and control of fou ling and scaling; research on RO prerrearmenr processes; energy recovery turbines and solar-assisted RO; as well as operational experience gained from desalination and water reuse applications. Lonsdale [7] also published a broad review of membrane technology development up to 1982 - incl uding reverse osmosis - and rhe reader is referred to chis paper for more derail. Further optimisation of RO membranes, process equipment, and operating strategies since rhe early 1980s has led to a worldwide shift away fro m thermal desalination processes in favour of reverse osmosis processes. Three of rhe most significant technological innovations responsible for chis shift are briefly described here. Improved membrane materials

Since rhe Loeb Sourirajan membrane, rhe next major breakthrough in reverse osmosis

membrane materials came in the 1970s and 1980s, when Cadotte and cowo rkers at FilmTec Corporation developed thin film composite RO (8, 9] and later N F (10) membranes by interfacialiy polymerising aromatic polyamide films on porous supports. T he key advantage of chin film composite membranes was char each polymer layer could be independently optimised to achieve rhe best water permeability, solute selectivity, and stability (chemical, mechanical, and thermal). T hin film composite RO membranes exh ibited much greater fluxes due to rhe very thin (50 to 250 nm) active layer, which could nor be achieved with cellulose acetate. The excellent performance of polyamide composite membranes has enabled widespread appl ications ranging from seawater desal ination, water softening, brackish water reclamation, and wastewater recycli ng to separations in the sem iconductor, pharmaceutical, mining, food, and beverage ind ustries [7] . However, a major co ncern with polyamide TFC membranes in ail applications is their propensity to foul and their poor tolerance to oxidising biocides, especially chlorine. Fouling mitigation and chlorine tolerance remai n major elements of modern RO membrane materials research in addition to continually improving selectivity and permeability. Optimised membrane modules

Design of rhe first multi-leaf spiral wound mod ule by D.T. Bray at Universal Oil Products, Inc. in 1968 sta nds our amo ng che most important modern developments (1 1). W hile hollow fibre membranes dominated early applications, advantages of sp iral wound modules have led to lowpressure, high recovery RO applications with relatively small "footprint." Many improvements in spiral wound mod ule design have come from theoretical studies of fl uid flow, pressure drop, and mass transfer in RO modul es, as well as from extensive practical resting (12, 13]. These improvements have reduced concentration polarisation, scaling, and fou ling concerns and, more recently, led to development and resting of large diameter RO modules (14). Followi ng several years of development and resting in coliaboracion wich che Metropolitan Water Discricc of Southern Califo rnia (MWDSC), Koch Membranes in San Diego, California now sells an opti mised 18 inch diameter module. Early test results ac MWDSC suggest a significant cost savings may be realised from large diameter RO modules predominantly due to smaller system footprint and fewer mechanical connections.

A CDS Group Company

• Stormwater treatment • Filtration and reuse • Flow regulation

• Sewage treatment plants • Sewer overflow screening • Flow regulation and storage


• Compact treatment plants • Sewer mining plants • Water reuse systems

and rhe United Stares Environmental Protection Agency. Extensive laboraro ry, pilot, and fu ll scale resting of various pre-treatment, reverse osmosis, and posr-rrearment processes has created an imp roved understanding of rhe design and operation of in tegrated membrane systems - decreasing produced water cost, 111creas111g water recovery, and extending rhe range of economically and environmentally acceptable applications fo r RO desalination in California.

Integrated membrane systems

T he Desalinatio n Research and Innovation Partnership (DRIP) - a $35-million cooperative research and development program under direction of the Metropolitan Water District of Southern California - began research nearly l O years ago on the use of integrated membrane systems fo r potable water production from brackish surface waters, as well as recycling of brackish agricultural waters and municipal wasrewarers. Figure 3. Founders of the UCLA W ater Tech no logy Research Critical Limitations and Members of the DRIP (WaTeR) Center (from left to right - Yoram Cohen, Bud Glater, Need for Innovation research program incl ude the a nd Eric Hoek) sta nd in front of a pilot system in one of the Universiry of California, Los Th ree additional W aTeR Center labora tories. Angeles and Riverside technological limitations must campuses, as well as numero us be overcome to enable Department of Water Resources, Califo rnia municipal water and successfu l, economic, and sustainable wasrewarer districts throughout rhe Stare of applicatio n of RO membrane desalination Energy Commissio n, American Water Califo rnia. Fund ing fo r rhe DRIP research and water recycling in Cali fo rnia. First, Works Association Research Foundation, biofouling in wastewater reclamation must program has come from the California the United Srares Bureau of Reclamation,



leading manufacturer of UV lamps for water + wastewater ultraviolet disinfection systems. low pressu re - high ou tput - amalgam medium press ure - electroni c control s replacemen t lamps + custom - design OEM service for domest ic + industrial UV disinfection systems + mu nicipal treatment plants. Heraeus - specialists in Ultraviolet + Infrared Technologies

V,cton.i 03 987,l 7-lS~

NSW 02 4 294 468?

l 02 MARCH 2006 water

www her,ieus-~mb.i com .iu

For over 80 years we have p rovided critical w ater infrastructure for our clients and communities. Using our speciali st skills in engineering, environmental, risk, w ater resources and water cycle mangem ent, we provide integrated solut io ns for yo ur business. To find out more please contact Selwyn Mcfaul or George Khouri in Brisbane o n (07) 3244 9600 or Gidi Azar in Sydney on (02) 8923 1555.

Success through Partnersh ip


be reduced because it drives the cost of recycled water. Biofouling has been described as the "Achilles heel" of membrane processes because it often can not be prevented by pre-creacmenc, membrane modification, or process optimisation (15). Second, product water recovery in brackish water desalination must be increased. Mi neral scale forma tion typically limits brackish water RO recovery co 50-80 per cent, which leads co major concentrate management challenges, especially fo r inland desalination facilities without ready access co permitted ocean discharge facilities. Thi rd , the cost of seawater desalination muse be red uced. The high pressures required co desalt seawater (u p co 1,000 psi) create excessive energy dema nds. Boch capital and operating costs are furth er exacerbated by concerns over fou ling; however, there is some hope char sub-surface intakes (i. e., "beach wells") will reduce fouling concerns as well as many environmental issues. T he Water Technology Research (WaTeR) Center was recently form ed in rhe Henry Samueli School of Engineering and Applied

Science at UCLA (www.wacercenter. ucla. edu). T he center includes facul ty and student researchers from UCLA and a number of universities around the world. In add ition, affiliate members include numerous scientists, engineers, and researchers fro m industry, a U.S. national lab, and local water utilities. The scope of research and ed ucation activities at rhe WaTeR Cen ter strives co equal the creacivicy and productivity of UCLA's efforcs in the 1950s and I 960s. T he Center's mission is co develop the technology needed co create che "desalination plane of the futu re" and co train che next generation of desalination experts co serve the Scare of California. Current UCLA effo rts co address che three key RO technology limi tations just mentio ned - fouli ng, scal ing, and cost - are summarised below. Combating fouling in wastewater recycling

After high-pressure pumping coses, rhe bulk of operating and maintenance costs for wastewater recla mation by RO is attributed co biofo uling, which leads co expensive pre-

treatment requirements, frequent chemical cleaning, and accelerated memb rane replacement. Ar UCLA, we have established a major research program co understand and control membrane biofouling. Among the approaches bei ng studied, the application of nanotechnology to develop hydrophilic and antimicrob ial RO membranes is producing excitin g results. A patent application has been filed based on a new class of reverse osmosis membranes [16) . Preliminary results sugges t chat membranes hand-cas e in che laboratory have up co double che water permeability of rradirional polyamide composite membranes (also hand-case in th e laboratory at UCLA) without a sacrifice in sale rejection. In add ition, an advanced microscopy techn ique - developed co enable direct observation of membrane biofoul ing [17, 18) - has confirm ed chat bacteria deposit much more slowly on these new membranes, are more easi ly removed once deposited, and are inactivated upon conracc with rhe antimicrobial membrane materials. Further, the direct observation system developed co study membrane

We'll help you make the most of it.

Only1% the world's water can be used fo'drinking. of

MWH plans, designs and delivers water systems that fully maximize our precious resources. Whether it's China's T hree Gorges Dam, or an advanced wastewater treatment plant on Australia's east coast, our commicmenc never varies. We build better systems for people, and for the planet

That's the nature of MWH.

O MwH. www.mwhglobal.com.au


MARCH 2006

l 03

membrane technology biofouling is now being extended for use as an early warning fou ling detector in full scale systems, which may greatly improve process optimisation, control, and cleaning operations. Increasing recovery in brackish water desalination

The only currently feasible approach co increase produce water recovery and thereby mini mise the generation of RO concentrate is co integrate brine minimisation technology into mulri-srage desalination systems. T he concept of accelerated chemical precipitation (ACP) by seeded crystallisation has been explored as a means of semi-selectively reducing the concentration of common scale forming mineral ions [19] . Accelerated precipitation studies on Colorado River water (CRW) and brackish agricultural drainage water (ADW) showed chat Ca2+, Ba2+, and Sr2 •, concentrations could be reduced by >90%, >80%, >75%, respectively. Moderate reductions (10%-20%) were also observed for magnesium, silica, and aluminum. In more recent UCLA/MWDSC DRIP studies, recovery up co 95% has been demonstrated in RO desalination of CR W and up co 95% and 98% recoveries have been achieved in UCLA laboratory studies for RO desalting of ADW and CRW, respectively. Work is continuing in collaboration with MWDSC and the California Department of Water Resources co optimise high recovery multi-stage approaches co brackish water desalination. In related research, an ex situ scale observation detector (EXSO D) was recently

developed at UCLA as an early warning system co detect the onset of mineral scale formation. The EXSOD approach also enables optimisation of pre-treatment process, RO operating co nditions, and membrane cleaning protocols. Ongoing research efforts at UCLA explore integration of ACP and the EXSOD scale detector into multi-stage RO/NF membrane systems co increase recovery and efficiency of both brackish water and seawater desalination. Reducing cost in seawater desalination

Desalination coses have declined largely due co high-efficiency energy recovery devices and more permeable seawater RO membranes, but improved process designs such as multi-stage RO/NF arrays, split partial permeate treatment, and high pH operation for boron removal have also con tributed co cost reduction [20] . T he cost of seawater desalination has reportedly dropped from about $1.5 (U.S.) per m3 in I 994 co about $0.53 (U.S.) per rn 3 in 2005 [21]. However, even if such cost effective seawater desalination becomes the norm, it remains at lease double rhe cost of traditionally imported waters in Cali fo rnia and more than three rimes the cost of reclaiming local brackish waters and municipal wastewaters [22]. The drought resistant supply of seawater remains all uring, and hence, researchers at the UCLA WaTeR Center are exploring the use of multi-stage NF/RO array configurations co red uce the cost of seawater desalination. For example, the Long Beach Water Department (LBWD) -

a UCLA WaTeR Center affiliate - has developed a parent-pending "rwo-pass NF" process capable of producing water with TDS of 200-400 ppm at operating pressures below 600 psi. UCLA WaTeR Center facu lty and students are working directly with LBWD engineers co help understand, improve, and scale-up the design of an optimised seawater desali nation facility. Potential benefits of multi-stage seawater NF/RO desalination systems may include lower operating pressures and improved product water quality due co the use of multiple membrane barriers.

Summary and Concluding Remarks The maturation of reverse osmosis technology has created an exciting era of technological innovation and new applications for water treatment professionals everywhere. The University of Cali fornia, Los Angeles has a proud history as a pioneer in desalination research and education. While the impact of the early research and development efforts is clear, perhaps of equal importance are the educational opporcuni ties and information disseminated from over fifty years of study at UCLA. In prepari ng chis review arricle, I am indebted to rwo colleagues at UCLA for providing details about the early history of reverse osmosis technology development, Dr. Julius "Bud" C later (emeritus researcher in the Department of C ivil and Environmental Engineering) and Dr. Yoram Cohen (Professor in the Deparrment of Chemical and Biomolecular

CONTAINMENT Minimal Maintenance Systems

~1 Jl PERMASTORE* ~ r •


104 MARCH 2006


4/8 Leighton Place

PO Box 240

Hornsby NSW 2077

Toll Free (from land lines) 1800 25 30 40 Fax 02 9477 7363

Engi neering and Director of the UCLA Water Technology Research Center). I am also graceful for che support and enco uragement I've received for my own research from many colleagues, coo numerous to name, in academia, industry, and government.

The Author Eric M V Hoek is Assistant Professor of Civil and Environmental Engineering and Fellow in the Henry Samueli School of Engineering and Applied Science, an d cofound er of che Water Technology Research Center at rhe University of Califo rnia, Los Angeles (UCLA), Email hoek@seas.ucla.ed u

[16] Hoek, E.M.V., Y. Yan, and B.H . Jeong, Nanocomposite membranes and methods of making and using same, US Paten t Application No. 60/660,428, 2005. (1 7) Kang, S.T., A. Subramani, E.M.V. Hoek, M .A. Deshusses, and M . R. Matsumoto, Direct observation of biofouling in cross/low m icrofiltration: mechanisms of deposition and release,/ Membr. Sci., 244 (2004) 15 I. (18] Wang, S., G. Guillen, and E.M.V. Hoek, Direct observation of microbial adhesion to membranes, Environ. Sci. Technol., 39 (2005) 646 l. [ I 9] Williams, M., R. Evangel ista, and Y. Cohen, Non-thermal Process fo r Recovering

Reverse Osmosis Concentrate: P rocess Chemistry and Kinetics, in Proceedings of the AWWA Water Qualiry T ech nology Conference, Seattle, WA, 2002. (20] W il f, M . and C. Barrels, opti misation of seawater RO systems design, Desalination, 173 (2005) I. (21] Wilf, M ., UCLA Graduate Seminar: New developments in RO seawater technology, Los Angeles, CA, 2005. (22] Gabelich, C., T. Yun, F. Gerringer, and A. Mofidi, The Desalination Research and

Innovation Partnership (DRIP) - Master Plan. 2003, Met ropolitan Water District of Southern California: LaVerne, CA.

References [ I] G later, J., The early history of reverse osmosis membrane development, Desalination, I 17 ( I 998) 297. [2] McCutchan , J.W. and B. Beorse, UCLA Dept. of Engineering Report 58-18. 1956, University of Cali fornia: Los Angeles, CA. (3) Glater, J. , Personal communication, 2005 . [4] Loeb, S. and S. Sourirajan, UCLA

Department ofEngineering Report 60-60. 1960, University of Cali fornia: Los Angeles. [5] Mnjikian, S., UCLA Dept. of Engineering Report 65-13. I 965, University of C alifornia: Los Angeles, CA. (6) C later, J ., A pioneer in desalination

education: The University of California programs, 4 ( I 994) I 0. [7) Lonsdale, H.K., T he Growth of Membrane Technology, j. Membr. Sci., IO ( 1982) 81. [8] Cadotte, J.E. , Reverse osmosis membrane, U.S. Patent 4 ,039,440, August 2, 1977. [9] Cadotte, J.E., R.J. Petersen, R.E. Larson, and E.E. Erickson, A new th in-film composite seawater reverse osmosis membrane, Desalination, 32 (l 980) 25. (10] Cadotte, J. , R. Forester, M. Kim, R. Petersen, and T. Stocker, Nanofiltrat ion Membra nes Broaden the Use of Memb rane Separation T echnology, Desalination, 70 ( I 988) 77. (11) Bray, D.T., Reverse osmosis purification apparatus, U.S. Patent 3,417,870, December 24, 1968. (12] Schwinge, J., P.R. Neal, D.E. Wi ley, D .F. Fletcher, and A.G . Fane, Spiml wound

modules and spacers: Review and analysis, 242 (2004) 129 . [ 13] Schock, G . and A. Miquel, Mass transfer and pressure loss in spiml wound modules, 64 (I 987) 339. (14] Yun, T.J., C.J. Gabelich, M.R. Cox, A.A. Mofidi, and R. Lesan, Reducing Costs far

Large-Scale Desalting Plants Using largeDiameter, Reverse-Osmosis Membranes, Desalination, in press (2005) [ I 5] Flemming, H .C., G. Schaule, T. Griebe, J. Schmitt, and A. Tamachkiarowa, Biofouling - the Achilles heel of membrane processes, Desalination, I 13 (1997) 215.

Clean water when needed! Save costs, energy, w ater and worries. Danfoss provides worldwide variable speed drive and soft starter expertise in water and wastewater processes, coveri ng the power range 0.37/1.SkW- 630/120OkW.


Danfoss (Australia) Pty Ltd - Motion Control Division Stand 303 Enviro 06 www.danfoss.com.au


MARCH 2006 l 05

MAGNETIC ISLAND WATER RECLAMATION PLANT D Young, C Hertle Abstract The first full scale Membrane Bioreactor (MBR) in Australia, the Magnetic Island Water Reclamation (MIWR) Plant, was commissioned in lace 2002. Citiwater Townsville owns and operates the MIWR plant. The plant was designed by GHD and constructed via two contracts. M ineforce were responsible for civil and electrical whilst Aquacec Maxcon (AM) did the mechanical installation. Kubota membrane technology was provided by Aquacec Maxcon via Aquator of the United Kingdom. G HD have supplied on-going technical support to Ciciwacer since chat nme. The MIWR has consistently produced effluent quality below licence requirements. Inter alia since commissioning in 2002 up to January 2006 the treated effluent has achieved BOD and suspended solids each < 5 mg/L (80th %ile) and turbidity <0.3, ammonia-Nd, Total N <2 and Total P < 0.2 mg/L (50th %ile). This paper summarises the performance of the plant since the introduction of raw sewage.

The results of three years operation summarised. Introduction Magnetic Island, a popular tourist destination located approximately 8 km north of T ownsville, Queensland, is comprised mostly of national parkland and is located within the Scace Marine Park of Queensland and the G reat Barrier Reef Marine Park. With an increasing populatio n and development, and concerns for environmental protection, the Picnic Bay STP was upgraded to a wastewater reclamatio n facil ity, with an MBR BNR process, in the year 2002.

The Magnetic Island Water Reclamation Plant The MBR BNR plant was commissioned in October 2002. The process, as installed, consists of: • Fine screening (3 mm), grit and scum removal;

106 MARCH 2006


gmµ Magnetic Island Water Recycling Plant Process Overview Odour Control



ShJd e Re

Anoxic zone Aerobic zone

= =

200 m 3 180 m3



= =

30 days 18 g/L

Figure 1. Process Flow Diagram. • Inlet storage and d ry weather flow balancing tank; • 4 stage Bardenpho-cype activated sludge process (P rimary anoxic - Primary aerobic Secondary anoxic - Secondary aerobic) configured in MBR format, where the secondary aerobic stage is a highly aerated reactor containing the membranes; • Membrane separation using flat sheet Kubota microfilcracion membranes (nominal pore size 0.4 mm; actual pore size in operatio n ~0.0lmm); • Recycle and wasting pumps and aeration blowers; • Drying beds for sludge d rying; • Alum dosing for P removal; • Caustic soda for pH correction; • Dosing facility for supplementary carbon source (e.g . sugar solution substrate); • Sodium hypochlorite for providing residual chlorine; • Storage and pumping of created water; and • Standby generator. An overview of the process is provided in Figure 1. T he ultimate design capacity of the plant is 8000 EP (Average Dry Weather Flow, ADWF = 2160 kL/d). Only the inlet works, however, was designed for the fu ll flow at 8000 EP. T he remainder of the plant was designed for 2000 EP (540 kL/d

or 6 .25 Lis). The full pl ant process capacity for 2000 EP was installed, except for the membrane b ioreactor MBR tanks where so far only one of two tanks has been fitted out. The M IWR was designed to achieve very low nutrient (TN<3 and TP<0. l mg/L) and faecal coliform levels (essentially zero) . The plant is also flexible enough to allow less stringent cri teria to be met. This allows for discharge to a sensitive environment during extreme wee weather events or reuse on an adjacent golf course in d ry weather.

Licence Requirements and Historical Performance Driven by concerns for the long-term protection of the sensitive marine waters of the Great Barrier Reef, which is a UNESCO Natural World Heritage Site, the environmental protection licence for this plant set stringent effluent quality standards. Inter alia the treated effluent is required to achieve BOD and suspended solids each < 5 mg/L (80th %ile) and ammonia-N <l , Total N <3 and Total P <0.1 mg/L (50th % ile) . Citiwater Townsville had further opted to conform to the South Australia reuse guidelines, targeting a clear effluent with <2 NTU turbidity (500/oile). At least 80% of the effluent was to be reused for irrigation of a local golf course adjoining the treatment plant near Picnic Bay. This required low

faecal coliform counts (5 no./ 100 mL, geometric mean) for land application of the effl uent. T he effl uent req uirements are summarised in Table I.

Table 1. MIWR Treated Effluent Q ua lity Req uireme nts.

Three years of operational data has been recorded fo r the MIWR. Daily on-site rests (> I 000 rest days) and fortnightly laboratory rests (89 cesc days) have been co nducted from October 2002 until January 2006. T he raw sewage data is summarised in Table 2 and effluent results are summarised in Table 3. Data presented in these cables excludes the commissioning period from October 200 2 to January 2003. T he in fl uent sewage is characterised by low actual influent BOD and COD, relative to design va lues (partly due to hoc cl imate and cha racteristics of the collection and pumpi ng system). T rends fo r effl uent quality data are shown in Figures 2 to 5. Performance has exceeded lice nce requirements since commissioning was co mpleted Oanuary 2003). T he plane has produced consistent excellent BO D, T SS, T N, T P, turbidity and faecal co liform removal. Removal of indicator bacteria (faecal coliforms, FC) has been good with only nine out of eighty- ni ne samples (up co July 2003) exceeding O FC/ l 00 mL. Six samples exceeded the max imum licence requirement (> 25 FC/ 100 mL). T here is no clear explanation for these sporadic spikes in faecal coliform levels; however ir is suggested chat ir may be due to sampling collection issues (ie dirty bottles). In theory there should be no transfe r of faecal colifo rms through membranes (unless there is an integrity breach, which wou ld show up in turbidi ty data) . Effl uent tu rbidi ty has been very good for a reclaimed wastewater, with no results exceed ing 1.8 NTU. N itrogen removal is often constrained by raw sewage characteristics, with a low ratio of COD (o r BOD) to T KN being unfavourable. Th is problem has occurred at Magnetic Island and is exacerbated by the relatively long retention rime in the sewage collection and pumping system. To achieve Total N requirements operations have focused on sugar solution dose rares, improved conrrol of recycle rares and aeration (DO) settings. P removal using simultaneous precipitation with alum has been very successful, aided by the long SRT s (giving efficient use of alum) as well as the excellent solids separation performance of the membranes. T he plant is most likely operating at or very close to the solubility limit of phosphorus, raking chemical equilibrium into considerations. T he high alum dose may be

Suspended Solids Ammonia - N Total - N Total - P (new) pH Turbidity Faecal Coliforms No.



Effluent Quality Requirement

Requirement Type

Maximum Value


mg/ l mg/ l mg/ l mg/ l mg/ l


80 %ile 80 %ile 50 %ile 50 %ile 50 %ile range 50 %ile geomean

10 10 3 10

<5 3 6.5 ¡ 8.0 <2 5

NTU per l 00 ml

5 25

Table 2. MIW R Raw Sewa ge Da ta (Feb 2003 - Ja n 2006). Parameter


50th %ile

80 %ile

Maximum Value

COD BODs BOD5 + sugar

mg/ l mg/ l mg/ l mg/ l mg/ l mg/ l mg/ l

245 86 428 57 47.2 59.9 7.4 7.2 1045 323

35 1 125 498 123 54.7 69.5 8.7 7.4 1181 362

10 10 255 758 428 72.0 84.0 10.2 7.9 3020 990

50th %ile

80 %ile

Maximum Value

2 0.15 2.0 0.20 7.4 0.2 0

3 5 0.54 10.0 0.48 8.7 0.5 0

9 25 2 26.6 7.7 10.2 1.8 3000

Suspended Solids Ammon ia - N Total - N Total - P pH Conductivity Alkalin ity

uS/cm mg/ l

Table 3. MIWR Performan ce (Feb 2003 - Ja n 2006). Parameter



mg/ l mg/ l mg/ l mg/ l mg/ l

Suspended Solids Ammonia - N Total - N Total - P pH Turbidity Faecal Coliforms

NTU No. per l 00 ml

Influent vs Effluent BOD 1000


,g_ E



L __



_ __





_ __


_ _ _ __ _ J

_ _ _ _ _ _ _ _ _ _ __


j-+- tnnuent BOO _.,_ Effluent BOD




Estimated Total BOD, Including sugar addition ]

Figure 2. BOD removal performance a t MIWR plant (October 2002- January 2006).


MARCH 2006 l 0 7

membrane technology contriburing significa nrly to sludge producrion in the sysrem and may be impacring on mixing perfor mance and membrane flux rares. The TP licence is currently being renegoriared from 0.1 mg/L to I mg/L (as noted in Table 1) due to the high levels of reuse on rhe golf course, where the nurrienr contenr has value. Membrane performance to dare has been excellenr. The trans-membrane pressure (TM P) has been consisrenrly lower rhan rhe design value. Si milarly, membrane flux has consisrenrly exceeded the design value. It appears that membrane perfo rmance has been aided by the warm temperatures at Magnetic Island (water remp range approx. 25 to 32°C), relative to planrs operaring in rhe temperare cl imates of UK, Europe or Japan.

Influent vs Efluent TSS

1000 ,-------------------------------,



10 LICENCE TSS 5 mg/l (80%1LE)

HISTORICAL TSS 5 mg/l (80%1LE)

' ., J. •

..... ...

-1----- - - - -- - -- - - - - - - - - - - - - - --~---'

0.1 1/0912002



2310<112004 L-+-1nnuent TSS





Figure 3. TSS removal performance at MIWR plant (October 2002- January 2006).


Influ ent vs Efluont Nitrogen

The MIWR has been operaring successfully fo r over rhree years and has consisrenrly met stringent licence requiremenrs, wirh excellent removal of BOD, TSS, rurbidity, TN, T P and faecal coliforms.


Notable characteristics ofplant performance over the past three years have included: • Continuous excellent BOD, TSS, TP, T N and faecal colifo rm removal;

!fl I

• Compliance with all design parameters over the lasr 3 years of operation;

M ""

• Ab ility to operate under low loads; 0.1 ,__ _ _ __ _ __

• Reliable performance of rhe membranes;



...,___ _ _ _ __ 611012003


I-+- Influent Ammonia as N _.,_ Influent TKN

• Ease of operability and maintainability;

-H-- - -- - - - -~

_ _ __ _ 911112004 Effluent TN


Effluent TKN -


Effluent NOX


• Ease of membrane cleaning;

Figure 4. Nitrogen remova l performance at MIWR plant (October 2002- January

Lessons learned have been:


• Highly intermittent flows have meant the balance rank is invaluable;

Influent vs Effluent TP and alum dose rato

• Aeration con trol is im po rtant;

100 - r - - - - --


• Supplementary carbo n source (sugar solution) has enabled low total N to be mer consisrenrly;


- - - - - - - - - - - - - - - - - -- -- - - - - ,


• High alum doses impact on sludge production, membranes and mixing; • The need fo r more regular cleaning after 3 years of operation; and • T he importance of QA on nitrogen and phosphorus analyses ar low levels.

Other positive aspects associated with plant operation have included: • Clienr sarisfacrion rhrough irs performance, simplicity of operation and low mainrenance of rhe plant; • Minimal odour, it looks great and is easy to ger around;

108 MARCH 2006


. ~




-1----- -- - - - - - - - - -- - - - - - - - -- - -- - -..J

0.01 1109/2002





l:1::.roTAL P (in) mg/I ::<>-TOTAL P (out) mgn -



Alum dose. kg/d )

Figure 5. Total P removal performa nce at M IWR plant (October 2002- January 2006) (QLD EPA licence for TP is currently being renegotiated to l mg/ L 50%ile).

UV For Taste and Odor Control • H igh profi le, with lots of visitors; • Supplier's design performance expectations exceeded; and • T he green-keeper of the adjacent golf course (who uses the water) proudly shows his before and after photos and can' t get enough of the water! Since the plant was constructed rhere has been a significant reduction in the cost of delivering this style of plant. The proven performance of this facility combined with increasing stringent licence conditions and a significant reduction in rhe cost of membrane systems, is likely to result in a larger number of MBR facilities in Australia over rhe next decade

Postscript As this paper is being written the Water Matters All iance is about to install the second set of membranes in the seco nd membrane rank. This will allow us to inspect the condition of membranes in tank I that have been in the water for over 3 years and subjected to a high dose of alum and addition of carbon to optimise nitrogen removal. The results of rh is investigation will be provided ar in a separate paper at Envito 06, May, 2006. Ciriwater is collabo rating with James Cook University, Kubota and Aquatec-Maxco n, to design, bu ild and ru n a pilot plant using the Kubota membrane to investigate fl ux efficiencies and biofou ling control and management. Dr. Phil Schieder fro m ]CU has been the principal technical force behind the study which is set to ru n for rhe nex t three years.

Acknow ledgments Many people have been involved in this project, from concept, to design, supply and comm ission ing. In particular, thanks are expressed to the fo ll owing people: • David de Haas and Rob Lowther, of GH D Pry Ltd, Brisba ne office. • Ken Hartley, Principal of Ken Hartley Inc. • Peter T uri and Henry Fracchia of Townsville CiriWater. • Karl Naumann of Aquatec Maxcon.

The Authors Dale Young is a process engi neer with fou r years experience in the fields of water, wasrewarer and aquatic science. dale.young@ghd.com.au; Chris Hertle is a Senior Chemical Engineer who has been involved with the Picnic Bay MIWR project from inception. Chris.herrle@ghd.co m.au. Chris and Dale have had an active role in monitoring the progress of the MIWR over rhe last three years. They continue to provide technical assistance and support to T ownsville Ciriwater to ensure the ongoing success, improvement and long-term sustainability of rhe Picnic Bay MIWR MBR facility, as well as involvement in rhe MBR design for Townsville's Horseshoe Bay and Cleveland Bay plants.

References Hertle, C.K. and P. Turi (2002) Picnic Bay Membrane Bioreactors for Wastewater Treatment at Magnetic Island, Australia (paper e2 l 409a) , !WA 3rd World Water Congress, Melbourne Victoria. Turi, P. and Crofts, J. (200 1) Australia's First Membrane Bioreactor Sewage

Treatment Plant to Protect the Reefat Magnetic Island, IPWEAQ Conference, Mareeba, Q ueensland. D e Haas, D. , Hertle, CK., Turi, P. (2004) Magnetic Island Water

Reclamation Plant - Membrane Bioreactor Nutrient Removal Technology One Year On, Enviro Conference.

Introducing Trojan UV-Oxidation - A Better Approach

Using a combination of ultraviolet light and hydrogen peroxide, th e TrojanUVSwifr"ECT delivers fu lly-validat ed, microbial disinfection for drinking wat er - and a proven way t o add ress seasonal taste & odor (T&O) events, alga l toxins and oth er chemica l conta minants. Our advanced UVoxidation process is a cost -effect ive, userfriendly way to t reat MIB, geosmin, and other T&O-causing compounds derived from decaying algae blooms in drinking wat er sources. For cost-effect ive, operator-friend ly drinking water treatment solutions th at deliver reliable d isinfect ion and taste & odor control, th e world is turning to Trojan. www.troj anuv.com. Represented by: Aquatec-Maxcon Pty Ltd 119 Toonga rra Road PO Box 455 IPSWICH QLD 4305 Phone: + 61 (7) 3813 7100 Fax: + 61 (7) 3813 7199 Email: enqu iries@aquatecmaxcon.com.au Website: www.aquatecmaxcon.com.au

..:• ··:..



MARCH 2006 109

AUSTRALIA'S TWO LARGEST MBR WATER RECYCLING PLANTS M Newland Summary Tenix Alliance Pty Ltd has recently designed and commissioned the two largest, state-of-the-art membrane bioreactor wastewater treatmen t plants in Australia: one treating domestic sewage in Victor Harbor, SA, and another treating ind ustrial wastewater in Forrestfield, WA. Both plants featu re aerated, submerged membrane modules (fl at place and hollow fibre memb ranes respectively) as a key enab ling technology to meet water recycling objectives. The fo rmer is achieving a high q uality treated wastewater, including total phosphorus of less than 0.05 mg/L, and provid ing water for agricultural re- use. The latter plant incorporates Reverse Osmosis (RO) to p rovide up to 80% industrial reuse and represents a cost-effective model chat could be adopted in a n umber of different industries. This paper provides some details on the design and operational aspects of each plant.

Design and commissioning data. Victor Harbor Wastewater Treatment Plant

Victor Harbor MBR aeration blowers and permeate piping. arc facil ity fro m the inception of th is project. T reared wastewater fro m the plant is pumped to the recently refurbished Hindmarsh Valley Reservoir and from th ere it is reticulated for irrigation re-use by neighbouring agriculturalists. H owever, if the reservoir is fu ll , the created wastewater will be discharged to the sensitive receiving waters of the local In man R iver. A particularly strict treated wastewater

p hospho rus req u irement has been imposed o n the plant to accommod ate re-use and , especially, any potential river d ischarge. T able l lists the treated wastewater quality specifications that the plant m ust meet. Sewage characteristics in Victor Harbor refl ect the holiday nature of the township and can b e d ivided into two categories: a relatively stable base load period (12,0 00

Victor Harbor is a rapidly growing seaside town and a popular holiday destination o ne and a half hours d rive southeast of Adelaide. Tenix Alliance designed, constructed and commissioned a Membrane Bioreactor plant fo r United U tilities Australia (UUA) in 2005. UUA own the facility and will now operate it for 20 years under a contract with Sou th Australia Water. T he plant is designed for an ultimate dry weather flow of 5.1 ML/day (peak wet weather 11.2 ML/day), with the current mem brane fit-out designed to accommodate dry weather flows up to 3.4 ML/day. Additional membrane modu les will be added as sewage flows in crease in futu re years. The desire of the lo cal community, and the requirements imposed by South Australia EPA, drove th e adoption of a state-of-the-

110 MARCH 2006


Victor Harbor MBR Membrane Module Aeration.

Ride the wave of expert water and wastewater solutions United KG , a leader in water and wastewater design, construction,

To find out more, visit

operations and maintenance services, will now be known as:


United Group Infrastructure.

or call +61 2 9492 8878.

The new branding reflects our continued innovation and increased expertise. After a boost to our design and construction skills with

And choose the organisation that's making waves in water and wastewater solutions.

t he acquisition of a highly regarded Water Projects company, our in-house design and engineering capabilities now offer even more innovative solutions. United Group Infrastructure can provide all aspects of design, construction, commissioning , operation, maintenance, and -where necessary - ownership and asset management.

www.unitedgroupltd.com Level 1 0, 2 Elizabeth Plaza NORTH SYDNEY NSW 2060

T: +61 2 9492 8878 F: +61 2 9492 8894 E: ugi.informat ion@unitedgrou p ltd.com

populatio n); and high concentration, h igher and variable flow periods during holidays or long weekends (peak population 35,000). T able 2 sum marises typical wastewater characteristics measured during 200 5 and up to end January 2006.

Table 1. Victor Harbo r WWTP Treated Wastewater Quality Specifications. Parameter

Median Concentration

90%ile Concentration

BOD TSS NH4-N TN TP F coliforms Turbidity Pathogens Enterococci

5 mg/l l mg/ l l mg/ l 5 mg/l 0.05 mg/l l Ec/100 ml

10 mg/l 2 mg/l 2 mg/l 10 mg/ l 0. 1 mg/l 10 Ec/ 100 ml Maximum 0.5 NTU Maximum l org/ 50 l Maximum 10/1 00 ml

since the comm issioni ng period (Table 3) . D uri ng the co mmissio ni ng period , samples of post-M BR m emb ranes collected prior to UY d isin fect ion were analysed and these results also confirmed that the membranes were p rovid ing a barrier to E-coli and, by inference, to any larger organisms such as Giardia and

The rrearmen r plane consists of Cryptosporidium. inlet works (3mm spiral sieve screens, grit rem oval), twin Joe White Maltings bioreactors and membrane rank Wastewater Treatment trains, UV d isin fection, treated Plant, Forrestfield, Perth wastewater d iurnal balan ce Joe W h ite Ma!tings is expand ing pond and sludge thickening. Table 2. Victor Harbor Sewage C haracteristics. the p rod uction capacity at the Several new remote pump stations were constructed as Perth malting facili ty by over Holiday Periods or Long Non Holiday Periods Parameter 120% . A key componen t of chis part of the project, alo ng w ith Weekends (range) (50%ile concentrations) expansion project is rhe provisio n approximately 12km o f new Flow (2005) 2.5 ML/day 3.0 - 3.6 ML/day of an integrated wastewater sewage rising m ain and treated BOD 240 mg/l 150 - 285 mg/ l wastewater tra nsfer and treatment and water recycling reticulation p ipeli nes. T he facility on site. COD 410 - 85 4 mg/l 580 mg/l bioreactors have continuous l 08 - 325 mg/l 297 mg/l TSS The malting process involves the inflow, bur are incermirrenrly germ ination of barley, where 54- 80 mg/L NH4-N 48.2 mg/l aerated to achieve n itrogen starches are converted to maltose. 61 - 85 mg/L 64.7 mg/l TKN removal. There are fou r Ultimately the m ale is shipped to 10 - 17 mg/ l TP 10.9mg/l m embrane tanks in total (two breweries both nationally and per b ioreactor) and each internationally. The malting and dosing of molasses to im prove the m em brane rank presently h as six Kubota operation is a 24 h ou r per day, 365 day per wasrewater's relatively low ca rbon to EK400 flat plate membrane modules year operation . T he process produces a nitrogen ratio and hence assist w ith installed. Each m embrane module p rovides relatively h igh volume of wastewater as the nitrogen removal. After just fi ve weeks of 320m 2 of membrane area. Design average germi nation process is initiated by steeping commissioni ng the plant dem onstrated fl ux rate is 0.6m/day, with a shon-rerm the barley in copio us amo unts of water for consistent treated wastewater q uality of maxim um of J .1 m/day acceptable during approxim ately 2 1 hours and then d raining BOD < 2 mg/L, T SS < J mg/L (essenrially wet weather infl ows. Permeate is d riven off th e resultant liquor. 0 mg/L), TN < 5 mg/Land total th rough rhe membranes by the maintenance A combination o f Memb rane Bio reactor phospho rus resul ts less than 0 .05 mg/L. of a 0.3m - 0.8m head of water across the and Reverse Osmosis technology has been mem brane (this trans-membrane pressure is T he combination of permeate ul rrafil tration adopted fo r chis p lant to p rovide a treared dependent on th e fl ux rare and the in the MBR mem branes and UV water stream of a quality "better than Perth condition of rhe membranes), and it then disinfection has p roven reliable in achieving drin king water". U tilising d ata from a pilot fl ows by gravity to a UV Technologies the treated wastewater microb iological trial undertaken in early 200 5, and Australia (UVT A) 'Term inator' ultraviolet specifications. Post UV d isin fect ion results experience fro m another full-scale maltings disinfection system. have been negative fo r all bacteria, viral and plant MBR facility in Belgium, Tenix The planr is provided with six 55kW path ogens in tests u ndertaken d uring and Alliance has designed, constructed and positive d isplacement aeration b lowers commissioned the new facil ity arranged with one per membrane rank, plus Table 3. Victor Harbor W W TP (commissioning completion March 2006) . one shared between bioreactors and one spare. The Kubota flat plate membranes require continuo us aeration (air scouring) to avoid fo uling. Perm eate extraction is critically interlocked w ith aeration as even a very short period of permeate fl ow u nder non-aerated conditions risks seriously fo uling rhe membranes and potentially requiring a labour intensive chem ical clean. Consequently, even with process adjustments mad e to reduce aeratio n rates during low flow periods, app roximately 60% of the total aeration power is consumed by the membrane modules on the Victor Harbor plant. O peration of the plant has been augmented with alum dosing for phosphorus removal

112 MARCH 2006


Microbiolog ical Results - Post UV Disinfecti on. Parameter


E coli Enterococci Adenovirus Reovirus Hepatitus A virus Norovirus Rotovirus Enterovirus Cryptosporidium Giordia Helm inths

0/ l00ml 0/ l00 ml < l /50 l < l /50 l Negative/] 0 l Negative/ l 0 l Negative/ l 0 l < l /50 l Negative/ ] 0 l Negative/ l 0 :l Not Detected

The design requirem ents are listed in Table 4 . As shown, the maltings wastewater BOD concentratio n is significan tly h igher than domestic sewage and hence th is project, to the best of our knowledge, represents rhe largest MBR in Australia when measured o n the basis of daily organic loading rate. The treatment process consists of screening

(J mm d rum screen), fl ow equalisatio n, membrane bio reactor, membrane fi ltratio n (two parallel membrane tan ks), reverse osmosis and mechanical biosolids dewatering. The bio reactor consists of a bioselector zone (to encourage the p roliferation of fl occulant-type biomass) and a main

aeration zo ne. T he coral bioreaccor volume is approximately 2,200kL. Aeration is provided by a removable, submerged diffuse r system and two (duty/standby) 75kW positive displacement blowers. The design MLSS concentration in rhe bioreaccor is 12,000mg/L. Recirculation of mixed liquor fro m the bioreaccor co rhe membrane ranks is provided by an axial flow mixer. Puron MBR membrane modules, manufactured by Koch Membrane Systems in Germany, were selected for this project. T hese membranes consist of reinforced hollow fibres with single end attachment. The aeration is centralised within each membrane bundle co ensure chat sludge accumulation is avoided. Each compact module consists of 180 membrane bundles. T he modules are suspended in the membran e tank from the cop and can be readily removed without the membrane rank being drained. Permeate and air co nnections are also fro m above. Aeration is provided by two (duty/standby) l 5kW positive displacement blowers. Figure 1 ill ustrates a standard Puron MBR membrane module. The plant incorporates a total of eight membrane modules, each with 588m2 of

Penneate removal in each row

Fibre support

Module row.;

Figure l. Puron Membra ne Bioreactor M embrane M od ules.

active membrane area (coral area 4,704m 2) . The modules are arranged in groups of fou r and placed into two process streams. T he design net fl ux rate adopted for the malrings wastewater is 15 L/m 2/ hr (note:

chis is considerably lower than would be adopted for th e treatment of domestic sewage). Aeration of rhe Puron membrane modules is intermittent and each module receives

Innovative Solutions Ciba Specialty Chemicals manufacture and supply an extensive range of products specifically designed to provide innovative solutions for you in sedimentation, clarification, filtration, flotation and dewatering applications. Our product range includes Coagulants, Flocculants, Defoamers, Deodorisers, Decolourants, Anti scalents and Dust Suppressants. We also offer systems such as Ciba AUTO-JETWET® and Ciba® ALCOTECH for the safe and effective, preparation and dosing of Ciba® ZETAG®, Ciba® MAGNAFLOC® and Ciba® MAGNASOL® products. We offer you our expertise in water treatment to help you deliver results. To find out more about how Ciba improves your process treatment

with innovative solutions, contact:

Ciba Specialty Chemicals 6 Donaldson Street Wyong NSW 2259 Tel: 02 43503200 Fax: 02 43532136 email: warrlck.green@clbasc.com or visit www.clbasc.com/ wws


Value beyond ch emistry


MARCH 2006 113

membrane technology Table 4. Joe White Mailings MBR WWTP Design Basis. Parameter




1,410 kl/day 1,600 mg/L 3,000 mg/L 327 mg/ L 70 mg/L 8 mg/L 5.5

l ,700 kl/day 2,000 mg/ L 3,850 mg/L 500 mg/L 100 mg/L 10 mg/L 8.0

1,600 mg/L

2,000 mg/ L

BOD COD TSS Total Nitrogen Total Phosphorus PH Total Dissolved Solids

aeration for 25% of the rime. This intermittent aeration provides considerable advantage in regard to operati ng costs and process d esign compared to other membrane systems rhar rely on constant, vigorous aeration for membrane scouring. The blower output is adjusted by variable frequency d rives on all rhe blower motors with control of rhe bioreactor blower speed set accordi ng to DO set points and the permeate flow requirements for the membrane module blowers. In 'Process Mode' the filtrate is d rawn fro m rhe modules via a rotary lobe permeate pump at a flow rare controlled to maintain a constant level in rhe bioreacror. In ' Backflush Mode' , rhe permeate pump direction is reversed fo r a sho rt period of time and permeate at a higher flow is used to loosen solids from the membrane fibres. T hese solids are then effectively flushed our of the membrane module by rhe integrated aeration system .

In situ cleaning is undertaken on a monthly basis by dosing cleaning chemicals (hypochlori re, caustic and separately cirric

Water Advertising To reach the decision-makers in the water field, consider advertising in Water Journal, the official journal of Australian Water Association. For information on advertising rates, please contact Brian Rault at Hallmark Ed itions, Tel (03) 8534 5014 or email brian. rault@halledit.com .au

114 MARCH 2006


Joe White Maltings MBR WWTP Bioreactor. acid, if required) in to rhe backflush sequence. T his chemical cleaning occurs automatically with no operator labour required. There is also no requirement to remove membranes or to rake a memb rane rank off-line and perform a chemical cl eaning soak. MBR permeate becomes the feed for a 2ML/day capacity Reverse Osmo sis plane. This package plant has been p rovided by Koch Membrane Systems and incorporates twelve 18inch diameter by 1.55m MegaMagnum, T FC, ULP, spiral wou nd RO elemen ts. Each spiral wound MegaMagnum RO element has approximately 260m2 of active membrane area. The RO elements are loaded in three 18 inch diameter S-5, FRP membrane housings rared for a maximum o f 300 PSIG operating pressure. The membrane housings are arranged in a two by one array. Anriscalant and pH adjustment chemicals are dosed to the RO feed stream and a clean-in-place (CIP) system is provided, with an expectation of a >3 monthly cleaning interval. A typical analysis of rhe MBR permeate from rhe malrings wastewater is shown in

Table 5. Typica l Concentratio n of Major Ions in Mailings Plant MBR Permeate.

Tab le 5. The d issolved salts in this water are dominated by potassium and chloride. The RO system d esign achieves 75% 80% recovery of water at a guaranteed TDS concentration of less than 400mg/L (design intent 250mg/L). The RO concen trate stream, with a TDS concentration of approximately 6,800 mg/L, is directed to rhe sewer. T he new facility will enable rhe maltings plant to conserve over 400ML of potable water each year. As a result, Joe White Malrings will no longer be one of the largest ind usrrial water users in Western Australia.

Summary The projects described in this paper represent the application of stare-of-the-arr membrane technology to wastewater treatment and water recycling. The technology has proven reliable and robust in these widely d ifferent applications and the processes are relatively simple to operate. Tenix Alliance is also presently completing several smaller town MBR sewage treatment plants in Bega Valley Shire, NSW as part of the Bega Valley Sewerage Program, as outli ned in the December ed ition.

The Author



Mark Newland is Principal Process

Calcium Chloride Fluoride Magnesium Potassium Sodium Sulphate

32 mg/L 340 mg/L l mg/L 31 mg/L 600 mg/L 200 mg/L 24 mg/L

Design Engineer with T enix Alliance and has over 18 years p ractical biological wastewater treatment plant d esign and commissioning experience. He leads rhe process engineering, tech nical innovation and commissioni ng ream within Tenix Alliance. Contact: Phone (08) 9470 40 04, mark.newland@renix.com

MEMCOR® MEMBRANE BIOREACTOR PRODUCTS - FROM RESEARCH TO COMMERCIALISATION F Zha the bulk phase in the simple air scour process. This process is, however, hindered by the mass transfer rate, especially when the membranes are densely packed. Such a passive refreshment process limits the membrane packing density in modules and in tanks. Loosely packed membranes will require more air to effectively sco ur the membranes [2) and to increase the membrane tank vo lume. Moreover, the simple air scour process often generates local dehydrated solids, blocking the air distributors and plugging the memb rane modules. T o overcome chose issues, Memcor developed a rwo-phase scru bbing mechanism. Gas and liquid are mixed first and then the rwo-phase flow of mixed liquor and gas bubbles scours th e memb rane surface. The positive liquid flow effectively refreshes the membrane surface and minimises the so lid concentration polarisation, while the gas bubbles perform the scouring effect. Figure l illustrates such a rwo-phase cleaning mechanism. The rwo-phase scouring process described above requires less energy to keep the membrane clean and reduces the risk of blocki ng air distributors. The air requirement fo r scouring membrane with

Abstract Memcor MBR products have been on the market for many years. T his paper provides an overview and summarises their distinctive feat ures.

Introduction With the increasing water sho rtage and more stringent discharge limit, the membrane bioreacto r (M BR) is playing a more and more important role in the treatment and recycling of wastewater. Sin ce the first publication of submerged membrane bioreactor by Yamamoto et al in I989 [l ], the technology has advanced very rapidly. Nowadays, there are many commercial MBR products in the market place and hundreds of full-scale MBR plants are in operation worldwide.

The two-phase scouring process requires less energy. Memcor's Research on MBR Initial MBR research conducted by Memcor's R&D group in Australia directly foc used on overcoming the problems occu rring in existing MBR plants. When the MBR technology first emerged, one of the major attractions was considered to be a simple process of immersing membranes in the bioreactor to replace the conventional clarifiers. Air is injected to the bottom of modules to scour the membrane thus preventing significant foul ing. T he membrane behaves as a physical barrier to completely reject the suspended solids and eliminate the problems of sludge bulking encountered in the conventional wastewater treatment processes. This in-tank MBR process with simple air scouring looks simple, but problems often appeared in the full scale plant operation. In an MBR system, the membrane fil ters mixed liquor that normally con rains suspended solids of more than 8,000 mg/L.

Figure 1. Concept of two-phase scouring . As the clean water is fil tered through the membrane, more solids are left on the surface and in the vicin ity of the membrane. T his creates a higher viscosity and hence a higher resistance to fil tration. T he membrane surface is only partially refreshed th rough the mass transfer with

FIitration Performance of MBR Module (M BR4 -7)

•oo 350 300




~ 200




. .-·-




l -.. ,...... . ~-~




Pr•••ur• Cnn1ducerf1ull







. ..


35 30 25



20 ;,

. .

i 15 10 5 0

0 4-Nov


I Flii••----slL11t11

~. . ~ ••













Figure 2. Membrane Performance in a MBR Pi lot.


MARCH 2006 115

two-phase flow is less than 0. 3 N m 3/hr/(m 2 membrane), significantly lower compared to >0.45 Nm3/hr/(m 2 membrane) required for the simple air scouring process. This has been validated on various pilo ts and fu ll scale MBR plants. Figure 2 shows the result from a pilot study conducted at Memcor R&D site in Windsor, Australia. During the chree-monch operation, the membrane's permeability was stable ac a fl ux of 30 L/m 2/h r. N o blockage in the system occurred during the cesc. Another issue for che in-tank memb rane MBR is the difficulty in perform ing fu ll chemical clean on the membranes, especially for large planes. No matter what type o f membranes is used in M BR, che membrane eventually requires a chemical clean to fully recover its permeability after operation for a certain period of time. One way to perform such a clean is to backwash the membrane with a high concencracion chemical solution. If chis meth od cannot recover the membrane's performance, then the modules have to be removed from the b ioreactor and cleaned in a separate rank. T his cleaning strategy has several negatives: (1) Inj ecting chemicals, normally high concentration of chlorine, w ill disturb the bio logical accivicies and generally create serious foaming problem; (2) Removal of mod ules out of the aerobic tank into another tank fo r cleaning causes a long down-time and involves operators d irectly handling of modules covered with sludge (non-friendly environment). Alexand er et al (3) described the cassette-by-cassette clean which was expensive, labou r intensive and potentially problematic for the memb ranes.

Figure 3. A Memcor B1OR Rack.

The Memcor MemJet system has traditionally adopted th e design with separate membrane cells from biological tanks. This design h as eliminated th e clean ing problem described above and provided a fl exible operation and control.

MemJet™ Immersed Membrane Bioreactor Systems After years of research and develop ment, Memcor launched its MBR product, MemJet™, in 2001. The product h as the features of:

Figure 4. Assembly of A B30R Ra ck.

• The unique patented two-phase efficient cleaning process; • Positive fluid t ransfer into fib re bundles and uniform d istrib ution of fl ow and solids; • Cross-flow dynamics with min imise energy consumption; • Automated, in-place membrane cleaning process; • Small foo tprint required for the membrane system; • A safe environ ment for plane operators.

Membrane and Modules The Memcor hollow fi bre membrane used in the MemJet™ MBR is hydroph ilic and chlorine tolerant. The membrane typ ically achieves a fil trace q uali cy of: • T urbidity <0 .2 NTU (typically around 0.1 N T U) • TSS <l mg/L • Faecal coliform <2 cfu/100 mL

Figure 5. Dua l Membrane Cells in Operation.

Joint studies with other companies have also demo nstrated a log reduction of viruses greater than 4. Hollow fibres are potted into modu les. Memcor supp lies two types of modules fo r MBR applications: BI OR and B30R. The effective membrane surface areas are 10 and 38 111 2 respectively.

size of a plane, a n u mber of racks are installed in a membrane tank to meet the flow capacity. Figu re 5 presents two cells of memb ranes in MBR operation side by side with uniform distribu tion o f air bubbles and mixed liq uor.

MemJet™ System The MBR mod ules are assembled into racks. Each rack can accommod ate up to a maximum of 40 Bl OR or 16 B30R modules. Figu re 3 is photo of a rack of 40 BI OR modules and Figu re 4 shows a rack of B30 R modules b eing assembled. The racks are then installed in parallel in a tank or memb rane cell. Depending on the

116 MARCH 2006


Memcor MBR Plants Since the launch ofMemcor MemJet™ in 200 1, the product has attracted a lot of attention from customers. So fa r the largest MBR project awarded to Memcor has a capacity o f 15 MLD . The fi rst MemJet plane started operation in USA in 200 2.

The first MemJer plant in Australia is Sydney Water's North Head MBR plan t with a capacity of 2 MLD which starred operation in July 200 5. Construction of the recycled water plant was undertaken o n behalf of Sydney Water by CHJY Freshwater, a joint venture between C H 2M Hill Australia Pry Ltd and John Young (Kelvinhaugh) Pry Ltd . T he design of MBR was based o n a single biological train (MLE process configuration) incorporating rwo trains of Memcor membranes [4] . The plant successfully passed a six-month performance-proving period at the end of January 2006 and is now operated by Sydney Water. U p ro 1.5 MLD of drinking water is being saved each day, with the commissioning o f a recycled water faci lity for onsite operations at North H ead Sewage T reatment Plant, according to Sydney Water Managing Director, David Evans [5].

Case Study The Calls C reek Wastewater Treatment Plant (WWTP) in USA was upgrad ed to MBR in April 20 04 using the Memcor MemJer product. In 2002, Oconee County recogn ised an urgent need to meet two

Existing EQTank

Upgraded 3 Channel Orbal Process (add a•ration capacity and rework trande r po rh for hydraulic•)

Figure 6. Schematic Layout of Calls Creek MBR Plant.

goals for rhe Calls Creek Wastewater Treatment Plant (WWTP): T o increase the treatment capacity with minimal footpri nt expansion, and T o meet stringent regulations for urban reuse foreseei ng the poten tial to reuse in the fu tu re. T he County looked to an ad vanced treatment solution combining superior

water quality in a small footprint. The Membrane Bioreactor (M BR) technology was chosen d ue to the ability to provide: • reliable, safe water q uality • compact fo otprint • decreased sludge waste The MEM COR MemJer system p roved to be the best technology to fir the need s of this growing co mmunity. This u nique

ORIHRU Unlimited Polymer Solutions

RAUBIOXON PLUS THE NEXT GENERTATION IN FINE BUBBLE AERATION In 1995, after closely analyzing the various failure modes of commonly used diffusers and listening to the experiences of wastewater treatment plant owners and operators, • Patented anti-microbial for- REHAU developed and manufactured the RAUBI0X0N silicone pipe diffuser. Today, mulation 10 years on since the first installation, the majority of all original diffusers are still in • Low susceptibility to encrus- operation. Double the service life of other commonly used diffusers! • Advanced German engineered silicone membrane

tatlon and no hardening or brittleness

• High oxygen transfer effi-

ciency • Quick and easy installation

Leading Edge International Technology with Local Support and Knowledge: With our local partners, REHAU provides the complete aeration system service from design, fabrication, installation, commissioning and operation.

Tried and Tested: Installed at hundreds of wastewater treatment across the globe. Some of our recent clients include Sydney Water Corporation, Redland Water, Toou• Economical conversion of existing systems uoomba Water Alliance, Collex, Tauranga City Council, and many others. • High chemical and temperature resistance

For more information, please contact REHAU Pty ltd: 91 Derby Street Silverwater • NSW 2128 • Email: elliott.burke@rehau.com • Tel: +61 2 9748 1788

Construction Automotive Industry

wastewater treatment process co mbines a single activated sludge biological treatment process with an integrated, immersed membrane system. A basic description of the flow path is shown in Figu re 6 and the main process seeps include: Seep 1 - The existi ng equalisation basin feeds a 3-channel Orbal System. The Orbal system is designed ro provide biological nutrient removal. Seep 2 - The mixed liquor from the Orbal is fed through a micron screen. The screen removes crash and inert solids from the mixed liquor minimising the rocal amount of biological sludge removed from the facility. Seep 3 - Once screened, the mixed liquor flows co the membrane tanks th rough che patented MemJec technology, which combines mixed liquor with air co create a uniform MLSS co ncentration throughout the MBR basin. Step 4 - Effluent leaves the MBR system with very low turbidity and cransmissivicy and sent co an existing UV system fo r fi nal disin fection. With the addition of Memco r MBR, the Calls Creek WWTP achieves:

• increased capacity from 1.5 MLD co 2.5 MLD; • future expansion capacity for 4 MLD fo r commercial and residential growth needs; • superior biological treatment; • improved turbidity and cransmissivicy, chus no UV system expansion requ ired; and • decreased sludge waste prod uction. The typical effluent quali ty from che Calls Creek MBR plant is summarised as: <2 mg/L BO D N H3-N <0.5 mg/L TP <0. 5 mg/L T SS: <2 mg/L Coliform <2 cfu/l00mL (before disinfection)

Conclusion Through years of research and develop ment, Memcor has successfu lly applied its MBR produces in to the wastewater market. With distinctive feat ures, the Memcor MBR produce is designed co be sui table for the treatment of wastewater from small packed systems co large scale planes. The Memcor R&D team continues co work on improving the tech nology co achieve high quality created

effluent at the same ti me further red ucing coses.

References I. K. Yamamoto, M. Hiasa, T. Mahmood and T. Matsuo, D irect solid-liquid separation using hollow fibre membrane in an activated sludge aeration tank, Wat. Sci. Tech., Vol. 21, p43, 1989. 2. T. Ueda, K. H aca, Y. Kikuoka and 0. Sei no, Effects of aeration on suction pressure in a submerged membrane bioreactor, Wat. Res. Vol. 3, P489, I 997. 3. K. Alexander, B. M cBride, R. Jackson and J.Wade, Membrane bioreacror design: problems and solutions for a plant upgrade in Ant hem, Arizona, Proceedings of WEFTEC 200 I, Session 3. 4. David Evans, "Recycled water plant on line at North Head STP" Media Release, December 6, 2005 . 5. Stephen Chapman and Ian Gabriel, "The Application of Membrane B ioreactor (M BR) in Australia", AWA Operator Conference, Queensland, November 2005.

The Author Dr Fufang Zha is Global Director of Wastewater Process Technology fo r Siemens Water T echn ologies - Memcor Prod ucts, 1 Memtec Parkway, Windsor 2756, Australia, Email: fufang.z ha@siemens.com


www.forearth:coro.au Aerobic Primary Effluent Pond

QLD, NSW & VIC FOR EARTH PTY LTD Alan Mckibbin / Shane Mckibbin Tel: 02 65814353 Mob: 0417 694 844 Email: alanm@ecogreen.net.au Web: www.forearth.com.au

WA, Environmental Products Australia Pty Ltd Peter Spry/ Gordan Motherwell Tel: 08 9347 3300 Mob: 0403 578 684 Email: gordan@ecogreen.com.au Web: www.ecogreen.com.au



MARCH 2006


wwwm . emcor.usfilter.com


WHY SMALL TOWNS ARE CHOOSING MEMBRANES A Patterson Abstract This paper reviews rhe success of low pressure membrane filtration in offerin g a simple, low-cost, low-maintenance technology for ensuring clean water especially at sires with huge variations in source water rurbidiry. The technolo gy has proven itself reliable enough to be installed in remote sires, demonstrated in analysed case studies. In some re more sires local people operate the plants with visits from a service person once or twice a year.

Introduction Over rhe past decade, the requirements fo r drinking water quality in Australia have seen many changes. The awareness of risks like pathogens, narural organic matter and algal toxins has increased and so too the technologies used for drinking water treatment have changed . Ir is co mmonplace now to see cities and large municipalities choosing large scale low pressure memb rane systems for their drinking water treatment to remove particles and pathogens. At rhe same rime some smaller co mmunities have been raking advantage of low pressure membrane filtra tion to reduce their risk while reducing cost o f treatmen t. As more local authorities beco me aware of the risks of pathogens and other poll utants in their water supply, they are demanding a small scale system that offers the prorecrion rhar a membrane treatment can provide together with a robust process suited to remote locations and changing water condition s. This article looks at the particular requirements of a smaller, remote water treatment plants and how current membrane systems h ave evolved to suit this application.

Why the push for improved, small membrane treatment systems? 1. Equity - Governments must face th e questio n of whether small communities deserve the same drinking water quality standards as those who live in cities. Around the world governments are committing themselves to improving the d rinking water for indigenous communities, especially where these communities are outside metropolitan areas. In Australia too, the pressu re is on to

122 MARCH 2006 water





0 09


0 00





Feed water turbidity varying between 2 to 80 NTU

0 07

Clean water (fi ltrate) turbidity stays less than 0.03 NTU

O (I;

"' €::,


0 05



0 04


[ ~


,.'2.. ::,





0 03


0 02


0 01


o 0,/19,03








Figure 1. Feed Water Turbidity and Clean Water (Filtrate) Turb idity from a Memcor® CMF-S plant. provide all Australians with safe drinking water. 2 . H ealth concerns. Research is continually highlighting new concerns for health in water so urces. Anecdotal data from health professionals suggests that people in small communities suffer from more water- borne d iseases than their city counterparts, bur epidemiological data is hard to obtain as o nly 5% of these illnesses are believed ro be reported.

A robust process suited to remote locations and changing water conditions. 3. Cost. New technologies and product improvements have meant the rapid reduction in rhe cost of membrane systems. Bur it should also be recognised that influential bodies like the World Health Organisation calculate that that the economic and health cost benefits resulting fro m safe d rinking water ou tweigh rhe costs o f provision by a facror of 3 ro 5. 4. Simplicity. M embranes p lants sometimes come with the repu tation of b eing a 'high tech ' solution bur in reality their op eration is quite simple wh en you compare against th e amount of training and operaror

attention required ro run a conventional drinking water plant like a coagulationsand fi lter plant.

What are the challenges posed by small, remote plants? 1. Local authorities are sometimes not equipped with rhe resources ro manage their system, identi fy and manage risks and select the right treatment solution. Local authori ties have wide areas of responsibility and for small capacity systems its di fficult ro justify the fees ro engage consultants to evaluate and select technologies. 2. Remote locations can often be dependent on the one water source and sometimes th ese single water sources can be ' flashy' with sudden h igh turbidity periods or seasonal algae events. Water sources in rural areas often have agriculrural activities close ro the catchment area which can lead to an increased risk of pathogen contam ination and even p esticides. These types of issues add to th e complexity of choosing th e treatment.

What are the advantages of membranes? M embrane systems have demonstrated their ability ro cope with changing feed water q uality without impact on rhe fi nal water quality. O ne of rhe strongest, or ni che, app lications for MEMCOR® CMF, CMFS and CM F-L Co ntinuous Membrane Filtration products, has been their ability ro

Creating Water Solutions Present in Au stralia for over 70 years, Veolia Water Solutions & Technologies is a world leader in the Design & Build of water and wastewater treatment plants. Our unique portfolio of differentiating technologies includes drinking water, industrial process water, and wastewater. We provide Design & Build, Engineered Systems and Services capabilities, focusing on

reuse and desalination. Our approach for delivering innovative w ater solutions makes us the partner of choice for municipal customers as well as the Pulp & Paper, Mining, Power and Food & Beverage industries.

Visit us at ENVIRO 06 in Melbourne, May 9-11, booth e232

Veolia Water Solutions

& Technologies

Tel. + 61 2 8572 0400 I Fax: + 61 2 8572 0410 systems@veoliawater.com .au


G)veOLIA WATER Solutions & Technologies

membrane technology operate o n "flashy" feeds. T hese are water sources chat are generally good quality with less than 10 NTU feed turbidity bu t are subject to short term turbidity spikes up to and even over 1000 NTU. At numerous installations it has been demonstrated that the process is robust and able to handle the feed variatio ns whilst still maintaining a consistent created water quality, as illustrated in Figure 1. The approach of the Australian D rinking Water Guidelines has always been to provide a multi ple barrier app roach to water treatment. Similarly in the US, the Long Term 2 Surface Water Treatment Rule applies the same methodology when crediting different technologies with removal of pathogens. M emcor's low pressure membrane products have been credited with log 4 removal o f pathogens by the California D epartment of H ealth Services (CDHS) and this rating is accepted by many ocher regulatory bodies around the world. This removal is also able to be regularly verified by in tegrity testing. Using the MEMCOR Pressure Decay Test, the LRV (Log Removal Value) of the plant can be quantified before you fil ter water to guarantee pathogen removal. This cannot be d one with a sand fi lter.

Case Studies Argyle Diamonds, WA The very first large range MEMCO R® CM F plant was installed at A rgyle Diamond M ines in 1991 and is still in operation today. l e treats water chat duri ng flood periods can have turbidities up to 200NT U and 200mg/L suspended solids. Argyle Diamond Mines is located in one of the most remote and inaccessible locations in Australia. It lies in the tropical northwest of Austral ia, and is exposed to high wind loadings (resulting from tropical cyclo nes) and an average daily maximum temperatu re of 36°C.











Figure 2. Process Flow Diagram from Argyle Diamonds Mine Water Trea tment Plant, WA.

As the plant is located 35km fro m the mine site, a telemetry control system allows the plant to be controlled remotely. D uring construction of the WTP the C M F unit supplied was a pre-engineered skid mou nted packaged unit. The unit was facto ry built and tested thus min imising complexity of installation and reducing the time requi red onsite for installation and commiss ioning.

Collarenebri WTP, NSW The Collarenebri Water T reatment Plant has a capacity of 0. 5M L/day and supplies drinking water for the town of Collarenebri in NSW. It was built for the NSW D epartment Public Works in 1996. The feed water is the Barwon River (NSW) and the turbidity, though said to be typically 20

NTU, can regularly spike up to 40 0 NTU. High turbidity alo ne doesn't d escribe all the problems with this feed water, the suspended solids themselves are very fine, 96% less than 1 micron in size, making them difficult to filter by conventional means. The small WTP plant at Collarenebri contains a single MEMCOR® 20Ml OC C M F unit which is 20 membrane modules connected in a block or skid. The raw water is filtered down to 0.2 micron by the membranes, which then fl ows into six man ually backwashed GAC beds that remove caste and odour. The water is chlorinated and then goes to clean water storage. The plan t is serviced regularly by Veolia Water Systems (VWS) service personnel. The challenge at this plant was

Fresh water for both the process and potable needs of the site is pumped from Lake Argyle, a large man-mad e lake about 35 km from the mine. The water is requi red to supply the potable water needs of 300-400 site personnel. The raw lake water is pre-screened through a 500 micro n strainer before entering a feed tank. Water is then fed to the MEMCOR® 60Ml0C CMF uni t by a pump. T he CMF unit produces a filtrate virtually free of suspended solids and bacteria. This filtrate is fed to a large filtered water tank and dosed on-line with ch lorine prior to delivery to service. Filtration is achieved withou t the use of floccu lating chemicals (such as alum).

124 MARCH 2006


Figure 3. The Laura Water Supply and Treatment Plant, showi ng the aeration tower and tank outside the main shed.

the changing nature of the feed water. Prior the installation of the membrane plane, the town had used conventional treatment, bur the bypass that occurred during these wrbidity spikes meant that the plane was no longer removing pathogens to an acceptable level - they had to go for a barrier technology. to

T he MEMCORÂŽ 20MIOC plane provided valuable experience in such a difficult application which fed back to Memcor product development - our of which came the Intermediate Packing Density (!PD) or Ml0D module. T he IPD is a module specifically developed to cope with high solids conditions at the same time lowering operating costs and exrending the membrane life. In 2001, the Collarenebri plant was upgraded with these new modules. The performance since the upgrade has validated the new design and led to the sale of more planes using IPD membrane modules in high suspended solids applications.

Laura, QLD The water treatment plant at Laura is another example of a membrane installation adapting to the needs of a remote site.

Figure 4. One of the CMF-L 6L l OV units (right) from the Laura Water Treatment Plant and the computer controlled pumps that supply the reticulation system.

Laura is a township of approximately 140 people, located in Cape York Peninsula in Far North Queensland, and the plant is shown in Figure 3. The project to build a water treatment plam at Laura was fully

supported by the Department ofN awral Resources, the Department of Families, Youth and Community Care, Ang-gnarra Aboriginal Corporation, and Cook Shire Council. The entire project was designed,

membrane technology built and constructed by Veolia Water Systems. The sire is several hours drive from the nearest large town and is frequencly cut off by road during rhe wet season. One of the first challenges was the lack of a three phase power supply at the site. Two small MEMCOR® CMF 6M l 0V units were installed, each only requiring a single phase power supply. The bore water char supplies the p lant is high in iron so there is an aeration cower and tanks to oxidise as much as possible of rhe iron befo re rhe membranes.



~ .



3C, 95 Metret







1610112006 11:21 AM

( Ail ........ Dll

A.etat_,.,. ,i::11

Tc,pT. .

Cook Shue Counars Laura Wator Treatment P1ant



Wain lank

{] 1.32 N.ties8-4 2:

i]{] _[] CH,,.,.

companies or the membrane supplier. T he site can be visited every six months to a year by experien ced technical experts to help support rhe operators and provide operational assistance.

Ano ther of th e key factors in rhe stress-free operation of these ~ IJjM-.HMfotJ If.i1• _,,I mgl!ai1n planes is rhe constant quality of I ll"' 1.24 Met,es832: 2.02: NMIHS-t 2: 2.l2W14fet 1002: the water produced, regard less Disital Out Oft'• AwC..a.l O. • S-,All Oa• Off• A•tfC.111tnl 1,,...,An C.1ttnl 0 means }fish or On of the feed water conditions. By Force Orr Fore• On example, the Cape York Se..t Pw4:, D 0 OFF Q ~ 0 OFF Q ON D ,.,_.,.,_. 1 l'reffllNl,....l comm unities are subject to the BerePIUllipl 0 OFF Q ON 51..-l'li.N, 0 OFF Q ON D D ,,,__,....l 0 Off Q OH 0 OFF Q ON Pft,......... l very d istinct Wet and Dry 0 OFF Q ON llotd 8..-to,n Oep110, 1.d hw 10 0 OFf Q ON hNlfulpl Benl"a.p2 SIICOl'ldalfAotoR.,.,ing Pft.,,.,.,....,J seasons. During the Wee rhe raw l'ftttv. ...... :J 0 OFF 0 ON 0 OFF Q OH 0 0FF Q ON 0 OFF Q OH water conditions deteriorate, C~rliuitr ,_,.,.,..,4 Pnnve,....4 Clrlltdu.ttr 0 OFF Q ON 0 0FF Q 2 0 OFF Q ON 0 0FF 0 P. changing rapidly in rhe wet Figure 5. Sc reen Shot from the Remote Monitoring System for conditions prod ucing turbidities Similar to the situation at the Laura Water Treatment Plant. in excess of 1000 NTU. Collarenabri , in 2005 the M emcor's membrane memb ranes had reached the end technology is used on several of their life and the plant was sites in chis region, including the dam on Oscar C reek. T he dam is an upgraded from CMF to rhe new CMF-L the water rrearmenr plants at Bamaga, open ' turkeys nest' and chis water becomes p rocess. The two CMF 6Ml0V units managed by Sunwater and W ugal Wugal, very heavily fouled with algae and the became CMF-L 6Ll 0V utilising the laresr managed by rhe local community. As was natural arsenic levels are high. Ar the Coen membrane modules and membrane shown in Figure 1 raw water turbidity has WTP the arsenic is removed by constantly materials, thus reducing the operating no impact on rhe filtered water quality aerating part of the dam, causi ng the coses (see photo Figure 4). Part of the from the membrane system, so no matter arsenic ro drop to the bottom. The feed is product development cycle at Memcor is what rhe season, the operator knows the drawn from close to the top of the dam. co enable users co upgrade to new end user is protected. there is a Prior to rhe membranes technologies and cake advantage of cost conventional pressure filter with alum What are the benefits for local red uctions. dosing for colour removal as an extra authorities? The Cook Council, who operate the plant, barrier to pollutants. This clarified water Local authorities receive rhe benefits of: do not have an operator on sire. Robert goes co a MEMCOR® CMF 16Ml0C Fenn, from the Cook Council, monitors • Higher quality, lower risk potable water plant as a barrier against pathogens. There rhe plant remotely by telemetry from the for their customers; is also a DAF unit upstream used in rimes office at Cooktown, 150km away. Figure 5 • A technology chat constanrly improves when algae concentration is high . and Figure 6 show examples of the to not only meet today's requirements but Th is fairly comp lex plant is operated by monitoring screens from the control also future requirements , and; one person, with the CMF plant caking system for the Laura plant. Fenn organises • A system that is simpler and ch eaper to the least operator attention of all the unit a monthly visit where they do the CIP operate. operations on the sire. (clean in place) for the membranes, take Pathogens are o ne of the biggest risks to These case studies were selected from water quality samples, and any ocher jobs people in these communities. Typically the that have come up. The plant is visited by Australian experiences, but the technology rivers char form the raw water source are experienced service personnel from has been applied in the same way all places where pigs, stock and native animals Memcor or Veolia Water Systems annually around world like the US, Europe and both drink and d ie. The pathogen and co ch eck the operation and offer technical Asia. coliform counts of the water can often be advice. What are the benefits for operators? very h igh. Though tri-halomethane Coen, Qld · A complex water source The plants highligh ted in chis article are p recursors have been much discussed in recent years, it is Cryptosporidium and While removing solids and pathogens can managed by different bodies - local Giardia chat are proving to b e rhe greatest d rastically improve the water quali ty, councils, water consortia, government risk, and have still to be addressed in many applications arise where membrane entities and industry. The operators all fi ltratio n is nor enough. Water sources remote parts of Australia. have varying backgrounds and education levels. T he automation chat a membrane that contain a signifi cant dissolved Membranes have a finite life and at some component like colour and metal salts plant provides means there is minimal day srage need co be replaced. However this to day supervision co operate a membrane require additional processes. One way of gives rhe owner the opportunity to plant. The control systems self monitors cackling these issues is with enhanced upgrade their plane, taking advantage of coagulation, allowing the coagulant to and prompts the operator. Th e expertise membrane technology advances char have required to operate a membrane plant is bind up the dissolved species in solid fo rm occurred since the first membranes were and allow them to be filtering chem out by often less than even simple conventional installed. Observers of rhe development of rhe membrane. systems. low pressure membrane produces for The community at Coen, also in Far North Queensland has two water sources to choose from, the Tankelly Creek and

126 MARCH 2006



Operators can also be backed with tech nical support through service agreements provided by specialised service

, . . . Ill

......... ....,...,,

d rinking water will have seen the rapid development of membranes over the past 20 years, producing:

• higher flows; • lower cost membranes; • better chemical resistance; • new processes chat are cheaper co operate, and; • require fewer resources. For small WTPs the major operating cost is the operational labour content. Unattended or minimal attendance at membrane planes reduces the nu mber of operational staff required. In certain regions these systems can also allow one operator co have respo nsibility for a network of planes.

The Future for Small Communities

Memcor's experience wich remote locations shows that MEMCOR® CMF, CMF-S and CMF-L planes offer a simple, easy co run installation, robust enough co cope with varying feed water qualiry.


Figure 6. Trend Graphs from the Remote Monitorin g System for the Laura Water Treatment Plant.

It is reasonable co expect that all Austral ian communities should have good qual ity d rinking water. le seems that the more we learn about drinking water supply, the more we identify new concerns co publi c health ch at need co be managed. In spice of ch is, many small co mmunities still have co

T he author would like to acknowledge the valuable help of the following people in researching this paper: Mr Tony MacCormack, Mr Wayne Hislop, Mr Craig Chamulko and Mr Mark Thompson of Siemens Water Technologies, Memcor Products Asia; Mr Daniel Brown of Veolia Water Systems; and Mr Robert Fenn of Cook Shire Council.

cackle the well-known issues in their drinking water.

The Author

Low pressure membrane filtratio n is a widely used and globally recognised technology that provides high qual iry water and greater chan 99.99% removal of Cryptosporidium and Giardia.

Alexandra Patterson is a Senio r Developmenc Engineer in che Research and Development Department of Siemens Water T echnology, Memcor Prod uces. alexandra. parrerson@siemens.co m

Leaders in Water Management \i1aunsell, one of 17 global AECOM operating companies, is a leader in the development and management of water systems ,ind water businesses, and is the fastest growing engineering service provider in the Asia-Pacific and Middle East region. Jur services in the water sector cover the lifecycle of infrastructure from planning and advisory services through to asset T1anagement. The calibre of our people combined with the strength of our internal systems has resulted in award winning, ::ost effective design of many of the region's iconic water projects. Nith more than half a century of experience in the water industry, Maunsell taps into a proud heritage to provide advanced ,olutions for the water industry in areas of water reuse, sustainable water management, modelling and master planning, and jesalination.

{isit www.maunsell.com



Treated gas


O dour has been a prominent issue for decades and with residential development encroaching on wastewater treatment plants it is becoming more significant. T here are many technologies available to either prevent odour generation or to remove odour from foul air streams. Some technologies h ave very limited applications and others can be used with success in a wide variety of situations. T he most commonly used fou l air treatment technologies and key facto rs in selecti ng the right tech nology for a given application are the subject of this paper.


Caustic soda solution

Bleed line Sewer gas Overflow Scrubbing column

Introduction Odour is caused by the presence of one or more compounds char exist (or are produced) naturally in sewage, and migrate from the liquid phase into the gas phase (air) . Sulphides, mercapcans, d isulphides and volatile fatty acids all contribute to odour. Naturally occurring bacteria u tilise different metabolic pathways in the presence or absence of oxygen and other electron acceptors such as nitrate. Hydrogen sulphide and other odorous compounds are formed as by-products under anaerobic (zero dissolved oxygen and zero nitrate) conditions. H ydrogen sulphide is often the cause of odo ur complaints from sewerage systems and treatment plants. Sulphide generation typically occurs under che following conditions: • H igh BOD (biochemical oxygen demand) of rhe wastewater • Warm weather • Long reten tion rimes in sewers • Low dissolved oxygen concentration of the wastewater.

Odour Control There are two primary mechanisms for odour control - preventative and curative. Prevenrative technologies work to prevent the generation of hydrogen sulphide in rhe system and typically include dosing of chemicals. Many of the preventative technologies provide only temporary

128 MARCH 2006

Liquid pump


Make up water

Figure 1. Chemical scrubber process.

control of sulph ides and dosing must be provided at several points in a system however they have the benefit of protecting the sewer from hydrogen sulphide induced corros10n.

A critical review of available technology. Common chemicals used and how they work include: • Magnesium hydroxide - Raises the pH to approximately 8.5, which increases the solub ili ty of hydrogen sulphide and decreases the level of hydrogen sulphide in the gaseous phase. • Nitrate - Prevents the formation of the anaerobic state under which sulphide generation occurs. • Oxygen or hydrogen peroxide - Oxidises sulphides and so prevents the generation of gaseous hydrogen sulphide. • Iron - Binds sulphides in an insoluble iron sulphide compound. C urative tech nologies work by removing the hydrogen sulphide (and other odorous

compounds) from a fou l air stream and include chemical scrubbers, soil bed fi lters and activated carbon. Fo r municipal wastewater applications where hydrogen sulphide is the p rimary odour compound there are really only five curative type technologies commonly used. These are briefly described below. There are other technologies including ozonation, wh ich oxidises most odorous compounds and is most commonly used to treat the small foul air fl ows at sewer pump stations, and catalytic incineration, where the odorous compounds are converted to carbon dioxide and water via a catalyst at temperatures of between 300°( and 400°(, which have more limited application d ue to their high capital and operating costs and high use of power.

Chemical scrubbers C hemical scrubbers are packed towers that u tilise chemicals to remove/oxidise contaminants. W hen used for odour control, they typically have counter-current fl ow (i.e. air is passed in the opposite direction of rhe liquid sprays). The tower is filled with a packing material char provides a greater surface area for reaction. Foul air passes upwards through the column and

rerereed paper

odour management con tacts the scrubbing chemical as it trickles down. Odorous compounds dissolve in to the scrubbing chemical and are oxidised to less odorous or odourless compounds. T he process is shown diagrammatically in Figure I. Typical scrubbing chemicals are sodium hydroxide and sodium hypochlorice. In recent years peroxide has been used as a scrubbing chemical co remove hydrogen su lphide from the sewer gas because it does not generate organo-chlorine compounds, many of which are carci nogenic and difficult co biodegrade in the downstream treatment system. T he choice of chemical is dependent upon the compound to be removed.

Sprinkler systcn

Mo.kc-up water Pl11s1i

edia with biologicol


Sewer gas Blov~-cr

Leachate collection to droin or to sump

The advantages of ch is system include: • Immediately effective • Up co three cowers can be con nected in series to treat large odour loads

Figure 2. Bioscrubber system.

• Capable of creating high concentrations of hydrogen sulphide

• Does not treat organic compounds

• Capable of creating high volumes of fo ul air • Can remove over 99% of the hydrogen su lphide in the air. T he limi tations of chemical scrubbers include: • High capital cost • Mediu m co high operati ng cost (dependent on load) • Requi res skilled operators (complex con trols) • Careful control of chem icals is requi red under fluctuating odour concentra tions

• Presence of chemical odou r when hypochlorite is used • Safety hazard in transport and storage of chemicals.

Bioscrubbers Bioscrubbers are often vertical cowers chat utilise mi croorganisms and biological processes co remove hydrogen sulphide (and ocher odoro us compounds) from sewer gases. The microorga nisms are immobi lised on a plastic packing media con tained in the cower. The scrubbing water contains organics and provides food and nutrien ts for the growth of the bacteria as it trickles

n for the wate1;

down the cower. The fou l air is passed up ch rough the cower, coun cer-current co the scrubb ing water. Odorous compounds dissolve inco the scrubbing water and are ox idised into odourless compou nds by the microorganisms, which develop and grow on the packing media surface. T he system is shown schematically in Figure 2. The removal efficiency is dependent on the gas retention time, maxi mum load ing race, pH, characteristics of che odorous co mpounds and successfu l development of microbial growth on the packing. Temperature and rype of packing material used are also influential.

'l:;ater and environment !

011 water monitors, c produced dig ital;


SCA DA and 5ystems for remot --1" process control problem.


ontrols yo , c n be assurred of our fullest attentio11

u/1 backu~ nnd

.,,.,;c, on all p,oduct, Australia.

Ph; 6173376 4086 Email; act.inst.salesfiiozemail.com.au Fax; 6173376 3192

refereed paper


MARCH 2006 129

odour management Bioscrubbers can also be used as a pretreatment co technologies such as activated carbon, d ry scrubbers and biofilcers co treat fl uccuacing loads and increase removal efficiency. The advantages of this tech nology include: • le is less labour intensive than a chemical scrubber • C hemicals are not used • le has low operating coses. The d isadvantages of bioscrubbers include:

A ir distribution


Figure 3. Biofilter system.

• Medium to high capital cost • A bioscrubber will typically generate an inherent residual odour • Noc suitable for highly fl uctuating loads (except as a pre-treatment scrubber system).

Soil bed filters (biofilters) T he soil o r compose biofilter is a bulk media biological scrubbing system, which utilises microorganisms to remove odorous compounds from fou l air. The media su pporcs a microbial pop ulation, which muse be kep t reasonably moist to ensure p roper functioning of the soil beds (microb ial viab ility) . Overwacering needs to be avoided as ic can cause large pressure d rop across che bed and can also result in the development of anaerobic cond itions within the bed. Under watering, resulting in too liccle moisture can dry the bed out, k ill microbes and allow breakthrough of che bed. The med ia can be made up of a m ix of materials (wood chips, topsoil, peat, sand, shells, etc; stable biosolids from wastewater treatment planes can also be used). Foul air is d istributed into the bed via a subsoil plenum arrangement, and is passed through che soil, which, like the bioscrubber, supporcs a large pop ulation of microorganisms. Compounds d iffuse into che moisture/water layer, where they are degraded by the microbes. Figure 3 illustrates a typical the bulk media biofilcer system.

• Dependent on load to the biofilcer, media o nly needs changing every 4 - 6 years. The limitations include: • Medium capital cost • Suscep tibility to fluctuations in odour load (although will consiscencly achieve stable percentage odour red uction) • le is not approp riate for hydrogen sulph ide concentrations above 20 ppm • A large surface area is required.

Dry scrubbers

• Can tolerate high humidity in the sewer gas without affecting efficiency.

T he scrubber is able to remove hydrogen sulphide, organic sulphides, mercaptans and amines with a combined concentration of 10 ppm or less. When the gas concentration is higher, it is recommended to use a wet scrubber to remove the bulk of the odorous gas and che Purafil ESD drum or cub

• Simple treatment process. • Can remove greater chan 99% of mercaptans, organic nitrogen compou nds and sulphides from the air.

• Handles a wide range of loads. • Medium capital cost. The limitations of chis technology include: • High operating cost - che operating cost increases with the q uantity of odorous compounds in the air. • Disposal o f spent media. • Bed depth and d iameter are limited so che flow may need to be split co multiple units .

Activated carbon Activated carbon is utilised to adsorb and treat odorous compounds. It can be made from a variety of carbon sources including wood and coal. The highly porous carbon particles provide a large surface area fo r in teraction and reaction.

• High reductio n of sulp hur compounds


• Simple co operate


These systems are typically dual media and che media is selected based on the contaminants co be removed. For sewer gas, where che main contaminant of concern is usually hydrogen sulphide, the gas would first be passed th rough media designed co remove chis compound. A second media would be used to remove ocher contaminants such as mercaptans.

T he Purafil drum or tub scrubber contains activated alumina, sodium bicarbonate, potassium hydroxide, potassium permanganate, chemical binders, activated carbon and water. The scrubber is able to work in a h igh hu midity environment and has a relatively low pressu re drop.

• There is no chemical requirement

130 MARC H 2006

Figure 4 is a schematic of che Purafil drum scrubber system.

T he advantages o f the Purafil d rum o r cub scrubber technology include:

The advantages of the biofilter include:

• Low o perating costs

scrubber to po lish che created gas from the wet scrubber so chat the system can be cost effective.

Dry scrubbers do not require either water or chemicals for operation as do both chemical scrubbers and bioscrubbers. An example of che d ry scrubber tech nology is the Purafil d rum scrubber for low air flow races or cub scrubber for high fou l air fl ow rates.

Biofilcers are capable of removing many odorous compounds, such as ammonia, hyd rogen sulphide and volatile o rganic carbons (VOCs). T he odorous compounds are removed via oxidatio n by the microorganisms present in the bed. Odou r removal efficiencies of up co 98% and hydrogen sulphide removal up to 99.9% can be achieved with good design

• Suitable for application to a wide array of odorous compounds


Figure 4. ESD drum sc rubber process .

Activated carbon aces as both a catalyst and an adsorbent. Organic compounds such as aldehydes and skatoles are physically adsorbed while hydrogen sulphid e and mercapcans are oxid ised to less odorous compounds on the carbon surface. Activated carbon can also be impregnated with ocher compounds such as sodium hydroxide to improve removal efficiencies.

efereed paper

odour management Untreated activated carbon is generally thought to remove approximately I 0% by weight of hydrogen sulphide, although treated activated carbon can achieve much higher levels of removal. T he basic process involves passing foul air over the activated carbon and ensuring an adequate residence time for adsorption and reaction. Recommended contact times range from 0.5 to 4 seconds and bed depth is typically between 30 cm and 90 cm. Given adequate contact times, activated carbon beds are capable of removing up to 95% - 99% of general organics and sulphides. Due to rhe high cost of activated carbon, it is not generally used for hydrogen sulphide concentrations greater than 3 ppm. T he humidity of the sewer gas fed to the activated carbon fil ter should be co ntrolled to less rhan 60% for efficient operation. Otherwise, the capacity of the carbon bed will reduce significan tly.


Activated carbon bed

Sewer air Drain

Figure 5. Activated ca rbon process.

• Handles a wide range of loads • Low capital cost.

T he advantages of this technology include:

The limitations of this technology include:

• Well-known and simple treatment process

• High operati ng cost - the operating cosr increases with the quantity of odorous compounds in the air

• Can remove 95 to 99% of organics and sulph ides from the foul air

• Disposal of spent activated carbon

• Regeneration/replacement of carbon every 12 - 18 months dependent on load and carbon volume • Bed depth and diameter are limited so the flow may need to be spli t to multiple units • Dehum idification of sewer gas is required for efficient operation in most cases.


The CDS Nipper delivers giant performance in a pint-size polymer unit. Because it's made from tough polymer, it's easier to handle and install and with no moving parts it goes on performing for years . The Nipper is easy to maintain with cleaning undertaken by suction vehicles. Gross weight 200kg Risers available to allow depth of 2.7m to pipe invert Easy installation for all pipes up to 375mm diameter For use in systems where pipe and fittings under 100mm are not permitted Captures up to 95 % of pollutant particles over 1mm For catchments up to 1 ha Trafficable and non-trafficable versions available

Model: PL0506 CDS Unit - The Nipper

refereed paper


MARCH 2006 131

odour management Selecting a Suitable Technology for Foul Air Treatment Key factors rhar determine rhe basic suitability of a technology to a particular application are coral input odour level (typically hydrogen sulphide concentration), foul air flow rare and odour reductio n target co be achieved. Certainty of the first two factors is essential in the selection of a suitable technology fo r foul air treatment. The foul air flow rate can ei rher be d etermined using a suitable number of air changeovers p er hour for the volume or from an existing fa n size. T o d etermine the odour level for an existing system, hydrogen sulphide monitoring should be conducted over a minimum two-week period at each location requiring odour treatment. The monitoring will provide information on the average and peak hydrogen sulphide levels reached fo r each location and on rhe variability of levels. For n ew systems, rhe hydrogen sulph ide level may be esti mated through sulphide generatio n modelling. Whilst nor essential iris also beneficial ro co nduct plume modelling co ensure char rhe created foul air odour level will be sufficiently diluted in the atmosphere raking into account the local built environment and weather conditions co meet the desired odour level at the boundary or n earest residence. Plume modell ing can assist in determining the required treatment removal efficiency.

Input Odour level / Foul Air Flow Rote low/low


High/ Low


Dry scrubbers Biofilter Bioscrubber Bioscrubber + AC/ dry scrubber/ biofilter Chemical scrubber

Figure 6. Schematic representation of the con ditions for which popular technologies are typical ly most economically viable .

Figure 6 is a simplified representation of the conditions for which each of the popular technologies, described in this paper, are typically most economically viable. The low input odour level would typically be less than approximately 3ppm, the mid between 3 and 20ppm and the h igh greater than 20ppm. The low fo ul air flow rare would usually be less than app roximately 2 ,00 0 cubic metres per hour. We note that the guidelines provided d o nor mean rhar a chemical scrubber, for example, is nor a good treatment choice for a low odour level - low foul air flow application. However it does mean that eg. biofilters are typically the more eco nomical choice fo r chis application neglecting all ocher considerations. T his general guide can

assist in an initial short-listing of the many technologies available. The other important factors for co nsideration are: • Footprint and available land area • Occupational health and safety • Availability of power and/or water supply co the sire • Proximity co sewer for liquid waste streams • Extent of operator availability. These factors can be used co either eliminate the short- listed technologies from further consideration or to differentiate between options during evaluation. For example the biofilcer may be eliminated due co space restrictions or rhe chemical scrubber for occupational health and safety reasons.


Water Water and not a drop will be delivered .... Without: Engineers of all disciplines, Project Managers, Contract Administrators, Designers, Asset Managers, Environmental Scientists, Procurement Managers, Commissioning Engineers, Instrumentation Specialists, Chemists, Community Liaison Consultants, the list is endless.

For your Water Focused Recruitment Specialist Call Peter Brown on 0418 746 964

Odour from wastewater treatment plan ts and sewerage systems is an increasingly significant issue for Water Authorities and Councils. There is a w ide array of technologies available to either p revent odour generation or ro treat odour once p resent. This paper has described so me of th e more commonly utilised technologies for fou l air treatment and provid ed some initial guidance in the selection and assessment of tech nologies suitable for a specifi c application. Input odour level and foul air fl ow rare are two key factors in the initial selection of odour control and ensuring maximum understanding of their values is essential.

The Authors


Jenelle Watson is a Senior Chemical Engineer and Robbert van Oorschot is Stmcgic Consulting & R«ruicmcnr

132 MARCH 2006

Mid/ High

Activated carbon



Mid/ Low


Manager Water Technology at GHD Melbourne. Email: robberr.van.oorschot@ghd.com.au

BIOTRICKLING FILTERS CUT THE COST OF ODOUR CONTROL G Finke shown tremendous success. Many of these "performance Biotrickling filte rs provide the Treated gas enhanced" biotrickling fi lters ability for creating a variety of have been employed overseas, waste gas streams in a very t::::::::==::j,___ although their application here efficient, low cost and in Australia has been relatively environmentally friendly Liquid new. man ner. T o determine whether recirculation Biofilters typically have fil ters of the application of biocrickling Inert pacl<ing organic materials, such as soil or fil ters is in face appropriate and 'l{\tQ _p!OJTlj\1$5' compost or of inorganic whether additional treatment materials. T hey rely on both stages are required using Waste gas absorption and adsorption of alternative technologies will Water odorous gases into rhe water depend on a number of issues. Nutrients phase and solid matrix co enab le These include characterisation of biodegradation by microthe waste gas scream to be Drain '---------' organisms. Biotrickl ing filters treated, the required ou clec Recirculation pump operate by the absorption of concentrations of odorous pollutants into an aqueous Figure 1. Diagra mmatic representation of biotrickling filter. species, as well as a number of phase which is recirculated other critical design and either continuously or operational considerations. Once Design and operational intermittencly over a packing material (refer these have been suitably dealt with, the Figure l ). The absorbed pollutants are likel ihood of achievi ng effective and reliable factors are critically oxidised by micro-organisms living on the performance will be greatly enhanced. packing material. Where intermittent discussed. recirculation is employed, adsorption onto Introduction the biomass or the surfaces of certain media Si nce the early 1990s, signifi cant advances the poll utant requiring treatment (e.g. can also play a significant role in the have been made in the understand ing and hydrogen sulphide) together with removal of pollutants. W ith biotrickl ing use of biological treatment so that it now optimisation of the operati ng pH has filters, the control of pH is easily achieved offers the ability to treat a wide variety of waste gas streams in a very efficient, low cost and environmentally friend ly manner. Table 1. Advantages and disadva ntages of biotrickling filters compared w ith Biological treatment uses the natural ability biofilters. of bacteria to degrade air-borne poll utants. Disadvantages Advantages T he bacteria used are fou nd generally in nature and in the case of wastewater Lower gas-liquid surface area for mass transfer Smaller footprint and overall facility size treatment, many of the species required Effectively treats hydrogen sulphide concentrations Reduction efficiencies relatively law far exist in the process ranks used fo r treatment pollutants with low solubility and high volatility greater than 20 ppm without problems of of the liquid waste. Two of the most acidi fica tion typical of biofilters common biological treatment systems are Nutrient dosing required (or use of suitable Effectively treats acid producing pollutants biofilters (this incl udes soil bed filters) and reclaimed water) biorrickling filters (i.e. a fi xed film Reduction efficiency may be affected to a pH readily controlled bioscrubber). greater extent by hig hly varying loads W hile there has been much written about Overall odour reduction efficiency tends to be Lower occurrence of toxic concentrations within biofilters over the last 40 years (Finke, lower where a multitude of odorous compounds water phase 2004), there is relatively little informatio n are present on the application and performa nce of Med ia compaction less likely to occur biotrickling filte rs. In fact, biotrickling Clogg ing of filter media less likely to occur filters were first employed over 30 years ago Lower pressure drop using rock media in a similar configuration to wasrewarer trickling filters. Until Significantly longer media life recently, biorrickling filters were generally More rel iable operation operated at neutral pH with varying levels Where concentrations of hydrogen sulphide ore of success. However, the design of these relatively high, whole of life costs are less systems with specific bacteria applicable to


Entrainment separator




refereed paper


MARCH 2006


odour management and so reaction processes chat result in che formation of acids or ammonia and nitrates (through che oxidation of o rganic n itrogen compounds) do not create che same problems as in biofilcers. U pstream humidification of che air scream is not required nor are air filte rs (in most cases) because particulates and mi nor grease deposits are readily removed during blowdown of recirculati ng fluid. Table l p resents che main advantages and disadvantages of biocrickling filters compared with biofilcers.

Absorption into



Gas phase

Adsorption onto media surface (intennittent tricklinQ)

Figure 2. Mass transfer concept.

In order co determine whether che application of biocrickling filte rs is in face appropriate and whether additional polishing stages using ocher treatment technologies (e.g. activated carbon or biofilcracion) or pre-treatment are required will depend on a n umber of issues. W hilst two of the most important issues are appropriate characterisation of the waste gas stream co be created and an understanding of che ouclec concentrations of odorous species required, there are also a number of ocher critical design and operational considerations.

Micro-organisms and pentane), breweries (e.g. hydrogen sulphide) and within che tob acco industry. Many ocher compounds (e.g. aliphatic, aromatic, oxygenated and chlorinated hydrocarbons, sulphur and nitrogen based compounds) have been successfu lly treated by biotrickling filters. The requirement co treat these compounds is an ongoing p roblem for many industries, and further applicatio n of biocrickling filters is therefore expected in the near futu re. Figu re 3 shows examples of some biocrickling fi lter installations.

Principles of Biotrickling Filters

Packing material

In b iocrickling filters, pollutan ts are absorbed from che gas co the free liquid phase and then onto the biofilm (or directly onto che biofilm) covering the packing material. Depending on che packing material employed (e.g. carbon) and che state of the biomass, adsorption may also occur as presented in Figure 2.

A number of packing materials have been used in biotrickling fil ters such as lava rock, pall rings, structured plastic or stainless steel packings, clay and activated carbon based packings, and open-pore polyurethane foam (refer Figu re 4). Whilst lava rock has been used in many of the earlier installations, recent use of polyurethane foam has shown h igh performance which is due co che relatively high specific surface area (approx. 600 m 2/ m 3) and low pressure drop.

Organic compounds + Oxygen - - > CO2 + H2O + Salts + Energy + Biomass In the degradation of hydrogen sulphide, carbo n d ioxide is utilised co form acid and sales.

bacterial immobilisation, high water retention capacity co p rovide a moist environ ment for bio-accivicy, sufficient void spaces co reduce excessive bed pressu re losses and be of a light weight (Tay, et al., 2005). On the ocher hand, che packing m use also be strong enough co limit compression due co the weigh t of the packing material above it, the attached biomass and che recirculating water. I n this case, the required packing height may need co be split into a number of smaller packed bed sections separately su pported.

T he packing material employed should have a high specific surface area co achieve higher

Since the packing media is inert, ic will need co be inoculated with pollu tantspecific micro-organisms. Activated sludge has a wide variety of bacteria capable of degrading a mulcicude of compounds and ic is generally sufficient for easily biodegradable compounds or waste gas mixtures. H owever, selection o f che inoculum becomes increasingly important for compounds chat are harder co degrade. This may be achieved through the use of p ure cultures, either m icro-organism samples chat are enriched in the laboratory or samples from ocher biocrickling filters creating similar wastes. Use of che appropriate inoculum for che poll ucancs co b e degraded rather than simply allowing micro-organisms co adapt from, say, an activated sludge could dramatically d ecrease the time it cakes for high removal effi ciencies co be achieved. Operating experience for biorrickling filters at several wastewater treatment planes has shown chat acclimatisation of che packing with Thiobaciilus spp. cultures derived from activated sludge usually cakes 2-4 weeks (refer Figure 5). In contrast, systems which are not inoculated may cake three times as

Hydrogen sulphide + CO2 - - > H• + Salts + Energy + Biomass Full-scale biocrickling filters have been successfully applied co the t reatmen t of odours (principally containing hydrogen sulphide and ocher volatile organic compounds) from wastewater treatment planes and sewerage systems. Although chis has mainly occurred within che USA, Eu rope and Asia, a number of systems have recen cly b een installed in Australia. In ad dition co wastewater treatment, biocrickling filters have also been applied co the chemical processi ng industries (e.g. removal of styrene and resin vapours), textile industries (e.g. removal of carbon disulphide and hydrogen sulphide), oil seed extraction (e.g. hydrogen sulphide, hexane

134 MARCH 2006


Figure 3. Exa mples of biotrickling filter installations: vertical flow system, left, and horizontal flow system, right.

refereed paper

odour management long co achieve the required efficiency levels.


Key Design Parameters Concentrations

Because a recirculation stream is sprayed over the packing, bio trickling filters are able co contend with high pollutant concentrations which could cause toxic inhibition co micro-organisms (e.g. in biofilters). Inlet concentrations in the range of0.2-1 ,600 g/m3 can be treated (USEPA 2003). However, co achieve the required performance will need consideration of the allowable elimination capaciry of the poll utant- media combination. (Refer below).

Pa ll Rings (16mm) ecific surface area: 356 m2/m3


As is the case with biofilters, the vast majoriry of biotrickling filte r systems operate with mesophilic bacteria within the l 5-40°C temperature range. At higher temperatures, mass transfer from the gas phase into the liquid phase is generally limited whereas the biological oxidation rate is enhanced. Use of thermophilic micro-organisms within biotrickling filte rs has been successfully applied and furth er application is expected (Kennes and Veiga, 2001 ). Oxygen

Although oxygen co ncentrations are generally several orders of magnitude higher than the poll utant concentration, oxygen transfer into and di ffus ion within the biofilm may become limiting due co its low solubiliry. This will depend on its utilisation rate and the diffusion rates of both oxygen and the pollutant. At high pollutant concentrations, this possibiliry increases and al though oxygen requirements fo llow reaction stoichiometry, secondary processes may significa ntly contribute co the overall oxygen requirement. Therefore, oxygen requirements need co be assessed fo r each treatment application. Nutrients

The pollutant is generally the source of energy for the bacteria. Carbon is also provided by either the pollutant or carbon dioxide in the case of heterotrophic and aucotrophic bacteria, respectively. However, other elements such as nitrogen, sulphur, phosphorus, potassium and magnesium, together with other trace elements, are also required for cell

rerereeCI paper

HD Q-PAC (4x4) S ecific s urface area: 433 m2/m3

Polyurethane Foam Cubes (40mm cube) 3 S ecific s urface area: 600 m2/m

Figure 4. Examples of biotrickling filter poc king materials.

functioning. Since the packing material is inert, an external nutrient supply is required in order co provide these elements in excess and ensure effective biodegredation. Nutrient dosing is typically achieved through the use of a small storage rank and dosing pump. Liquid recycling Liquid recirculation is necessary co supply water and nutrients co the biofi lm as well as co remove reaction by-products which may be toxic co the micro-organisms. Recirculation rates are usually specified in terms of the superficial velociry through the packing and the gas- liquid ratio, which ranges between 0.5-20 m/ h, and 1001,000:1 (volume units) respectively. The recirculation rate needs to be high enough to remove potentially toxic by-products although the upper limit will depend on

flooding of the packing and will also need to accou nt fo r biomass accumulation within rhe packing. T oo high a rare may also increase the thickness of the water film over the biofilm and lead to increased mass transfer resistance. pH requirements

Most of the micro-organisms creating volatile organ ic compounds require neutral pH conditions to function effectively. However, some species of T hiobacillus (e.g. T hiobacillus thiooxidans and Thiobacillus ferroxidans) have shown an optimum pH range of 2-6 (Koe 2002). Currently, several fu ll-scale biotrickling filter plants are treating sewer gas in the 1-4 pH range whilst achieving removal efficiencies for hydrogen sulphide in excess of 99%. Where reduction of a number of compounds is required to achieve high odour removal rates, a co mpromise may need to be fou nd with respect to the normal operating pH. Alternatively, multiple packing sections (either combined or separate), each operating at a different pH (and hence supporting a di fferent bacterial population), may be necessary. Drainage

Figure 5. Scanning e lectron micrograph of Th iobacillus spp. (within biotrickling filter recirculating fluid ).

A constant bleed stream will need co be discharged


MARCH 2006 135

odour management from the biotrickling filter in order to comp letely remove reaction by-products, replenish fres h water supplies and control pH. T his stream is usually discharged from the sump through an overflow arrangement or by a solenoid controlled bleed-off line from the recirculation pipework. This drainage scream will need co be suitably disposed of especially if its pH is low in the case of acid producing biodegradation systems.

:2 <')


C: O>


.:, .?;-

Cl! · c: u ·Cl! E a. :.; co

LU (.)

Media sizing

EC max

Empty bed residence time, elimination capacity, substrate mass loading rate and removal efficiency are parameters typically used in the design and to report operation and performance ofbiorrickling filters. Each may be used on its own although it is prudent to consider all parameters together during design co ensure appropriate sizing and system configuration. Sufficient time muse be provided for the pollu tants co be absorbed and degraded within the packing media. While che accual residence time will be the vo id volume within the media divided by the ai r flow race, the void volume is not easily meas ured . Therefore, empty bed residence time (EBRT) is usually quoted in literature. EBRT = Packed volume of biorrickli ng filter (minutes or seconds) Bulk air flow race

Substrate mass loading rate (SMLR) is the mass rate of poll utant delivered to the biotrickling filter per unit volume of media. SMLR = Bulk air flow race x Pollutant concentration (g/m3.h) Volume of media Elimination capacity (EC) is the amo unt of pollutant chat the biocrickling fi lter is able to remove per unit volume of media and per unit time. EC= Bulk air flow race x (Inlet concentration - Outler concentration) (g/m3.h) Volume of media Many studies have shown chat an EBRT within the range of 10-60 s is generally sufficient to achieve suitable elimi nation capacities and good reduction efficiencies for easily biodegradable compounds such as hydrogen sulph ide and carbon d isul phide (USEPA 2003), (Gabriel 2003), (Kraakman 2004).

Decease in inlet concentrations

Substrate Loading Rate (g/m3.h)

Figure 6. Typi cal elimination capacity versus load curve. concentrations, with perfo rmance increasing proportionally with rhe loading (refer Figure 6). Ac some po int, a higher loading will result in breakthrough of the poll utants so chat removal still occurs but to a lesser extent compared to the loadi ng. A maximum elimination capacity will eventually be reached, at which point a further increase in load wi ll not bring about increased pollutant removal. In this case, the infl uent race of pollutant flow will exceed the rate of decomposition and limit the availability of sires for absorption, leading in rum to breakthrough of poll utants. Ir should be noted however chat at low concentrations of pollutants (i.e. < 0.05-0. l mg/m3) and for poll utants chat exhibit high Henry's constants, the elimination capacity will be decreased by reduced biological reaction races (Kennes and Veiga, 2001) . Different medias will exhibit di fferent el imination capacities for pollutants (assuming moiscure and nutrients are sufficienrly available). T herefore, in order to achieve high removal efficiencies, rhe required el imination capacity will need to be less than che max imum ach ievable by rhe media in question.

It is important to understand chat biocrickl ing fi lters will typically exh ibit near complete removal efficiencies at relatively low pollutant

Biomass growth

Remove wastewater odours naturally with the Little River Pond Mill Wind driven aerator

A vailable in wind, solar & electric models

Benefits • • • • •

Reduces/eliminates odours Reduces/eliminates a]gae Reduced energy costs Reduces aquatic weeds Reduce silt/sludge buildup • Minimal maintenance • Chemical free • Portable

'Wini/ii[[ Pty Lta 136 MARCH 2006


While nutrient addition is necessary to achieve high performance, cloggi ng of the media by accumulated biomass has been reported by a number of investigators (Kennes and Veiga, 200 1) . In field applications, chis is generally controlled through recirculation rares and optimisation of nutrient addition. Ocher control strategies which have been investigated include use of growth inhibitors, predation by protozoa or mites, and physical and chemical cleaning (e.g. sodium hydroxide, sodium hypochlorite, hydrogen peroxide). However, these have only been applied to laboratory and pilot plant systems.


Phone: 08 9454 5334 Email: enquiries@ windhiU.com.au Fax: 08 9454 5339 Web: www.windhiU.com.au

Biocrickling filter performance requires temperatu re, oxygen, nutrient supply, recirculation rares, and operating pH to be sufficient or suitably controlled. The type and volume of packing material are also important in limiting

refereed paper

odour management high pressure drops and in providing a large surface area for sufficienr microbial populations ro exist. Provided these parameters are satisfied, high reduction efficiencies can be achieved for a variety of organic and inorganic compounds. Perfo rmance data for hydrogen sulphide removal from two systems operating within Australia is show in Figure 7 (Aromarrix Australia Pry Led). For these systems, removal efficiencies of 80% were recorded upon start-up, with susta ined average reduction efficiencies of 99.8% being achieved. Effective perfo rmance is shown throughout the normal diurnal cycle at concenrrarions approximately three rimes the average which is typical of most sewer systems. However, in th e unusual case where concenrra rions exceeded the average by around five rimes, the reduction efficiency was reduced to around 98%. Odour removal perform ance was measured at around 87%. Tab le 2 reports removal efficiencies for hydrogen sulphide, odour and ocher volati le organ ic compounds at va rious full-scale biorrickling filter insrallarions. Reduction efficiencies for odour will depend heavily on the make-up of

AroBIOS Biotrickling filter - H2S Removal Performance (EBRT = 9 sec) 50.0

- -




100% 95%





S 30.0



E u






u, 20.0









·.~.•. ~.

- :--·

5.0 0.0





' '




'~, '~, ,~

~ ~



' '


----~ ·-


. .








~ ~




' '




., C


ic w C








55% 50%








~ ~

f"::iniet (ppm) • Outlet (ppm) • Reduction Efficiency (%)

Figure 7. Biotrickli ng fi lter performance data for hydrogen sulphide (Aromatrix Austra lia Pty Ltd).

co mpounds within the waste gas scream. For application 4 with in Table 2, fou l air is drawn from various pares of the wastewater rreatmenr plant which receives domestic sewage with a significa nt industrial component. As a result, inlet

concentrations of volatile organic compounds (VOC) were relatively high. Removal of voes by the biorrickling filter was typica lly less than 20% since the system was designed to optimise removal of hydrogen sul phide and not other voe


Humegard"' - Gross Pollutant Trap


Captures and retains gross litter, debris, vegetation and organics, medium and coarse sediment.

Captures and retains fine suspended solids, hydrocarbons, petroleum products, heavy metals, oil and grease.

• • • • • •

• Extensive performance data under field conditions • Retention of 80% TSS including particles < 60µm • 98% hydrocarbon capture • Very low operational velocity < 0.007m/s • Market leader - 18,000 units worldwide

Large volumetric capacity Low operational velocities Litter capture at high flow Very low headloss Performance proven in field 50 year design life

Waterce/1 ™

ph: 1300 361 601 I email: info@humes.com .au I www.humes.com.au I Enviro 06 Booth 55

odour management chemical scrubbers were converted to biotrickling fi lters treating airfl ow rates fro m 15,800- 68,000 m 3/ h. T he resultant EBRT was in the order of 1.6- 2.2 s since chemical scrubbers are designed o n 20 ~ lllO 10 100 0 a much lower residence time than 1 4 H,S load (9 111 h' ) biological systems. Based on inlet -- .-.~concentrations for hydrogen sulphide ranging fro m 0.1-100 W hile residence times have ppm (50 ppm average), removal generally been between 10 and 60 efficiencies of >97% were reported 0 . j C ~ - , - - . . . . - -.......---.---.---T---,,----4 s, recent studies have shown that t~O 180 180 80 80 100 12D 20 40 month period. The over an eight 0 effective removal efficiencies for l(.S lo•d (g mi h4 ) elimination capacity fo r hyd rogen hyd rogen sulphide in excess of sulphide over this period is shown 97%, at inlet concentrations in the Figure 8. Elimination capacity profile for hydrogen in Figure 8. While a maximum range of 20- 100 ppm and mass sulphide (Orange County Sanitation Districts, Californ ia, elimination capacity of around 120 3 loading rates of up to 80 g/m .h USA) (printed w ith permission, Gabriel 2003). g/m 3 .h was demonstrated, can be achieved at residence times red uction efficien cy dropped off At two wastewater treatment planes as low as 1.6-3 s which is comparable to significantly (i.e. <97%) beyond an Orange County Sanitation operated by the chemical scrubbers (Koe 2002), (Gabriel elimination capacity of 80 g/m3.h. number of Districts (California, USA), a 2003). Reduction effi ciencies for odour, carbonyl sulphide, methyl mercaptan and carbon disulphide of 65%, 44%, 67%, and 35% Table 2. Summary of removal efficiencies recorded for some biotrickling filter were also reported with corresponding inlet installations . concentrations of 1,980 OU, 67.2 ppb, Inlet concentralion/ range Average removal Applicalion 192.5 ppb and 70. 3 ppb respectively. efficiency (%) (ppm, unless stated) T he success of biotrickling filters in l. Wastewater treatment, Queensland, achieving high removal effi ciencies for both Australia IAromatrix Aust.) hydrogen sulphide and sewage odour at 23,300 OU 87 Odour EBRTs close co that of chemical scrubbers, 99.8% 12 !av.), 47 (peak) Hydrogen sulphide has challenged the viability of the latter due 2. Wastewater treatment, Queensland, co the inherent problems of high capital and Australia IAromatrix Aust.) operating costs and the hazardous nature of 99.5% 8 !av.), 29 (peak) Hydrogen sulphide reaction chemicals and waste streams. 3. Wastewater treatment, Steenderen, Based on a survey by the author of several The Netherlands IKroakman 2003) water authorities operating biotrickling 99 69 Hydrogen sulphide filter plants for control of sewage odours, 0.6 18 >93 Methyl mercaptan the fo llowing conclusions were noted: 49 0.035 Dimethyl sulphide • Biotrickling filters were well suited co the 64 0.01 l effective removal of hydrogen sulphide at Dimethyl disulph ide concentrations up co 500 ppm and removal 4. Wastewater treatment, California, USA effi ciencies in excess of 98 %; !KBR 2004) 3,400-29,000 OU 58 (inlet < l 0,000 OU) Odour • Odo ur removal efficiencies of 58%-94% 68 (inlet >l 0,000 OU) were achieved although this was heavily influenced by the concentration of VO Cs 98 53 (av.), l 00 !peak) Hydrogen sulphide within the waste gas scream. In order co 5. Wastewater treatment, Florida, USA !KBR 2004) achieve higher odour removal efficiencies, a 400- 500 >99.8 Hydrogen sulphide secondary treatment step such as carbon 6. Wastewater treatment, Florida, USA (KBR 2004) absorption or a biofilter may be required; >99.5 200-500 Hydrogen sulphide • Fluctuations in inlet concentration are 7. Wastewater treatment, Florida, USA (Aromatrix Aust. ) handled effectively, with little impact on >99 Hydrogen sulphide 18 !av.), 44 !peak) the outlet concentrations; 8. Wastewater treatment, Singapore IAromatrix Aust.) • Acclimatisation periods were typically 94 16,400 OU !av.) Odour 2-4 weeks before fu ll performance was 98 12 !av.), 31 !peak) Hydrogen sulph ide achieved (hence, an alternative treatment

components; overall odour removal was consequently low. For each of the other wastewater treatment applications, sewage was essentially of a domestic nature with hydrogen sulphide the more dominant of the odorous compounds. Therefore, odour reduction efficiencies were relatively high.

9. Chemical processing, Victoria, Australia Styrene l 0. Textiles, Ohio, USA (Kraakman 2003) Hydrogen sulphide Carbon disulphide






200-400 200-500

> 90 >


Notes: OU = odour units {number of dilutions required to achieve threshold concentration)

138 MARCH 2006


system may be required during this period, e.g. carbon fil tration); • Biological systems have the potential co be poisoned by toxic compounds within the air stream, mainly in catchments dealing with significant industrial waste screams. However, where this was experienced,

refereed paper

odour management d urations were typically less than 24 hours prior to fu ll performance bei ng restored . Additional nurrienr d osing may also be required d uring this time; • Biotrickli ng filte rs require very little operator attention and very low maintenance.

So me treatment methods (e.g. biofiltration, biotrickling filters, chemical scrubbi ng) m ay also need pre-treatment or po lishing of the waste gas stream in order co ach ieve h igh reduction efficiencies wh ile other methods may not (e.g. thermal combustion).

Figure 9 provides a comparative representation of the capital and The overall capital and operating operating costs for various Capital cost ($) coses fo r biological treatm en t (both treatment technologies. It m ust be biofi lters and biocrickling filte rs) stressed char coses will d epend o n Figure 9. Relative cost comparison for various treatment are generally low compared with many faccors such as airflow rate, technologies. other proven treatment the range of compounds co be tech nologies such as chemical treated and their influent scrubb ing, activated carbon concentration , tempera rnre and biofiters will result in biotrickl ing fi lters ad sorp tion and thermal combustion. It is the required rem oval efficiencies. being more cost effective. Similarly, difficult, however, co co mpare the costs activated carbon is suitable fo r low Conclusions associated with d ifferent treatment methods concentratio ns of pollutants o r for effluen t since all tech nologies may not be su itable Biocrick li ng filters provide the abi lity of polishi ng and generally has low co medium for treatment of a particular waste gas treating a variety of waste gas stream s in a cost compared with ocher methods. scream. For example, biofilters wi ll very efficien t, low co st and H owever, for high pollu tan t generally exhi bit low capital and operating envi ro nmentally fr iendly ma nner. In fac e, concentrations, operating costs will be h igh. costs for waste gas stream s w ith low the success of biotrickl ing filters in Capital costs may also be increased concentrations of hydrogen su lphide. achieving high removal efficien c ies for sign ifi cantly by the need to provide enough H owever, for high concen trations, the carbon mass to achieve a su itable bed life b oth hyd rogen sulphide and sewage problems with acid accu mulation within od o ur at EBRTs close to chem i cal between carbon change-overs.


odour management scrubbers, has challenged the viab ility of the latter due to the inherent problems of high capital and operating costs and the hazardous nature of reaction chemicals and waste screams. However, for any ap plication requiring treatment, ic is critical that che problem requiring a solution is fully defined. This will include q uantification of such items as airflow, temp erature, co mpositio n of che waste gas scream and pollutan t concentrations and variations in em1ss10ns. For biological treatment to be effective, pollutants will need to be biodegradable and also water soluble to some degree to enable transfer to the biofilm. If biodegradation leads to appreciable toxic by-products because of high influent pollutant load, then biotrickling filters may be more appropriate than biofilters. Selection of the for m of biological treatment will depend on the removal efficiency required, the availability of land, pre-treatment and pose treatment steps required to achieve the treatment objective (i.e. a secondary treatment seep may be required to achieve effective odour


= = = = = z = ;:: = ... = u, ~

Odour Control Systems Odour Measurement and Analysis

removal), as well as the site-specific capital and operating cost implications. H aving made che selection, consideration of che critical design and operational issues discussed in this paper will in turn enable an effective and reliable biological treatment system to be implemented.

The Author Gary Finke is a chemical engineer and Managing Director of Aromatrix Australia Pry Led, 18/148 Chesterville Road, Moorabbin Victoria 3189, Tel: 03 9553 1171, Fax: 03 9553 1193, Email: gary.finke@aromatrix.com

References Finke, G.N. 2004 . Biofiltration and bioscrubbing. In A Training Course in lndusttial Odour Control, C lean Air Society of Australia and New Zealand Inc, November 8-10, 2004. Gabriel, D., Deshusses M .A. (2003). "Retrofitting Existing Chemical Scrubbers to Biotrickling Filters for H2S Emission Control " in Proc. Natl. Acad. Sci, U.S.A. 100(11): 6308-6312, 2003. KBR (2004) . fntemal report (BEG304 l l-B-DR004), Kellogg Brown & Root, March 2004.

Kennes, C., Veiga, M.C. eds. (2001). Bioreactors for Waste Gas Treatment, Kluwer Academic Publishers. Koe, L.C.C., Wu, L., Loo, Y.Y., Koh, Y.M., Wu, Y., Wei, J. (2002). "Chai Pilot-Scale Experiences in Converting a Chemical Scrubber to a Biotrickling Filter for the Treatment of Sewer Air", Cranfield University 2-6 March 2002. Kraakman, B. (2003). "Biotrickling Installations Applied for the Treatment of Municipal and Industrial Air Pollutions, 2nd !WA International Workshop and Conference "Odour & VOCs Measurement, Regulation and Control Techniques", 14-17 Singapore, September 2003. Kraakman N.J .R. (2004). "H2S and Odour Control at Wastewater Collection Systems-An On-Site Study of Biological T reatment", Enviro04 Convention and Exhibition Sydney, 28 March- I April 2004 . Tay T.C.H, C.C.L Koe, Y.Y. Loo, N.S. Lim, "Comparative field pilot biotrickling filter performance between packing media". Proceedings from Ozwater Convention & Exhibition, Brisbane, Queensland, 8-12 May 2005 (Paper no. 05017). USEPA (2003). Using Bioreactors to Control Air Pollution, USEPA EPA-456/R-03-003, September 2003.

• 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. • Manufacturers of AroBIOS™ biotrickling filters, AroFlL™ biofilters, AroCARB™ activated carbon filters, AroCHEM™ chemical scrubbers, and Odormat™ olfactometers. • Services include : Odour control strategies, sewer modelling, atmospheric dispersion modelling.

140 MARCH 2006


refereed paper

odour management


• Sulphu rous, e.g. H 2S (ro tten eggs) , methyl mercaptan (decayed cabbage)

Odour Emission Capaciry measuremenrs enable wastewater operators to quanrify odours and app ly suitable control methods to minimise their impact on local recepcors.

• Nitrogenous e.g. indole (faecal, repu lsive) methyamine (fis h y, rotten) • Aldehydes (frui t, apple) T h e generat io n of H 2S a l so causes infrastructure co n cerns du e ro sulphuric acid corrosion and OHS issues consequent o n co nfined space access by wastewater ut ility staff.

Introduction Odorous emissions from sewer systems can cause serious an noyance co local recepcors. and the number of public complaints is steadily rising . T hese impacts are influenced by the o rganic load of the wastewater being transported th rough the sewer system. Figure 1. Example of force choice olfactometer Odoran ts which are p resent in the wi th two sniffing ports (Courtesy of Aromatrix, liquid phase of a sewer system are Singapore). emi tted inco ambient air wherever th ere is a liquid-gaseous interface. The sulph ide (H 2 S) and also volatile organic emission rate is mainly dependent on two compo unds (VOCs) originating from the factors: (i) the physical-chemical food waste itself or due co fermentation circumstances such as turbulence, size of p rocessing durin g the sewage transport. interface etc. and (ii) amoun t of odorants To recognise and define odours, we present in the liquid phase due co microbial common ly use d escripcors. In the science transformation of sulphates in co hydrogen and practice o f odours (including perfumery and wine tasting) there are over 830 d escri ptors, but for the field of Industry wastewater industrial complai nts the major descripcors odours are different from are:

Odour Measurements Odour measurements can be divided into two broad areas of analyses: chemical measurements of specific compounds and sensory o r olfactory measurements relating co odour p erceptio n . Senso ry measurements employ panellists of human noses as the odour detector and hence relate d irectly co the properties of odo urs as experienced by humans. Either the European standard method (prEN 13725) is employed, giving the u nit OUE, or the more recent Ausrralian and New Zealand standard (AS/NZ 4323.3:2001).

n01·mal domestic Prt llmlnarytrtatmt nt


.. ..

wastewater odours. ,r









10000) .


1000 •


100 •



C;,.;• ~.11· R'•0.4'51

....,. ,

,. 0.01




.- '""






. .. W•0978


Figure 2. Example of hydrogen sulphide monito r - H2 S - Jerome 631-X (Courtesy of A rizona Instruments).



J ,., . .

c;..._ . 1~ ; ,;



,E ,





..., ... 1




Figure 3. Correlation of H2 S w ith Odour Units in wastewater treatment (Gostelow and Parsons, 2000) .


MARCH 2006 141

odour management There have been distinct improvements in the techniq ue and the apparatus available commercially in the past five years. An example of a commercial apparatus is shown in Figure 1. Chemical analysis of odorant concentrations potentially have the advantage in obj ectivity, repeatability and accuracy compared to olfactory measu rement. Since the odorous emission is a mixture of gases, separation techniques such as gas chromatography coupled with mass spectroscopy (GC-M S) are often used for qualitative or quanti tative measurements of od our co mposition. In some cases, it is often fou nd that, although many odorants are presen t for a particular odour source, a single odorant is dominant, such as H 2 S, which can readily be d etermined by a number of hand-held and bench instruments. (e.g. Figure 2) . This can be very useful, if the measure of the single odorant can be indicative of the odour as a whole and is easily measured. However, limi tations exist with the relationship between single compounds and overall olfactory responses. For example, when dealing with municipal wastewater, the predominant odour is due to H 2 S, and Figures 3 demonstrate that for most phases of wastewater treatment there is a strong correlation between Olfactory Analysis and the measured concentration of H 2 S. H owever, in the aeration tanks the H 2S has been oxid ised, and the odour is due to other compounds. The same factor often applies to odours from the food industry, which are rarely due to H 2S, so the simple H 2S instruments have li mited applicabil ity

Figure 4. Olfactory-ma ss spectroscopy showing an o lfactory port coupled to GC-MS.

More recently the integration of chemical and o lfactory techniques has been applied to odour analysis to allow the correlation of chemical and sensory measurements via the coupling of an olfactory port to a GC-M S. This technique (GCMS-O) allows annoyance odorants to be separated and identified individ ually as well as allowing the odour co ntribution for each compound to be characterised (Figure 4).

Odour Emission Capacity of a Liquid To measure the odour emission of a liquid wastewater sample, a methodology is needed char relates the emission of odorants to the volume of the respective liq uid rather than to the surface as determined by conventional window runnel and emission hood measu rements (see Gosrelow and Stuetz, 2004). T his parameter, known as the Odour Emission Capacity (OEC) of a liquid, is defined as the total amount of odorants present in that liquid char can be stripped from chis liquid under standardised condi tions (Frechen and Koster, 1998). This proced ure and experimental apparatus (Figu re 5) co nsists of aerating a liquid of known volume with odourless air. The stripped air is sampled immediately after the start of aeration, and then su bsequently after certain ri me intervals un til the end of the test. T he samples are analysed to determine thei r odorants co ncen tration using olfactory measurement and/or specific chemical measurement (i.e. H 2 S) . Figure 6 shows the olfactory measurement results of samples taken during the OEC test time for two different samples. The OEC values are calculated by integrating the area between the measurement values and the lower limit of 100 o u/m 3 over the

Figure 5. Odour emission capacity test reactors.

:2 ::,


{ Do }



"' E 111 ::I

lfactometric measurement results





..e C






500 ¡







The Odour Emission Capacity (OEC) of a liquid is the total amount of odorants present in that liquid, that can be stripped from this liquid under standardised conditions. Units: ou/m3 liquid



0 0




air flow in liter GEP.Ausw ertung_Exla

Figure 6. OEC test showing odour concentrations a nd differences on O EC measurement for two wastewater samples.

142 MARCH 2006


odour management amount of air in linear scale (Frechen and Koster, 1998). Figure 7 shows che comparison between a municipal wastewater and meat industry effluent for OEC measurement by odour concentration, hydrogen sulphide and dimechylsulphid e. The results support che argument chat many odorants can be present in a wastewater and chat H 2 S is not rhe most appropriate odour marker particularly for foo d industry-based wastewacers.

Table 1. O EC for different wastewater effl uents.









Chemical Industry






Meat Industry 121 ,000














4,000,000 Meat industry

Municioal sewer -0- Qdour

• Li miting d ischarge of odorants inro sewer systems and

-+- H2S

-0- 0dour

-0- Dimethylsulphide





OEC = 594,000 o.u.,!m 3


.€ 6 Cl

E 5 .!:


·;: 20000


.g 15000




-+- H2S

-o- Dlmethy lsulphlde

120000 100000

7 ,.,·ii

. 35000

'"'e 30000 'ill 6 25000

The odorous ch aracteristics of liquids, particularly wastewater generated fro m food industries, can exhibit significantly different odour emissions from municipal sewage. The OEC techniques enable wastewater operators co quantify odours and apply suitable control methods co minimise their impact on local receptors.

Samples plants


• Assessing relevancy of odours from different liquids;



3 0UE/ m Liquid

Table 1 shows the OEC for some typical wascewacers samples from different industries The OEC has a number of specifi c applicatio ns chat can be applied co the assessment of odours from sewers, these include:

• Assessing effi ciency of d osing chemicals into sewers for odour suppression.



"0 C :::,


OEC = 5,210,000 o.u.,{m




'"'e 'ill :::, 0



20 Cl

E .!:








.!: 60000


&. .g0






2 0


E 0 u

5000 0

0 0





air volume in m3




air volume in m3


Figure 7. OEC co mparison between a mun ic ipal wastewater and meat industry

References Gostelow P. and Parsons S.A. (2000) Sewage t reatment odour measurements. Water Science and Technology, 41 (6), 33-40. Gosrelow P and Scuetz R. (2004) Odour Sampling and Measurement: A R eview. Water. 31 (2) 107- 11 3 Frechen F-B and Koster W . ( 1998) Odour em ission capacity of wastewaters - standardisation of m ethod and application. Water Science and Technology 38 (3): 61-69


The Authors Assoc Professor Richard Stuetz is Deputy Director of the Centre for Water and Waste T echnology at the University of New South Wales. Email: r.stuecz@unsw.edu.au. He lead s che Environmental Emissions and Odour

Program at UNSW. www.odour.unsw.edu.au. Prof FranzBernd Frechen is Professor and H ead of the D epartment of Sanitary and Environmental Engineering at the University of Kassel, Germany. Email: frechen@uni-kassel.de


~~ c:=.J PCM


Scum, Sludge & other Unmentionables Tel. (02) 6581 0744

Fax. (02) 6581 0790





MARCH 2006 143

GOING, GONE, TRENCHLESS: THE HALLAM VALLEY MAIN SEWER EXTENSION C Corr Abstract The 9 kilometre extension of the Hallam Valley Main Sewer currently under co nstructio n in Victoria is one of South East Water Limi ted's most significant projects to date. The 1500 mm d iameter plastic lined sewer is being en tirely constructed by trenchless technology at d epths of between 5 and 12 metres. The sewer alignments will traverse paddocks, major open drains, railway lines, arterial and local roads, a major gas p ipeline, high voltage transmissions lines, wetland s and public open space reserves. Small lengths of the sewer will be in close proximiry to houses. The fl ood p lain is predo minancly low enviro nmental value cleared land used for grazing, however, because of th e sewer dep th, open cut co nstruction may have required up to a 60 metre wide band of major high impact earthworks. T he alignment also has a high ground water table, making excavatio n, and in particular re-com pactment, of the. excavated material very difficu lt T he adop tion of trenchless tech nology construction has minimised noise, dust, environmental impact, negative aesthetics, construction traffic, O H &S risks and general inconven ience to the community. The risks associated w ith significant rainfall events during constructio n causing environmental damage and public impact from sedimentation, possible increased flood levels, longer project timeframes and an increase in costs were all minimised.


VALLEY 'Offtcer


Cranbouroe Growth area

Existmg urban area

Urban growth boundary

Possible future development front

Figure 1. Lo cation of the Pakenha m G ro w th Corrido r.

Triple bottom line justifications for selection of trenchless construction. Corridor under the State Government strategy M elbourne 2030 - Planning for Sustainable Growth (Figure I ). T he area generally incorporates the development in the H allam Valley which includes subu rbs

such has Hallam, Hampton Park, Narre Warren, Berwick, Berwick Springs and rap idly expanding suburbs such as Officer and Pakenham.

This paper describes the sco pe of the project in some detail as well as exploring SEWL's Triple Bottom Line justifications for selection of tren ch less construction over the cheaper open cut alternative.

Th e area is the responsibiliry of South East Water Limited (SEWL), one of Metropolitan Melbourne's three retail water companies.

Introduction and Background General The South Eastern Growth Area has been previo usly identified by the Victorian Seate Government as one of the principal areas for Melbourne's future residential growth. T his has been intensifi ed by the area's designation as the Pakenh am G rowth

144 MARCH 2006


It is proposed to transfer sewage fro m O fficer and Pakenham South & West into the Hallam Valley T ransfer System (HVTS), which ultimately discharges into Melbourne Water's Eastern T reatment Plan t. A number of key elements o f the H VTS, however, are approaching capacity and require augmentation to carer fo r the p redicted growth within the Hallam Valley, and also for the add itional flows from Officer and Pakenham.

The previous Hallam Valley transfer system

Figure 2. The Herre nknecht tunnel bori ng machine .

Early sewerage network p lannin g for the area included a major trunk sewer at depth extending up the H allam

refereed paper

sewerage Valley and eventually as far as Pakenham. Around 1980, rhe then Melbourne and Metropolitan Board of Works (now Melbourne Water), constructed an 1800 mm internal diameter gravity sewer as far as Hampto n Park by tunnelli ng. T his sewer was intended to extend further ar chat time but was cut shore as a result of unfavourable geological conditions ahead on the selected alignment. As an interim measure, a series of gravity sewers, pump stations and rising mai ns were co nstructed to serve the Hallam Vall ey. These key pump stations as shown in Figu re 2 include T CPS, SRPS, GLRPS and W RPS. Except for TC PS these stations pump rhe sewage seq uentially from one station to rhe next before the flow discharges in to Melbourne Water's 600 mm diameter South H allam Branch Sewer in Hallam. The limited capacity of ch is sewer has increas ingly become a major hi ndrance to rhe effective operation of the system and has limi ted the potential fo r capacity upgrades of the pump stations of wh ich many are operating contin uously in wet weather to carer fo r peak flows. During initial planning, the project was split into fo ur distinct sections of work with the major pump station connection poi nts fo rming the section boundaries. These distinct sections could be used to stage rhe works over a period of ri me. Section numbering was later revised to that shown in Figure 2. Worki ng from the downstream end (page left), the design flows required the in ternal diameters of the fou r sections to be 1500, 1350, 1350 and 1200 mm respectively. Options for system augmentation

Prior to the decisio n to include the Officer and Pakenham flows a number of broad options fo r augmenting the system had been assessed. The rwo main options were as follows: Gravity sewer option proposed the co nstruction of a single gravity sewer to receive flows fro m the Hallam Valley, with the subsequent decommissioning and diversion of several of rhe existing pump sranons.


KEY Sl - Section 1 S2 - Section 2 S3 - Section 3 S4 - Section 4 S... Plan Source: GHO Pty Ltd. (2002). Na= Wamn Pakenlwn Corridor Sewerage SU.tegy-R.port on

FurctioMI Design, Figur, J TranJfar .S:,.st,ms



Figure 2. Previous Hallam Valley Transfer System Infrastructure.

requiring a very substantial increase in the capacity of the proposed augmentati on. A review report recommended that fun ctional design proceed on the basis of a majo r gravity sewer and ch is recom mendation was subsequenrly adopted by SEWL.

General sewer route

used for grazing. The majority of the land is owned by Melbourne Water. T he Hallam Main Drain runs the length of the route and there are a number of other drainage schemes in the area. Immediately on the southern sid e of the flood plai n is established resi dential housing and there are predominanrly commercial/ind ustrial sires to the north with similar develop ment continuing.

The general route selected fo r the sewer was along the Hallam Valley flood plain (Figure 3) which incorporates the Hallam Valley Main Drain.

T owards the upstream end, rhe route traverses narrower drainage and sewerage reserves adjacent to the main drai n an d close to new residential estates .

Hallam Valley Main Sewer Extension Route and Alignment

The flood plain is rhe logical route as it is the lowest land in the area being typically between 100 and 500 metres wide and relatively flat. T he general route and later the derailed alignment always required some limited elements of trench less construction. Route surroundings and land use

The flood plain is predo minanrly highly modified and cleared pasture land presenrly

Potential alignment restrictions and obstacles

The alignment is governed by the downstream connection point in Oakrree Drive in Hamp ton Park, and th e required connections with the fo ur major pump statio ns shown in Figure 2 as well as rhe existing 600 mm diameter gravity T i-Tree Creek Main Sewer in Berwick Springs.

Pump/storage option proposed to optimise the use of the existing assets as much as possible by progressively upgrad ing existing pump stations and rising mains to transfer sewage to the head of rhe Hallam Valley gravity sewer, with the construction of associated detention storages to attenuate the peak flows.

In May 2002, a decision was made to transfer existing and futu re flows fro m growth corridor development in Officer and Pakenham to the Hallam Valley system,

refereed paper

Figure 3. The Hallam Va lley flood plain including the Hallam Valley Main Drain.


MARCH 2006 145

sewerage These connection points are all located within the flood plain. At the commencement at Oaktree Drive, a relatively narrow and awkwardly angled floodway on River Gum Creek has been utilised to gain access to the flood plain. Alignment obstacles co be crossed include: the Hallam Main Drain in H ampton Park, Hallam So uth Road, multiple high voltage transmission lines, Troups Creek West Branch, the Gippsland Railway, Sh rives Road, the Narre Warren Creek, the trunk Longford natural gas p ipeline, Narre Warren - Cra nbourne Road, Golf Links Road Wetland, the Hallam Valley Main Drain in Narre Warren South, Centre Road and the Hallam Valley Main Drain in Berwick. With all of the above obstacles, it was a requirement of rhe asset owners that all crossings be constructed by trenchless technology co minimise inconvenience to users, risk or damage to th e assets and any negative environmental impacts. In addition to the above, four parcels of land in the flood plain are privately owned requiring consultation and minimisation of construction impact.

Detailed sewer alignment Commencing from the downstream end, the selected Sections 1 and 2 sewer alignment tracks immediately to the south of O'Grady and Centre Roads within rhe Melbourne Water owned (Section 1) and privately owned (Section 2) flood plain as shown in Figure 4. A proposed 900 mm diameter sewer connects with rhe Troups C reek Pump Station and includes a crossing of the Gippsland Railway.

Figure 4. Alignment for HVMS Extension Sections 1 and 2. An add itional benefit of the alignment is that it passes relatively closely to five smaller existing pump stations. The three closest ones can b e easily connected and decommissioned , another will have its rising main connected and thus shortened and the fifth could be connected in the future when major expenditure such as replacement o f p umps, motors and electrics becomes necessary. In all, the length of the Hallam Valley Main Sewer align ment is approximately 9 k ilometres with more th an 1 kilometre of additional major connecting sewers.

Sewer Construction Methodology General Two main options were available for construction of the main sewer: open cut and trenchless.

As discussed above, there were many segments where trenchless construction was required as a result of asset owner requirements. These sections comp rised in total some 1500 metres or approxi mately 16% of the overall alignment. The most common construction methodology could therefore have been expected to be a combination of the two major options.

Open cut construction Open cut construction would be possible but quire difficul t at depths averaging 8 to 10 metres. Large earthmoving machinery would be required with a construction width of app roximately 60 metres (Figure 6). The selected alignment offset discussed earlier would allow such a construction method. Much of the route could accommodate a 60 metre wide construction zone, however, there would be a number of areas, particularly in Section 4, where such a width would not be possible.

T he offset adopted where possible throughout was 17 metres from the road reserve boundary to enable possible open cut construction. The hatched area on Figures 4 and 5 indicates approximately a 60 metre wide zone of influence for open cut construction.

Open cut construction would require a large number of wo rkers and items of plant, would be highly disrup tive and damaging to the flood plain, would create a large amount of noise, dust and mud, would be a visual eyesore and be a general an noyance to nearby commercial and residential properties.

The Section 3 alignment tracks west of Centre Road and then south of Golf Links Road, as shown in Figure 5 , and through Melbourne Water's Golf Links Road Wetland constructed in 2003. At the eastern end of Golf Links Road private property is traversed and the alignment deviates so uth to avoid an area of proposed development.

O pen cut construction would also be more prone co delays due to wet weather and the cohesive clay soils in the area would be expected to be extremely difficult to adequately recompact given the varying moisture content, the dep th of excavation and the amou nt of refill required.

The Section 4 alignment tracks through a landscaped drainage and sewerage reserve and in close proximity to residential houses as shown in Figure 5.

Trenchless construction tech nology is typically used where there are either obstacles on the surface along the rou te or the depth makes open cur

146 MARCH 2006


Trenchless construction

Figure 5 . Alignment for HVMS Extension Sections 3 and 4 .

refereed paper

sewerage co nstruction impossible or uneconomic. Trenchless conscruccion is required for elements of chis project as a result of surface obstacles as discussed earlier. This co nstruction methodology could be extended to che whole project to minimise the impact on the su rrounding area, however, the low envi ronmental value of the adjacent land surface does not make such conscruccion a requirement. Construction by crenchless technology involves a series of smaller works sites as opposed to a linear and continuous construction site. Each work site may req uire an area in the order of 1,000 3,000 m2. The smaller works areas wi ll be highly beneficial in Section 4 due to che limited space available and the proximity to housing. T renchless construction brings with it a different set of challenges and risks. The equipment being utilised is more "high tech", work spaces such as che shafts and tunnels are confined and there is che need to carefully manage che repetitive casks associated with cunnelling operations co ensure safety and efficiency are maintained. T renchless construction may be more







Note: Approximate dimensions shown are in metres. 60

Figure 6. Typical open cut construction footprint.

susceptible co technological and georechnical risks buc is less affected by weather conditions, and in particular, rai n and flood ing rhac can be regularly expected with a works si te located entirely within a flood plain. The tendering process

fc had been identified ea rly chat che final conscruccion solution would include a significant portion of crenchless technology

as discussed above. SEWL, and their consultants GHD, elected co call for tenders based on cwo scenarios. T hese scenarios were: Scenario 1: Predominantly open-cue construction with some for m of crenchless technology used for crossing roads, watercourses, wetlands and ocher unavoidable surface obstacles. Scenario 2: Maximum use of rrenchless technology with excavatio ns limited co

maces The force i n flow.


sewerage causing snagging and odour and sulphide generation issues. Construction on the Hallam Valley Main Sewer Exten sion has continu ed almost uninterrupted for over 18 months in a flood plain through what has been a t rimes unusually wet weather. Two major flood even cs of approximate average return intervals I in 100 and I in 150 only caused m inor delays of up co 36 hours. T he average pipeline installation race has been 18 metres per 24 hours. Construction is expected to be completed in early to mid 2006, ahead of rime and under budget.

Conclusion The decision by SEWL co construct che extension of che Hallam Valley Main Sewer entirely by trenchless technology at additional cost co the open cue alternative and despite chere being some disadvantages is a sign of the way forward in construction of our major infrastructure. W hilst every site has environmental value, the Hallam Valley flood plain is not an example of a

pristine, undiswrbed environment nor is che route littered with insurmou ntable obstacles. It is however a fast growing residential corridor.

The Author

As a result of the Hallam Valley Main Sewer Extension - Going, Gone, T renchless, the effect on the area and ics residents and visitors is being minimised, as are the risks of flooding and escalations in the cost of construction. In addition, the knowledge and experience relating to trenchless technology within the Victoria n and Australian construction industry is being furthered, thus assisting future major infrastructure projects also to - Go T renchless.



Melbourne 2030 - Planning for Sustainable Growth, Figure 1 - Managing Urban Growth. GHD Pry Lrd. (2002). Narre Warren Pakenham Corridor Sewerage Strategy Report on Functional Design, Figure 1 Transfer Systems. Melway Publishing. Melway Street Directory -

T he auchor would like co thank the followi ng individuals and companies for their invaluable assistance with technical input and/or peer review of chis paper; • Chris Geehman - South East Water Ltd. • Colin White - GHD Pry Led. • GWC Group Pry Ltd.

Chris Corr is a Senior C ivil

Engineer with

GHD in the Melbourne office.

Corr, C. (2005), Australian Water Association Victorian Branch Regional Conference, Philip Island, Victoria, Australia, 6-8 October 200 5. Geehman, C. (2004) . Proc. 12th Aust ralian Tunnelling Conference, Brisbane, Australia, 17-20 April 2005 . Corr, C. (2004), Proc. 6th National Australasian Society for T renchless Technology (ASTT) Conference, Melbourne Australia, 27-29 September 2004. Department of Infrastructure. (2002).

Edit ion 29. British Standard BS591 l - 1989.

With six of the best brand names in municipal and industrial water and waste water treatment, and over 100 years of experience, it is clear that we can find a solution for you. Tel: 02 4320 4755 • e-mail: mark.houghton@glv.com • www.glv.com


sewerage Twin hydraulic rams in rhe launch shaft push the pipe sering and TBM with up to 700 to nnes thrust. Intermediate jacking stations were also sourced but have nor been used to date. Typically jacking forces have remained relatively low at under 300 tonnes with a few notable exceptions. Every fifth pipe has three bentoni te injection ports. Lubrication reduces jacking loads by lowering friction forces and increasing buoyancy and provides support to surrounding so il. An automatic bentonite injection system is used commencing from when pipe jacking starts. Each port has an independently co ntrolled delivery line giving great flexibility. Bento nite usage varies from virtually nil to 0.7 cubic metres per pipe. Pipe jacking continues 24 hours per day, seve n days per week to maxi mise production and minimise risks of the runnel becom ing sruck in the cohesive clays. Drilling foa m is ap plied to the clay material at the face as it is excavated preventing the clay from clogging the screw co nveyor rh ar transports rhe spo il from the excavation chamber in the T BM to the belt co nveyor rhar loads the muck trucks. Foa m dosage rate is typically 2-3 litres per metre advanced. Reinforced concrete pipes

GWC elected ro import the 3.0 metre long 1500mm internal diameter jacking pipes from T hai land. T he pipes have an outs ide diameter of 1800mm, with an 8mm thick integrated plain mild steel-jointing band, are fully PE-lined and weigh approximately 5.8 ton nes each. These are made to Bri tish Standard B5591 l- 1989, as there is not an equivalent Australian Standard. All tests required under the Standard are independently wirnessed and verified by Connell Wagner on behalf of GWC. In areas identi fied as having more aggressive groundwater, rhe steel bands and exposed concrete within the socket join t are epoxy coated. A nu mber of other jointing methods, materials and treatments were also investigated. Sewer diameter and maintenance hole spacing

Whilst rhe design called for sections of sewer of th ree different diameters ( 1500, 1350 & 1200mm), GWC offered to install the complete 9.1 km solely using rhe largest diameter pipe, with resul tant additional sewer capacity, at no exrra cost to SEWL avoiding porentially costly issues, includi ng: • Purchase of addirional TBM's and addirional training for staff.

refereed paper

• A slurry system for spoil disposal (smaller TB M's not being suitable for a muck truck system), Maintenance Holes (M Hs) were designed ar changes of direction or every 250 metres. GWC proposed the maximum spacing be extended to 400 metres and "savings" be shared equally between rhe cl ient/contractor for each MH eliminated. This has realised savings to SEWL of more rhan $0.5M over the project as well as minimising environmental disturbance. T he MH shafts are constructed usi ng a number of 2500mm internal diameter, th ree metre long concrete jacking pipes each weighing 16 tonnes. These MH riser pipes are also sourced from Thailand and are PE-lined. One reason for selection of jacking pipes fo r rhe MH risers was so rhar rhe joints could be easily secured with multiple external stainless steel straps to minimise the possibility of damage due to surcharge or flotation. Construction accuracy and progress

Ar the rime of writi ng (Feb 2006), all 27 main sewer pipe jack drives had been completed with an average drive length of approximately 300 metres and mul tiple

longest drives ar 400 metres. Only three of the 27 drives required mid-drive rescue (excavation) of the TBM. The three rescues were due to a mechanical failure, pipe roll co mplications resulting from serup errors at the start of the drive and encountering rock. In addi tion more than 1,200 merres of 900 sewer and 200 metres of 225 to 600 sewer have been installed. All 27 main sewer MHs have now been constructed and sire restoration works are underway. Most temporary construction shafts were co nstructed utlising sheer piling whilst shafts close ro structures and houses were constructed by ring-and-timber. Accuracy has generally been very good with receipt of the TBM typ ically within approximately I 0-30 mm vertically and 50- 100 mm horizontally of design. The maximum deviations from design have been 270 111111 vertically (low) and 420 111111 horizontally as a result of unfavourable ground co nditions. Where there were "dips" greater than 50 mm in depth in the installed lines, invert infill repairs were undertaken to maximise the hydraulic performance of rhe sewer by minimising th e likelihood of sed iment bu ild up possibly

Pure, safe water.

Alwa • • • • • • • •

Introduci ng th e Upstreamrn series Self Cleaning, mai ntenance-free Built in microprocessor 40mJ/cm 2 at 75% and 50% UVT Lab validated 4 log inactivation Up to 150m 3/ hr in parallel Sta inless steel and plastic housing Suitable for potable, grey water, pool and cooling tower applicatio ns

licensed Trademark of UV Pure Technologies Inc.

Call 1800 669 596 or visit www.uvsafewater.com.au


uvsafewater !W.~ •.e .• PACIFIC


MARCH 2006 149

causing snagging and odour and sulphide generation issues. Construction on the H allam Valley Main Sewer Extension has continued almost uni nterrupted fo r over 18 months in a flood plain through what has been at times unusually wet weather. Two major flood events of ap proximate average return intervals 1 in 100 and 1 in 150 only caused minor delays of up to 36 hours. The average pipeline installation rate has been 18 metres per 24 hours. Co nstruction is expected to be completed in early to mid 2006, ahead of time and under budget.

Conclusion The decision by SEWL to construct che extension of the Hallam Valley Main Sewer entirely by crenchless technology at additional cost to the open cut alternative and despite there being some disadvantages is a sign of the way fo rward in construction of our major infrastructure. Whilst every site has environmental value, the Hallam Valley flood plain is not an example of a

pristine, undisturbed environment nor is the route littered with insurmountable obstacles. It is however a fast growing residential corridor.

The Author

As a result of the H allam Valley Main Sewer Extension - Going, Gone, Trenchless, the effect on the area and its residents and visitors is being minimised, as are the risks of flood ing and escalations in the cost of co nstruction. In addition, the knowledge and experience relating to crenchless tech nology within the Victorian and Australian construction industry is bei ng furthered, thus assisting future major infrastructu re projects also to - Go Trenchless.



Melbourne 2030- Planning/or Sustainable Growth, Figure I - Managing Urban Growth. GHD Pty Ltd. (2002). Narre Warren Pakenham Corridor Sewerage Strategy Report on Functional Design, Figure I Transfer Systems. Melway Publishing. Me/way Street Directory -

The author wo uld like to thank the fo llowing individuals and companies for their invaluable assistance with technical input and/or peer review of this paper; • Chris Geehman - South East Water Ltd . • Colin White - GHD Pry Led. • GWC Group Pry Ltd.

Chris Corr is a Senior Civil Engineer with GH D in the Melbourne office.

Corr, C. (2005), Australian Water Association Victorian Branch Regional Conference, Philip Island, Victoria, Australia, 6-8 October 2005 . Geehman, C. (2004). Proc. 12th Ausrralian T unnelling Conference, Brisbane, Australia, 17-20 April 2005. Corr, C. (2004), Proc. 6th National Australasian Society for Trenchless T echnology (ASTT) Conference, Melbourne Australia, 27-29 September 2004. Department of Infrastructure. (2002).

Edition 29. British Standard 8 S5911 - 1989.

With six of the best brand names in municipal and industrial water and waste water treatment, and over 100 years of experience, it is clear that we can find a solution for you . Tel: 02 4320 4755 • e-mail: mark.houghton@g lv.com • www.glv.com



ACCEPTANCE OF WATER RECYCLING IN AUSTRALIA: NATIONAL BASELINE DATA JS Marks, B Martin, M Zadoroznyj Abstract Water recycling is recognised by policy makers and the water supply industry as being an important strategy in the management of urban water supplies. Following prolonged drought conditions, co mbined with predicted water demand, sewer mining has been given the official goahead in Sydney, indirect potable reuse is back on the planning agenda and various alternatives, including desalination, are either established, planned, or in the process of being implemented. With wide acknowledgement of the need for triple bottom line outcomes, rhe social aspect of water recycling needs to be co nsidered along with the economic and environmental benefits. To what extent do Australians accept the introduction of water recycling as a water conservation strategy? Key results of a national survey (n=2504) are presented in this paper that confirm the historical preference for non potable (not for drinking) uses of reclaimed water. In addition, this research finds that householders are will ing to use a range of alternative sources of water in the home and for purposes that will involve direct bodily contact.

Background While there is no published literature in Australia on national findings relating to

public acceptance of recycled water, a few scholarly stud ies have demonstrated that various communities, in Sydney (Syd ney Water 1996; Roseth 2000), Perch (ARCWIS 1999), and Adelaide's residential reuse sites (Marks 2003; McKay & H urlimann 2003:48), are enthusiastic about non potable (not fo r drinking) uses of reclaimed water (water sourced from sewage effluent treated to a standard suitable for its end use). T he same co mm unities are more reluctant to recycle the water for potable uses. T he percentages in Table l represent the proportion of each sample that support reclaimed water for a range of non potable and potable uses. Of some interest is that one national telephone survey on public perceptions of health risks (Scar, Langley & T aylor 2000; n=2008) included questions on "reuse of treated sewage on crops" and parklands. Despite the negative bias (omitting that the water is treated to suit its end use), around half the sample of Australians believed that the reuse fo r crop irrigation and parklands wo uld involve low or minimal health risks (approximately 47 per cent for crops, and 53 per cent for parklands) . Another more recent telephone survey (UMR Research, n=600) on behalf of the Government of New South Wales reports that 29 per cent of Sydney residents were

very, or, mildly comfortable with "drinking recycled sewage, including toilet water, that is treated to drinking water quali ty" (Davies 2005) . H owever when the process of indirect potable reuse was described, as involving the mixing of reclaimed water with rainwater in Warragamba Dam, "48 per cent supported the 'shandying' option" (Davies 2005). A similar result was generated through a Clean Up Australia poll that reported Sydney people were "46.4 per cent in favour" of "pumping treated sewage into Warragamba Dam" (Clean Water Campaign 2005). It has been argued that findi ngs from general population surveys chat focus on policy-type questions need to be differentiated from those chat present more sal ient options. In the case of recycling water, the salience of a proposal means chat its implementation is imminent (Bruvold 1988) or that bodily contact, personal use or ingestion is specified (Marks 2004:46). For example, recent work completed by CSIRO (Po et al 2005) in Melbourne and Perth queried respondents' intended acti on, rather than support for proposals. In relation to the Werribee Irrigation District's use of recycled water for crop irrigation, 35.0 per cent of respondents (n=400) indicated they would buy the vegetables without hesitation, and a further 55.5 per

Table 1. Accepta nce of recycling reclaimed water (total percentages favour/ag ree).

Golf courses Recreational parks Household gardens Car washing Home toilet flushing Vegetable crops Home laundry Showering Cooking Drinki ng

Sydney 1996 n: 1000

Sydney 1999 n=1 300

94 95 96 96

97 97

New Haven 2000 n:20 1

Mawson L. 200 1 n=20 1






95 2




58 2

45 2

32 2

Perth 1999 n=666

Mawson L. 2002 n:136


96 77 55 33 27

96 96 94 75 52 34 26


95 51 31 16



Note: 1. Face to face interviews (qualitative research) with randomly selected householders. 2. Both types of applications included in the one statement.

refereed paper


MARCH 2006 151

cent were unsure (Po et al 2005:83). Most of the reasons (n=223) fo r being unsure related to the need for reassurances that rhe water would be treated properly or that safety would be guaranteed (2005:84). In Penh, respondents considered the planned Managed Aquifer Recharge (MAR) indirect potable reuse scheme and 31.3 per cent were unconditionally willing to drink the water (200 5:52). Another 51 .0 per cent had reservations, and 17.8 per cent indicated they would nor drink the water. A total of 72.1 per cent of reasons (n=205) for being unsure related to respondents' concern about health risks (2005:53), as fou nd in the Melbourne research. As far as ocher alternative sources are concerned, the ARCWIS (I 999) study mentioned earlier found that Perth residents preferred recycling water sourced from stormwarer than reclaimed water. And in rhe ongoing recycled water vs. desalination debate, the aforementioned UMR Research srudy fou nd that "65 per cent of people support desalination" (Davies 2005). The risk literature, involving psychometric, social and cultural studies, confirms that risks are more acceptable if they are visible, voluntary, fami liar, controllable, fair, fo rgettable, acute (as opposed to long term, ch ronic effects), immediate (rather than delayed) and narural (see for example Fischhoff et al 1978, Otway & von Winterfeldt 1982, Marris & Langford 1996). Risks that are most dreaded or unacceptable are new, higher technologies experienced involuntarily that have delayed consequences for masses of people (Fischhoff et al 1978: 143; also Beck 1992, G iddens 1990). Curren t sociological though t on embodiment of risks is also relevant, as found in the growing literature on biotechnology, where acceptance is shaped by usefulness, moral and ethical considerations (e.g. Paula 2001 ). People (consumers) are particular about what they consume. To furthe r complicate matters, risk perceptions of recycled water are likely

to be mediated by trust in the technology and those that control the ongoing management of the risks (Marks 200 3, 2004; Po et al 2005; Hurlimann 2005).

Research Design and Methodology The objectives of rhe survey were to establish national benchmark data on acceptability of a range of water recycling options and to explore the influence of context (prior experience, trust), values, beliefs and relevant demographics. T he target population for the survey included householders who had experienced water restrictions over the p revious summer period. T herefo re, all capital cities except Darwin were included. The telephone survey was developed by the Flinders University researchers and administered through the University of Queensland Social Research Centre from November to December 2004 and completed in January 200 5. The total finished sample was 2504, comp rising approximately 357 respondents in each city, with an overall response rate of 29.6 per cent (that is, of 8,467 persons contacted, 2504 responded to rhe survey) . This response was lower than expected, al though it is com parable to that experienced for similar surveys (e.g. Po et al 2005). Representativeness of the sample to the seven target populations was assessed by comparison with relevant Census data (2001 ) revealing some biases on age, education and gender (our sample was older, more highly educated, with gender evenly distributed). Post hoc weights were then co nstructed to correct these biases. Sampli ng weights based on rhe relative sizes of cities were also calculated to permit estimation of representative national results. In addition, a fo llow-up survey of no n respondents (total n=400) was conducted in Sep tember 2005. The results confirm that the initial non respondents d o not vary greatly from respondents in the main sample. All resul ts reported here are weighted using post hoc and sampling weights .

Due to the large sample size, all figures reported are representative of responses that would be fou nd in the target population. That is, it can be claimed , at rhe 95 per cent confidence level, char average percentages in the populatio n would vary only slightly from the sample percentages (within an interval ranging from plus or minus one to two per cent above and below rhe percentages shown) . This paper will derail the questions put to respondents and their responses to accep ting alternative options to traditio nal mains water supply. Note that all 'don't know' responses were not offered, bur volunteered. Reporting the resulcs for each capital city is beyond the scope of chis article and will be undertaken elsewhere. Space limitations also prohibit presentation and discussion of the influence of a range of factors as well as a mo re detailed analysis of rhe reasons given by respondents for their responses.

General Acceptance of Reclaimed Water To gauge attitudes to the policy of introducing reclaimed water for non potable uses, respondents were presented with the following explanation and uses of recycled water. The results are recorded alongside each item in Table 2, and the level of support towards the applications is graphically illustrated in Figure 1. Treated wastewater or sewage effluent is normally discharged to rivers or the ocean, bur can be recla imed fo r recycling. These uses for reclaimed water are not for drinking but the reclaimed wastewater is highly rreared ro suir rhe end use. Please indicate whether you strongly favour, favour, oppose, or strongly oppose these uses of reclaimed water [as listed in Table 2]. The questions were presented to respondents in the order they appear in Table 2. Figure 1 ranks the results which confirm that recycling reclaimed water for the irrigation of golf courses, parks, gardens and flushi ng of public toilets is strongly

Table 2. Acceptance of recycling reclaimed water (percentages, n=2504). Strongly favour


Total in favour

Don't know


Strongly oppose

56.5 32.4 60.5 45.0 34.9 28.9 29.6

37.7 43.6 36.1 41.6 40.2 39.3 44.3

94. 1 76.0 96.6 86.6 75.1 68.2 74.0

1.4 3.7 0.4 1.7 3.2 3.4 3.5

3.6 18.6 2.7 10.8 19.4 24.6 19.7

0.9 1.7 0.3 0.9 2.3 3.8 2.9

For flushing toilets in public buildings Commercial laundries For irrigating golf courses, parks & gardens school yards and playing fields da iry, beef and sheep pasture vegetables a nd fruit crops vineyards

Note: - Refusals and missing data not included {involving one respondent on four responses). - At the 95% confidence level, population % would fall within :t0.1% through to ÂąI. 9% of 'total in favour' figures.

152 MARCH 2006


refereed paper

favo ured by over half the sample. Support for the remain ing uses is also confirmed with the median fall ing within the ' favo ur' category. Less enthusiasm ('strongly favou r') is indicated fo r irrigating school yards and playing fields, followed by personal contact through laundering of cloches and uses involving ind irect ingestion. Awareness of water recycling was explored with respo ndents being asked whether they had experienced any of these uses of reclai med water. A coral of 22.7 per cent had some experience, or said that they were aware of examples of water recycl ing, and 73.9 had no experience.

Willingness to Recycle Reclaimed Water Recycling reclaimed water within households was queried as fo llows: If reclai med water became available

Golf courses parks gardens Flushing toilets public buildings School yards and playing fields Commercial laundries Dairy, beef and sheep pasture Vineyards Vegetable and fruit crops








[• Strongly Favour I!!! Favour Figure 1. Percentage 'strongly favour/favour' recycling recla imed water (n =2504). Box 1. Questions on sa lient options.


your household, how wil ling would you be to use it for [Table 3) : willing without hesitation, willing with some qual ifi cations, or nor willi ng?

The various uses were given in the order they appear in Table 3, from applications char may involve indirect contact, through to chose chat are more salient, chat is, uses involving bod ily contact or indirect ingestion. As predicted, closeness of contact governed wi llingness to recycle. Even so, there is liccle difference between levels of acceptance for those uses where contact can be avoided (to ilet fl ushi ng and garden irrigation) and some ochers where contact is inevitable (hand watering and car washing). The medians for these first fo ur uses fa ll within the highest level: 'will ing without hesitation'. Obviously, there is some resistance to using reclaimed water in the laundry, with more than a quarter of the sample indicating they are not willing to do so. While a derailed analysis of reasons relating co hesitation on laundry use can not be undertaken here, the overarching theme for co ncerns with respect co one or more of the applications listed in Table 3 is concern for health risks (a rticulated as chemicals, water quality, safety, health risks, etc).

There ore at least six ways of recycling different sources of water in the home that ore currently being considered. l . Ind ividual greywoter un its con treat used water from the laundry and shower for household garden watering and toilet flushing. How willi ng would you be to hove this type of unit installed?

2. Recla imed water - that is highly treated wastewater · could be recycled for outdoor uses only, such as watering your garden. It would be distributed from a municipal treatment plant through purple pipes. How wi lling would you be to connect to this service?

3. Reclaimed water could be recycled for toilet flushi ng as well as for watering your garden, while treated rainwater from roofs could be used for other uses. If you were in the market for a new home, how willing would you be to buy into this type of housing development ii other features met your needs? 4. Stormwoter, that is, water drained from streets and other areas, con be treated for recycli ng. How willing would you be to use it for toilet flushing and garden watering?

5. Stormwoter con also be treated to a high, drinking water quality. How will ing would you be to use it for all your water needs in your household, includi ng dri nki ng?

6. Finally, seawater con be used. Desalination removes the salt and hos been on expensive option that uses a lot of energy, but is becoming more economical. If desalinated water become available, how willing would you be to use it in the some way as you use the current mains water?

Six water recycling options To increase rhe salience of water recycling six scenarios were put before respo ndents (Box l ). T hese centred on the use of different sources of water char incl uded reclai med water, greywarer, rainwater, srormwater and desalinated seawater. Rather than providing detail char would be unmanageable within rhe allotted time and could tax respondents' patience, the

different options were summarised co incl ude distinguishing features of those that are either established or under consideration in different pares of Australia. Results of this line of questioning on willingness are as listed in T able 4, in the order given to respondents. City householders are willing to recycle all these sources of water with preference given (over 60% 'willing without hesitation', Table 4) fo r non potable uses of reclaimed

Table 3. W illingness to recycle recla imed water for th e househo ld (percentages, n=2504) . Willing without hesitation

Willing with some qualifications

Total willing

Don't know

Toilet flu shing


Gorden irrigation

80.8 80. 1

18.8 15.1 15. 1 14. l 34.6 32.6

96.5 95.9 95.1 91.3 73.3 67.4

l.O 0.8

Hand watering Car washing Washing machine Hand washing


77.2 38.7 34.7

0.5 l.3 0.8 l. 1

Not willing

2.6 3.2 4.4 7.4 26.0 31.5

· Refusals and missing data not included {involving from three to six respondents of the total sample). · At 95% confidence level, population percentages would foll within ± 0.7% through to ± 1. 9% of 'total willing'.

refereed paper


MARCH 2006


Table 4. Willingness to use water through sa lient options (percentages, n=25O4) Willing without hesitation

Willing with some qualifications

Grey water installed for garden a nd toilet flushing


Reclaimed water for garden


Reclaimed water for ga rden, toilet and rainwater for all other uses


Stormwater for garden and toilet


Total willing

Don't know












3 .1





1. 1




Stormwater for all uses including drinking




Desalinated seawater for all uses

5 1.4



Not willing

Note:· Refusals and missing data not included {involving from two to seven respondents of the total sample). · The medians fall within the 'willing without hesitation' category except for potable uses of stormwater where the median is within 'willing with some qualifications'. · At the 95% confidence level, percentages for the population would Fall within ±0.8% to ±I. 9% of 'total willing'. water and stormwarer. Taken together with 'som e qualifications', over 90 per cent of respondents would be willing to undertake recycling on their property for toiler flushing and garden watering, or accept the desalination alternative. The ranking of rhe high positive response is depicted in Figure 2 wh ich illustrates highest acceptance (willingness without hesitation) of non potable uses of stormwater while willingness to use stormwater for all household needs including d rinking is ranked well below all other options. Following the questions on stormwarer (Box 1, questions 4 and 5), respondents were asked for their level of con fidence in specific potable uses, and rhe resulrs are presented in the n ext section.

Stormwater toilet garden Reclaimed toilet garden, rainwater Grey water toilet garden Reclaimed outdoor only Desalination all uses Stormwater all uses

0 10 20 30 40 50 60 70 80 90 10 0

• Willing without hesitation 1:1 Willing with some qualifications Figure 2. Willingness to use water through sa lient options (percentages, n=25O4).

Potable Reuse Potable reuse involving reclaimed water was treated as a separate question from the six scenarios to m ake respondents more aware of what was being proposed. W hen finalising the wording to these questions, rhe advice of proponents of potable reuse was sought and some were concerned char the descriptions, that include the word 'sewage', would "put people off' or, worse, "would cause a stir". H owever, the aim of the exercise was to ensure transparency in communicating the concept, to assist people in making an informed choice, and to ensure rhar there would be no surprises for rhem in rhe future when indirect potable reuse migh t be considered for implementation in their city. The wording of the introductory statement is as fo llows: Reclaimed water - that is, water reclaimed from wastewater or sewage effluent - can also be treated to drinking water quality. It can then be mixed with traditional sources, such as water collected in reservoirs, and then treated and piped in the usual way to the whole ciry or town. Have you heard of this before?

154 MARCH 2006


A total o f 35.0 per cent of respondents had heard of the 'i nd irect potable reuse' concept, with 64. 3 per cent saying they had not heard of this form of recycling reclaimed water (17 were unsure, 0.7 per cent). Respondents were then asked if they would be willing to use the water: How willing would you be to use water mixed with reclaimed water, treated to dri nking water quality, for all your household needs? As with previous questions, there were few missing data (1 refusal, 4 missing data). A total of 22.2 per cent were willing to use the water without hesitation, 50.5 per cent were willing with some qualifications, 26.0 per cent are nor willing and only 1.3 per cent expressed uncertainty (unsure/don't know) . The median falls within the category 'willing with some qualifications' and a roral of72.7 per cent are willing to recycle the water in this way, albeit with some qualifications. To build salience of rhe issue, respondents were then asked how much confidence they would have in usi ng it for showering, cooking and drinking. The same question

was asked of rhem earlier in relation to using drinking water quality stormwater. Boch results, reflecting trust in potable reuse, are compared in Table 5, Figure 3 . This exercise provided respondents with the opportunity to give closer consideration to potable reuse and , not surprisi ngly, overall, the result is more conservative than the 'willingness co use' questions. The med ian result lies within ' moderate confidence ' for all but d rinking where ir rests within ' nor much confidence'. However, respondents are more 'confident' on rhe more salien t options of showering with stormwater or reclaimed water (Table 5 : T otal confident) than they are 'willing to use' either water source fo r all household needs (a more general, less salient measure), being 73 .6 per cent (stormwarer, Table 4) and 72.7 per cent (reclaimed water) respectively. Although a derailed analysis of qualified responses cannot be included here, the main comments given by respondents at rhe end of chis section can be summarised as concerns about health risks and the need to have more information made available.

refereed paper

Table 5. Confidence in potable uses of stormwater, or reclaimed water through indirect pota ble reuse (percentages, n=2504) . n=

Great confidence

Moderate confidence

Total confident

Don't know

Not much confidence

No confidence

2498 2499 2496

42 .0 25. l 19.4

39.6 31.0 26.8

81.6 56.1 46.2

0.5 0.7 0.5

11.8 18.8 19.6

6.2 24.4 33 .6

2498 2500 2498

36.5 21.1 14.7

39.3 33.0 26.9

75.8 54.1 41.5

0.5 0.4 0.7

13.0 17.2 20.0

10.7 28.3 37.8

Drinking water quality stormwater showering cooking drinking Water mixed with reclaimed water treated to drinking water quality showering cooking drinking

Note: - Refusals and missing dato not included {four to eight). - The medians fall within the 'total positive' category except for drinking where they rest in 'not much confidence'. - At the 95% confidence level, percentages for the population would fall within Âą 1.6% to Âą2% of 'total positive'.

In addition to the earlier question on prior awareness, respondenrs were given a description of 'unplanned indirect potable reuse'. T his was included to build knowledge while gauging public awareness of this aspect of rhe water cycle rhar is so fa miliar to water engineers. In terviewers were instructed to read the followi ng explanation very slowly: One way of chinking about recycling water is char ic already occurs wherever warer su pply for drinking is withdrawn downs tream of ocher cities and cowns chat discharge sewage effluent in co rivers; for example, the Thames River and rhe Murray River. H ave you ever though t abou t water s upply in chis way?

So me 67. l per cent had not heard of the concept while almost a third had (32.2 per

shower stom,water shower reclaimed water cook stormwater cook reclaimed water drink stormwater drink reclaimed water -,-






[- Great confidence





90 100

~ Moderate confidenc~

Figure 3. Confidence in stormwater a nd indirect potable reuse (percentages, n=2504).

Beneficial Solutions to Organic Waste Pro Professional services for biosolids and organic waste management. Land application of orga,oic wastes to agriculture, fo~estry and land reha bilitation. Complete range of specialised biosolids injection and spreading equipment. Biosolids, soil and water monitoring and reporting. Biosolids and effluent management plans and EPA approvals. 19 Riverleig h Ave Gerroa NSW 2534 ph 02 42 34 4444 fax 02 4234 4422 email lrawlinson@lvra.com.au

refereed paper


MARCH 2006 155

cent); again, only 0 .7 per cent were unsure whether chey had o r had not.

Willingness to Pay Respondents were asked co consider paying a realistic price for water in the event chat recycled water had co b e implemented in order co supplem ent the water supply. Note chat chis question came at the end of all q uescions on water recycling and refers co scenarios chat include the reticulation of either reclaimed water or scormwacer, while not specifically nominating non potable or potable uses. The aim was co test resistance co these sources of water, while introducing che concep t of more realistic pricing of recycled water.

If we had co increase water supplies, and there were two options one - rhar you pay double the price you pay now fo r water so char additional water can be captured through new reservoirs, or from rivers, ere.; or, two - rhar you use some form of recycled wastewater or srormwacer and pay che same price for chat as you pay now for water which would you support: pay double what you pay now, or recycle?

A coral of79. l per cent chose recycling water sourced from wastewater or scormwacer charged at the same race chat they now pay fo r mains water. Another 11.5 per cellt would prefer co pay double chat price in order not co recycle water, and 6.6 per cellt chose neither option, wanting some alternative such as more water conservation (coral 'don't know' : 2.7 per cent, with only four refusals).

Discussion The general acceptance 'policy' questions on recycling reclaimed water yield high levels of support (Table 2). Acceptance grades from lease likelihood of direct contact - being che irrigation of golf courses, parks and gardens - through co chose involving contact with the body via clothing (commercial laundries) , or through indirect ingestion (meat, vegetables, fruit, wine). Looking at the results for the more salient proposition of willingness co recycle reclaimed water for non potable uses (Table 3), the pattern of ranking reveals chat people are prepared co handle the water for hand watering and car washing (bodily contact) but are less willing co use it for laundering cloches, where che level of acceptance is similar co chat for commercial

30th Hydrology & Water Resources

SYMPOSIUM Past Present & Future Hotel Grand Chancellor Launceston 4 - 7 December 2006

laundries. In pare, these results confirm previous find ings, particularly the comparable results for Sydney (Table 1): golf courses, recreational parks, home coilec flu shing and laundry. However, national support for the irrigation of vegetable crops is well sho re of chat previously reported for Sydney and Adelaide. This may refl ect general health risk concerns as found previously by Scar et al (2000) and Po et al (2005) combined with a lack of fam iliarity with working examples of water recycling. Reclaimed water is raced highly against ocher alternative .s ources when willingness co recycle water in the household is further explored (Table 4). For these salient options, che m edian for all non potable uses fa lls within 'wi lli ng without hesitation', although enthusiasm is dampened somewhat for existing properties co con nect co a reclaimed water service for outdoor uses, and for the inscallacion of greywacer units. Rainwater is the preferred drinking source, and desalination is favoured over scormwacer for all household uses. When the idea of potable reuse is explored, it is confirm ed chat: (a) on the face of ic, scormwacer is preferred over reclaimed water (Table 5), but not co the extent found in the ARCWIS (1999) Perch study (the difference is negligible: from 2 co 6 per cent); and (b) people have significancly less confidence in ingesting che water compared co using it for showering. Boch willingness and confidence in potable uses of reclaimed water, however, are much higher than chat indicated in previous surveys (Table 1). These national survey results are more in line with chose produced by the latest Sydney and Perch measures of acceptance: in all three cases, respondents were given a description of indirect potable reuse systems.

A fina l assessment of people's willingness co

Call for Papers available at www.cdesign.eom.au/hydrology2006 Abstracts due now Keynote Speakers: Dr Walter Boughton, Honorary Senior Fellow, Griffith University A/Professor Ron Cox, Director, Water Research Laboratory, University New South Wales Dr Rory Nathan, Principal Hydrologist, Sinclair Knight Merz General Enquiries: Conference Design Pty Ltd t: 03 6224 3773 f: 03 6224 3774 e: info@cdesign.com.au www.cdesign.com.au

156 MARCH 2006


recycle water is provided in the 'willingness co pay' question. If the choice had co be made co combat depleting water resources, 79 per cent would not only use recycled water but would be prepared co pay che same p rice as they currencly pay for mains water. On the ocher hand, in relation co current prices of mains water, 12 per cent opted co pay double the price than use recycled water. The results on prior experience or awareness are of interest in assessing the level of background familiarity people have with chis new technology. Only around a third or less of the national sample have had exposure co the idea of recycling water. A coca! of 74 per cent have had no experience with non potable reuse, 64 per cent have not heard of blending drinking water q uality reclaimed water with current

refereed paper

water supplies before ir is treated and reticulated (indirect potable reuse), and 67 per cent have never contemplated that their water supplies may be supplemented by upstream environmental discharges of sewage effluent (unplanned potable reuse).

Conclusions In rhe absence of comprehensive findings on perceptions of water recycling across Australia, these results provide national baseline data to outline potential public acceptance of alternatives to tradi tional water supplies. The study confirms the pattern of acceptance previously reported for Sydney, Perth and rwo sires in Adelaide: acceptance is ranked roughly in accordance with rhe degree of bodily co ntact. However, the current data clarifies chis ordering in chat the nature of rhe contact is important: there is a strong willi ngness to handle reclaimed water for car washing and hand watering; a decli ne in willi ngness when its use becomes personalised, in washing clothes and the body; and less confidence for uses that involve ingestion. The salience of rhe proposal helps exp lain this ordering. Th is study also provides evidence char Australians are willing to use a range of alrernarive sources of water and char, despite the lack of familiarity with rhe concept, and given a frank description of indirect potable reuse, they are more prepared to co nsider potable reuse than some commun ities have in rhe past. Further deliberations such as the role of prior experience and awareness, risk perceptions and trust, will be considered in subsequent reports on these national findings that will model a range of facto rs that influence levels of accepta nce.

CSIRO Urban Warer Program, CSIRO Land and W ater. Beck U. (1992) Risk society: Towards a new modernity, Sage, London. Bruvold W .H . (1988) Public opinion on water reuse options, Journal of Water Pollution Control Federation, Vol.60 (1), 45-49. C lean Water Campaign (2005) Newspoll survey shows Syd neysiders are open to water reuse, Media Centre, Today's News, Clean Water Campaign, 15 August. Accessed 23 August, URL http://www.cleanup.com.au. Davies A. (2005) Recycling water too distasteful, says Sartor, Sydney Morning Herald, 22 July. Accessed 22 July, URL http:// www.smh. com.au/ news/ national. Fischhoff B., P . Slovic, S. Lichtenstein, S. Read and B. Combs ( 1978) H ow safe is safe enough? A psychomet ric st udy of attitudes towards technological risks and benefits, Policy Sciences, 9, I 27- 152. Giddens A. ( 1990) The consequences of modernity, Poli ty Press, Cambridge. H urlimann A. (2005) Is recycled water use risky? An urban Aust ralian community perspective, Environmental Research Even t, Hobart, 29 November - 2 December. Marks J.S. (2003) The experience of urban water recycling and the development of t rust, PhD Thesis, T he Flinders University of South Australia. Marks J .S. (2004) Advancing community acceptance of reclaimed water, Water 3 1:5, August, 46-51.

Marris C. and I. Langford ( 1996) No cause for alarm, New Scientist, 28 September, 36-39. M cKay J. and A Hurlimann (2003) Attitudes to reclaimed water for domestic use: Part l Age: Young people are marginally more supportive, Water, 30:5, August, 45-49. O rway H.J. and D. von Winterfeldr (I 982) Beyond acceptable risk: On the social accep tab ility of technologies, Policy Sciences, 14, 247-256. Paula L. (2001 ) Ethics: the key to public acceptance of biotechnology? Biotechnology and Development Monitor, 47, 22-23. Po M., B.E. Nancarrow, Z. Leviston, N .B. Porter, G.J. Syme and J.D. Kaercher (2005) Predicting community behaviour in relatio n to wastewater reuse: W hat drives decisions to accept or reject? Water fo r a Healthy C ountry National Research Flagship, CSIRO Land and Water, Perth. Roseth N. (2000) C ommunity views on recycled warer, Proceedings Enviro 2000, Towards Sustainability, 9- 13 April , Sydney. Scar G., A. Langley and A. Taylor (2000) Environmental healrh risk perception in Australia: A research report to the Commonwealth Department of H ealth and Aged Care, Centre for Population Studies in Epidemiology, Department of Human Services, Adelaide. Sydney Water (l 996) C ommuniry views on water reuse, Research Report, June, 1- 15.

The Authors Dr June Marks is a Research Fellow investigating public acceptance of water recycling with Assoc. Professor Bill Martin and Dr Maria Zadoroznyj in rhe Department of Sociology ar Flinders University of South Australia (email: june.marks@flinders.edu. au, [08] 820 1 2628) .

Acknowledgments An earlier edition of this paper was

presented at the 4th Australian Water Summit, February 2006, Melbourne. The research is supported by an Australian Research Counci l (ARC) Project Linkage grant in partnership with the Adelaide water services operator, United Water Inrernarional Pry Lrd.

References Australian Research Centre for Water in Society (I 999) The social basis for urban water provision in the 2 1st century, Report,

refereed paper

EARTH TECH . Delivering a susta inable future. As Australia¡s private sector leader in water recycling. Earth Tech develops sustainable solutions to today's challenges that improve the quality of life tomorrow. Our dedication to excellence has contributed to the following award -winning projects: Campaspe Water Reclamation Scheme IEAust !Vic] 2005 Engineering Excellence Award for Environment and Sustamability Sandhurst Club UDIA !Viel 2005 Urban Development Excellence Award for Water Sensitive Urban Design www.earthtech.com.au A BETTER TOMORROW made possible


MARCH 2006 157

GREEN BUILDING INNOVATION AWARD VRTX chemical-free water treatmen t has been awarded the prestigious AH R Expo Green Build ing Innovation Award 2006 by ASHRAE . This recognises the system as a sustainable solution for successfu lly controlling bacteria, scale, corrosion and, most importantly, delivering water and energy savings in building cooling water systems. The compet ition is conducted each year by ASHRAE journal and sponsored by ASHRAE, rhe Air-Conditioni ng and Refrigeration Institute (ARI) and rhe Internacional Exposition Company (IEC). Th e Green Building Award recognises products that make build ings 'green', favou ring the environment, conservation of natural resou rces and the reduction of energy consum p tion and pollution. VRTX uses a powerful combination of kinetic energy, intense hydrodynamic cavitation and mechanical shear to d rive chemistry changes in the water and destroy bacteria.

Water Business aims co keep readers alert to business news and new product releases within rhe water sector. Media releases should be emailed to Brian Raulr ar brian.raulr@halledir.com.au or Tel (03) 8534 5014. AWA wish to advise readers rhac Water Business information is supplied by third parties and as such, AWA are nor responsible fo r the accuracy, or oth erwise, of th e information submitted . picture of this effective and sustainable water treatment solution for rh e bu ilding environment.

Contact Roland van Amstel, National Safes Manager Mobile: 0438 077 572, Fax: 02 9984 9187 Email: water@imatech.com.au

DIAPHRAGM CONTROL VALVEFOR WATER A self-adjusting valve with zero dead volume is n ow available from Bu rkert Fluid Control Systems. T h e pneumatically operated Type 2730 Diaphragm Control Valve, with plastic body, comprises hermetical separation of flu ids from rhe operating mechanism by the diaphragm , and is rated for use with p otable water and sterile applications as well as with aggressive waste fluids.

VRTX systems have a very small footprint.

T he unit includes a pneumatically operated p iston actuator, which is designed to facilitate frequent alteration to the defined stroke. The 2730 is thus suited to applicatio ns and p rocesses in which a controlled flow-rare is freque ntly altered.

Keeping systems clean from scale and biofilm imp roves the heat-exchange efficiency in rhe system, resulting in considerable energy savings. Bacteria counts are kept at very low levels by keeping the water clean from nutrients and contin ually destroying any new bacteria growth. Very good ro excellent corrosion results and im proved worker safety co mplete the

T he 2730 is suited to stand-alone operation, and is also designed for troublefree inclusion in a customised modular component or as a building block in a comprehensive water or waste system. T h e 2730 can be actuated by rop or side control posirioners, TopControl 8630, SideConrrol 1067 or SideControl 8635, for continuous regulation.

The 2730 Diaphragm Control Valve has an optio nal, integrated PI D controller, to enable linki ng with flow, level, pressure or temperature sensors to perform stand-alon e p rocess control. Bu rkert offers a comprehensive range of instrumentation that ideally connects to the 2730 and is suitable for almost any application. With a global network spanning offices in more than 40 countries and operational outposts in an additional 80, Burkert Fluid Control Systems is an established and mature global organisation celebrating its 60th an niversary in 200 6.

Burkert technical experts may be contacted on sales.au@burkert.com, and 1300 888 868. Sales inquiries to sales.au@burkert.com or 1300 888 868.

ITT APPOINTS NEW DISTRIBUTORS ITT Fluid Technology Australia, a d ivision of ITT I ndustries Residential and Commercial Water Group, is pleased to ann ounce that ir has appointed two new distributors for the regions of Western Australia, and South Australia and the Northern Territory. The new d istributors will support the sales of ITT's diverse pump and associated product range which in clude the well know b rands of Lowara, Goulds, H ydrovar and Marlow in the clean water industries. Applied Pump Tech nologies, located in Welshpool, Perth, will look after the Western Australian market. It is headed by Kim Dean who has 25 years experience in business management including 14 years in the pump and irrigation markets.

WATER GEMSÂŽ THE TOP CHOICE AMONG WATER PROFESSIONALS WaterGEMS version 8 brings the only platform-independent solution to the world of water distribution, allowing engineers to build, analyse, and map water distribution models without being anchored to a specific engineering design or geospatial platform. For more information about this new version, see the inside front cover of the April issue of the A WA's Water Journal, visit our booth at Enviro06, visit www.bentley.com/awa, or e-mail anz.marketing @bentley.com

158 MARCH 2006