Water Journal September 2006

Page 1


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


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


Being anchored to a single platform is costly and risky. WaterGEMS V8 is the only solution that lets everyone work within their favorite platform, sha ring a single model file.

EASIER THAN EVER! The legendary ease of use of Haestad Methods solutions is magnified with WaterGEMS V8. Enjoy a fresh new stand-alone interface, an improved ArcGIS integration, and an all -new M icroStation platform with dozens of new features and ground breaking uti lities: • Critica lity analysis

• Stand-alone GIS demand allocation

• Network trace queries

• Network segmentation analysis

• Stand-alone Terrain Extraction

• Pressure-dependant demands

• New isolation val ve element

• Network nav igator

• Variable speed pump batteries

• Centralised water use management

• New hydrant element

• Unit demand engi neeri ng libraries

• Dynam ic SOL-based queries

• Auxil iary fire flow results


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

Register now for WaterGEMS courses: Brisbane: 14-17 Nov. 2006 ~ : 27-30 Nov. 2006 Visit www bentley com/ en-au/training or contact us fo r more informat ion or to register.

Journal of the Australian Water Association

Volume 33 No 6 September 2006

OPINION AND INDUSTRY NEWS OPINION Growing Our Networks - Unleashing Our Volunteers Management Models Under the Microscope? My Point of View

DDoy, President, AWA CDavis, CEO, AWA BSydes, Principal Solicitor with the Environment Defenders Office, Victoria

AWA NEWS Includes: Water Education Network; Young Water Professionals; Toowoomba Water Recycling Forum; National Specialist Networks; WaterAid Australia


5 6


CROSSCURRENT Water Industry News, National Issues & Policy, International, State, Personalia,






Master Class 2006 - Evaluating Water Recycling Projects


TECHNICAL FEATURES ( ·, indicates the paper has been refereed) HEALTH-RELATED ISSUES International Symposium on Health Aspects of Calcium and Magnesium in Drinking Water Conference Report, Baltimore, Maryland, USA, April 2006

Reported by MSinclair






J Poon, AKeegan, DDeere, ADavison, PMonis


HChapman, FLeusch


PSelmes, RBain, ADomanti, WZillmann, I Cameron


DFraser, PLambie


[ii NDMA - An Emerging Issue for the Water Industry Areviewof ifs formation, analysis, removal by UV Swimming and Spa Pool Atmospheres The formation of atmospheric contaminants by chlorination of indoor pools UV Disinfection for Class AWater Recycling Validated using the latest MS2 bioassay techniques

[i] Water Recycling: Endocrine Disrupting Compounds Negligible amounts found in treated effluents

RECYCLING Toowoomba Water Supply Options The preferred option was indirect potable recycle



Groundwater Models and Salinity Risk Assessment Comparing four tried and true groundwater models to assist project managers


[I] Tubular Polymeric Aerators for Wastewater Treatment Ukrainian tubular aerators are operating in 15 countries around the world

RGalich, YMeshengisser, VLos, PSpiridonov 72





OUR COVER Melbourne Water is producing Victorian EPA Class A recycled water for irrigation of vegetables which could be eaten uncooked. UV irradiation for the inactivation of small counts of Cryptosporidium was adopted and validated using the latest bioassay methods. Our photo is of the UV system which was installed, commissioned and validated against a challenging deadline. Full details on page 56. Photo courtesy of Melbourne Water. Journal of the Australian Water Association





'Promoting the sustainable manatTement O1water , ()'

POSTAL ADDRESS PO Box 388, ARTARMON NSW 1570 EMAIL info@awa.asn.au

WEBSITE http://www.awa.asn .au PRESIDENT Darryl Day - president@awa.asn.au

CHIEF EXECUTIVE OFFICER Chris Davis - cdavis@awa.asn.au

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

EVENTS Linda Phillips - 61 2 9495 9914 lphillips@awa.asn.au

MEMBERSHIP INFORMATION AND INQUIRIES Michael Seller - 02 658 1 3483 mseller@awa.asn.au

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

MEDIA AND MARKETING Jennifer Sage - jsage@awa.asn.au

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

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

NATIONAL SPECIALIST NETWORK Laura Evanson -6 1 2 9495 9917 levanson@awa.asn.au

AWA BRANCHES: AUSTRALIAN CAPITAL TERRITORY and NEW SOUTH WALES Errin Dryden - 61 2 9495 9908 edryden@awa.asn.au NORTHERN TERRITORY c/o Ian Jarman - 61 2 9495 99 11 ijarman@awa.asn.au SOUTH AUSTRALIA Sarah Carey - 61 8 8267 1783 sabranch@awa.asn.au QUEENSLAND Kathy Bourbon - 61 7 3397 5644 awaq@awa.asn.au TASMANIA c/o Ian Jarman - 6 1 2 9495 9911 ijarman@awa.asn.au VICTORIA Joe Owzinsky - 61 3 9509 2748 awwa@i.net. au WESTERN AUSTRALIA Cath Miller - 0416 289 075 cmiller@awa.asn.au INTERNATIONAL WATER ASSOCIATION, AUST. (IWAA) c/o Chris Davis - cdavis@awa.asn.au

DISCLAIMER Austral ian Water Association assumes no responsibility for opinion or sta tements of facts expressed by contributors or advertisers.

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



Journal of the Australian Water Association

Volume 33 No 6 September 2006

ISSN 0310-0367

AWA WATER JOURNAL MISSION STATEMENT 'To provide a print ;ournal that interests and informs on water matters, Australian and international, covering technological, environmental, economic and social aspects, and to provide a repository of useful refereed papers.' PUBLISH DATES Water Journal is published eight times per year: February, March, May, June, August, September, November and December EDITORIAL BOARD: Chairman: FR Bishop BN Anderson, TAnderson, CDiaper, GFinlayson, AGibson, GA Holder, BLabzo, MMuntisov, CPorter, DPower, FRoddick EDITORIAL SUBMISSIONS Water Journal invites editorial submissions for: Technical Papers and topical articles, Opinion, News, New Products and Business Information. Acceptance of editorial submissions is subject to editorial board discretion. Email your submissions to one of the following three categories: 1. TECHNICAL PAPERS AND FEATURES Bob Swinton, Technical Editor, Water Journal: bswinton@bigpond.net.au AND https://zeus.econ.umd.edu/wj (Editorial Express) Papers of 3000-4000 words (allowing for graphics); or topical stories of up to 2,000 words. relating to all areas of the water cycle and water business. Submissions are tabled at monthly editorial board meetings and where appropriale are assigned to referees. Referee comments will be forwarded to the principal aulhor for further action. See box on page 71 for more details. 2. OPINION, INDUSTRY NEWS, PROFESSIONAL DEVELOPMENT Jennifer Sage, jsage@awa.asn.au Articles of 1000 words or less 3. WATER BUSINESS Brian Rault, National Sales & Advertising Manager, Hallmark Editions brian.rault@halledit.com.au Water Business updates readers on new products and associated business news within the water sector. ADVERTISING Brian Rault, National Sales & Advertising Manager, Hallmark Editions Tel: 61 3 8534 5014 (direct), 6138534 5000 (switch), brian.rault@halledit.com.au Advertisements are included as an information service to readers and are reviewed before publication to ensure relevance to the water environment and objectives of AWA. PURCHASING WATER JOURNAL Single issues available@ $12.50 plus postage and handling; email dwiesner@awa.asn.au BACK ISSUES Water Journal back issues are available to AWA members at www.awa.asn.au PUBLISHER Hallmark Editions, PO BOX 84, HAMPTON, VICTORIA 3188 Tel: 61 3 8534 5000 Fax: 61 3 9530 8911 Email: hallmark.editions@halledit.com.au

Journal of the Australian Water Association


Since its inception in 2003 Veolia has sponsored the UWP. Veolia lend their support to the UWP viewing it as an important vehicle for promoting education and growth in the next generation of water professionals.

technical features


health-related issues

INTERNATIONAL SYMPOSIUM ON HEALTH ASPECTS OF CALCIUM AND MAGNESIUM IN DRINKING WATER BALTIMORE, MARYLAND, USA, APRIL 2006 Reported by Martha Sinclair This Symposium was organised by the World H ealth Organization (WHO) in response to requests from several member states for guidance on the possible long term health effects of d rinking demineralised water prepared by desalination processes. This has been rapidly expanding in recent decades d ue ro fa lling costs and scarcity of ocher water sources. Several countries in che Middle Ease now rely on desalination as cheir main source of drinking water. The Symposium followed an expert meeting held in Rome in November 2003, which produced a report Nutrients in Drinking

Water with a main focus on the relationship between water hardness and cardiovascular disease mortality (WHO 2005).

health-based recommendations could be made about minimum hardness levels in d rinking water.

Epidemiological studies on this topic dace back as far as 1957, and many have reported that populations living in hard water areas have lower mortality rates from cardiovascular d iseases than those living in soft water areas. Magnesium tended to show a stronger relatio nship with cardiovascular d isease mortality than calcium, however the quality of most studies has been poor. Prior reviews by WHO concluded that the evidence was insufficient and therefore no

However the participants at the 2003 WHO meeting concluded that: ... on balance, the hypothes is chat consumption of hard water is associated with a somewhat lowered risk of cardiovascular disease was probably valid, and chat magnesium was the more likely contributor of chose benefits.

The Baltimore Symposium was convened to further discuss this issue and help define what questions need to be resolved before


~Jater solut i ons

Commercial and Industrial Rainwater Harvesting Solutions • Free of water restrictions • Water free of charge • Catch water where it falls • Help supplement water supplies • Reduce stormwater runoff At ECS we can design: With an on-site packaged wastewater treatment system you can protect your large scale landscaping investment by using 'A Class qualify recycled water in your irrigation network. A packaged plant is compact, simple to operate and provides a continuous flow of water so you can keep those sprinklers going. The recycled water is also su itable for industrial applications. Contact us for a quotation on a customised plant to suit your specific requirements. Tel. 07 3279 3276 www.epco.com.au

'A Class water or basic treatment EPCO has the solution.



Journal of the Australian Wate r Association

• Rainwater tanks with no overflows • Peak rain flow diversion • Innovative storage solutions • Energy efficient interface with existing potable water supply.


Using sophisticated design models we can even tie it all intg__an Energy Performaoce Contract at no risk to you.

Interested? Thought about it? Then contact us today at ECS e w

02 9983 1144 water@ecsaustralia.com www.ecsaustralia.com

•• energy • • • conservation •• systems

minimum guideline values for calcium and magnesium in drinking water are justified on h ealth grounds. T he Symposium was opened by D r Jamie Bartram, WHO Coord inator fo r Water, Sanitation and H ealth, and was attended by more than 150 d elegates from 24 countries.

The Role of Magnesium in Cardiovascular Health Magnesium is a critical element for cardiovascular health. There is abundant evidence that severe deficiency of magnesium can cause irregular heart beat (arrhythmia) and ocher adverse changes in cardiovascular fu nction. H owever, clinical trials of magnesium supplementation for the rrearment of heart arrack, stroke and high blood p ressure have produced inconsistent results with no clear evidence of b enefi r. There is some evidence that high magnesium intake may have favou rable effects on blood cholesterol levels. Th ree mechanisms of action (preventing arrhythmia, lowering blood pressure and lowering cholesterol levels) have been proposed by d ifferent researchers to explain the observed inverse relationship berween water hardness and cardiovascular mortality, with the arrhythmia hypoth esis being the most popular over the years.

Requirements and Intake Dietary calcium requirements are well established in terms of che amounts needed for normal bone growth in children and maintaini ng the health of bones in adults. The recommended values are similar in different countries and lie in the range of 1000 to 1300 mg/day for both men and women . High calciu m intake may have beneficial effects on hypertension. Information on the required intake of magnesium is much less certain. Although magnesium has multiple biological roles, no simple biological marker has been established to assess magnesium status or determine whether dietary intake is sufficient. There is considerable variation in recommended intakes:

WHAT WE VALUE ... At a time of compellin g environmental challenges and business opportunities - our core values have never been more important:

·: .. uncompromising in our determination to achieve product excellence and in turn, improve the world's quality of life... Whether you're looking for solutions for your business or your next career move MWH offers a world of opportunities and proven outcomes .


• WHO/FAO has set th e Recommended Nutrient Intake (RNI) for magnesium at 260 mg/day fo r males and 22 0mg/d ay fo r females. These intakes are defined as "meeting the nutrient requirements of almost all (9 7 .5 per cent) apparently healthy individuals in an age and sex-specific population group". • in the US and Canada the Recommended Dietary Allowance (defi ned in the same way as the RNI) is sec at 420 and 320 mg/day respect ively fo r males and females. • the European Community defines a broad range of "acceptable intakes" for magnesium of 150-500 mg/day. Dairy products are the richest source of dietary calcium. Efficient calcium absorption requires Vitamin D (from the diet or synthes ised in the body in response to sun exposure). Dietary sources o f magnesium are more varied. Dairy products, vegetables, whole grains, fruit an d nuts are important contributors to dietary magnesium. The chlorophyll molecule contain s magnesium, so green vegetables are a significant source. Dietary intakes of calcium are reasonably well documented in developed nations, b ur information on intake of magnesium is more limited. H owever, it is estimated that many people have inadequate magnesium intake from fo od , although severe magnesium d eficiency attributable to poor diet appears to be rare. Such assessments of d ietary intakes usually do nor consider intake from water.


Increased flow capacity Long term hydrostatic strength Extraordinary resistance to impact Half the weight - reduces OH&S risks Size range 100 - 375mm Series 2, PN 12.5 to PN 20 Half the embodied energy Half the raw materials 100% recyclable No heavy metal stabilisers Energy savings in pumping costs Manufactured to comply with AS 4441 (lnt):2003 Minimum lifespan of 100 years Australian innovation

For more information on Supermain or other great products - Contact Vinidex TODAY!



13 11 69 info@vinidex.com.au www.vinidex.com.au

Vinicln Systems & Solutions

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



Company ISO 0001 Uc 570 SAIGlobol

Water as a Source of Calcium and Magnesium

Contra-Shear Technology



._.._ UV-techniek

Specialists in Inlet Works Supply • Complete Range of Fine Screens • 360° Pista Vortex System

The mineral content of reticulated drinking water varies greatly around the world and often within individual countries. Generally speaking, surface waters tend co be lower in calci um and magnesiu m than groundwaters. Even in countries where hardness is measured, it may not be reported co regulatory agencies and therefore data may nor be readily obtainable from a centralised source. T he mineral content of bottled waters is highly variable, as are the requi rements or constraints for labelling and treatment of such waters in different countries. The variability in water co mposition and individual water intake means chat the contribution of drinking water co coral mineral in cake is likely co be highly variable in different populations and between individuals within a population. In situations where dietary in cake of calci um and/or magnesium is low, drinking water with high content of these min erals may contribute a significant part of coral intake.

Magnesium Deficiency le is di fficult co attrib ute specific symptoms or diseases co dietary deficiency of a si ngle nutrient. Diets which are poor in magnesium may rend co be high in sugar and fats, and these components may contribute co obesity and ocher cardiovascular risk factors independently of any effect of lack of magnesium .

Combined Inlet

The level of magnesium in seru m can be measured, but ch is does not appear co correlate well co the levels of magnesium in soft tissues. However levels in seru m and bone do seem co co rrelate with each ocher. le is believed chat magnesium in bone may act as a reservoir chat can be released co supplement ocher tissues in response co inadequate dietary intakes.

Pista 360°

Tertiary Treatment • Microfi lters to 5 micron • Berson Ultraviolet Disinfection

• • • • •

WAM Lime Handling Sludge Dew atering High Rate Anaerobic Treatment Clarifiers Package Plants

CONTRA-SHEAR TECHNOLOGY www.cstechnology.com.au Tel: (02) 9427 1279 Fax: (02) 9427 3354 Bamajan P/L trading as Contra-Shear Technology ABN: 35 001 226 160

44 SEPTEMBER 2006 Water

Journal of the Australian Water Association

Epidemiological Evidence Dr Paul H unter from the UK discussed the fi nd ings of a review of the published epidemiological evidence co mmissioned by the UK Drinking Water Inspectorate (Carl ing, Abu baker et al. 2005). Despite che poor quality of most studies, he scared chat in his opinion there probab ly was a protective effect of magnesium on card iovascular morcalicy but nor on ca rdiovascular disease incidence. ln ocher words, people drinking water with high magnesium content are just as likely co develop cardiovascula r disease, but less likely co die from it, chan people drinking water with low magnesium content. He also felt there was no effect of magnesi um on cancer risks, or of calci um or hardness on cardiovascular disease or mortality races. He supported these conclusions by showing a graph ploccing risks of deacl1 from various cardiovascular diseases from che five case-co ntrol studies where a significant protective effect was observed for magnesi um at levels above 8 co 10 mg/L., and noted the similarity in fi ndings. However, chese five studies do not represent fi ve entirely independent settings nor did they consider smoking or ocher individual risk factors. My presentation on behalf of the Global Water Research Coalition outlined che shortcomings of studies conducted co dace, including use of routine cause of death scaciscics (which have been shown co have error races of up co 30% fo r cardiovascular deaths) and failure co account for the effects of risk factors such as smoking (which may exert an effect on cardiovascular disease races up co 20 times stronger than che hypothesised effect of magnesium).

l e was proposed in discussions chat the relationship between magnesium intake and risk was non-linear. Therefore shifting che whole population slighcly up the intake curve could improve the magnes ium scams of a mi nority of individuals who have very low magnesiu m intake and very high cardiovascu lar disease risk. However, no numerical models were presented on the hypothetical risk curve, nor estimates of how large chis high risk fractio n of the population might be, or their degree of increased risk. During discussions, Professor Ragnar Rylander of Sweden suggested chat bicarbonate in water may reduce the urinary excretion of magnesium by making urine more alkaline, th us increasing che quantity of magnesiu m in the body, and chis hypothesis could be explored in ocher epidemiological studies.

Water Treatment Aspects A presentation on behalf of the Water Qual ity Association (representing manufacturers of water softeners and fil ters in North America) delivered the message chat che chemical compositions of namrally soft rap waters and artificially softened waters were quite different. The speaker reported chat since publication of the WHO Minerals in Water report some legal sui ts had been th reatened against suppliers of water softeners an d reverse osmosis fil ters in the USA, on the grounds that removal of these minerals from drinking water had caused householders to suffer heart attacks. Three speakers from che Netherlands presented papers on centralised softening of hard waters and remineralisarion of nanofilcered or reverse osmosis fi ltered water. Cose savi ngs due to softening in terms of lifespan of hoc water appliances, reduced energy use and reduced detergent use have been documented. Conversely, the m ineral con tent of nanofilcered or reverse osmosis fi ltered water needs co be increased by marble fi ltration before distribution in order co prevent pipe corros10n. Ir was noted chat scaling problems would impose an upper limit on acceptable calcium levels in drinking water suppli es. Magnesium sales are more soluble than calcium salts, and therefore there are unlikely co be major technical barriers to magnesiu m addition if it was decided char chis was desirable.

Conclusions • Some people believe that the evidence is sufficient co show a beneficial role of

magnesium in drinking water with respect co morcaliry from cardiovascular diseases. • Ochers believe that the quality of the evidence is coo poor to draw such a concl usion. • Given the appropriate resources, it would be scientifically possible to produce higher quality evidence. • Supplementation of drinking water supplies with magnesium co the levels proposed co be beneficial (perhaps 10-20 mg/L) is likely to be technically feasible. • Even if the beneficial effect is real, the degree of benefit achievable by water supplementation would depend on the dietary intake of magnesiu m from food, and on cap water consumption patterns. Throughout the Symposium, a number of people expressed the view chat drinking water should not be used as a mea ns of correcting a d ietary deficiency, and chat if an increase in calcium and/or magnesium intake in populations was deemed desirable, fo r public health reasons, it should be achieved through foo d. Following the Symposi um, WHO convened a cwo-day meeting of an expert group co draft a co nsensus statement on the

Health Effects ofMagnesium and Calcium in Drinking Water. The consensus statement is expected to be released later in 2006, and will be followed by a more derailed report incorporati ng material from the Symposium. A separate WH O expert group is preparing advice on techn ical aspects of desalination (e.g. corrosion control).

The Author

Martha Sinclair is a Senior Research Fellow in che Department of Epidemiology and Preventive Medicine, Monash University, and Editor of Health Stream, the public health newsletter of the CRC for Water Q uality and T reatment, (03) 9903 0571, email: marcha.sinclair@med.monash.edu.au

References Presenrations from the Baltimore Symposium are available on the websice: www.camgwacer.org/defaulc.hcrn WHO (2005) Nucriencs in D rinking Wacer, Geneva. ISBN 92 4 159398 9. This report is available from: www.who.inc/wacer_sanicacion_healch/ d wq/ nu trien csindw/en/index.h cm I Carling L, Abubaker I, et al. (2005) Review of evidence for relationship between incidence of cardiovascular disease and water hardness. University of Ease Anglia and Drinking Water lnspeccorace. This report is available from: www.dwi.gov.uk/regs/infolecc/ int200 5.shcm

II Monadelphous We're a leading structural, mechanical and process concraccor wi th a solid reputation for: •

Delivering quality work Providing reliable and flexible service Developing long-term customer relari onships

We're now bringing chis reputation ro you. This might sound simple bur very few companies can actually deliver ir. We can.

If you want to set new standards on your projects, consider Monadelphous.

Tel (08 ) 9316 1255

www. monadel. com. au

Journal of the Australian Water Association



fereed paper

NDMA - AN EMERGING ISSUE FOR THE WATER INDUSTRY B Nicholson Abstract N-Nitrosodimerhylamine (NDMA) is an emerging issue for rhe water industry due to its identifi cation as a carcinogen ic disinfection by-product formed during chloramination. Ir is also an issue in terms of wastewater reuse. This paper summarises rhe mechanisms ofNDMA formati on, analysis and removal (by UV). No World Health Organization (WHO) or Australian drinking water guidelines for NDMA have yet been set. However, it is being co nsidered by WHO and a guideline may be published later in 2006.

Introduction N-ni trosodimethylamine (NDMA) (Figure 1) has been known to be a carcinogen for many years. Animal experiments have demonstrated its potent carcinogenicity and it has been classified as a probable human carcinogen (Group 2A) by the International Agency for Research on Cancer (IARC, 1978) . Consequencly expos ure to chis compound should be reduced as far as possible. Recencly it has been shown that this compound can be for med during disinfection using chloramines and is therefo re potentially more wid espread in drinking water. It is also formed d u ring chlorination of created wastewater containing ammonia and consequencly is an issue for reuse. Thus NDMA can b e considered a disinfection by-product and is therefore an issue for the water industry.

Routes of Exposure N DMA exposure can come from a number of sou rces apart from drinking water. I t has been found in tobacco smoke and car ex haust. It is also found in a number of foods , especially those that use nitrite for curing, e.g. bacon. Other foods fo und to contain N DMA include fish products, powdered milk, cheese and beer (Lijinsky, 1999) . NDMA is a contaminant of a number of common pesticides (CICADS, 2 002) so exposure to these pesticides can result in NDMA exposure. NDMA has had a number of industrial uses, as well as being form ed in some industrial processes;



NOMA Figure l. Chemical structure of NOMA. ¡

NDMA is a carcinogenic DBP. The impact on the water industry will depend on the guideline value being developed by WHO. therefore occupatio nal exposures have occurred. For example NDMA is formed during the curing of rubber seals manufactured fo r the automotive industry (Reh and Fajen, 1996).

NDMA as a Water Contaminant NDMA in water can arise from discharges from industries employing NDMA in their manufacturing activities, or forming NDMA as a by-product of thei r activities (CICADS, 2002). The most significant in the USA appear to have resulted from the use ofNDMA as an intermediate in the production of 1, 1-dimethylhydrazine (unsymmetrical dimethylhydrazine, UDMH ), a rocket fuel. Groundwater contamination with N DMA, an impurity in the fuel and possibly formed by degradatio n ofUDMH, at an aerospace facility in California in 1998 resulted in wells bei ng taken out of service (CDHS,

2006). The stabili ty ofNDMA in the absence of light (Tate and Alexander, 1975) has resulted in contamination from ind ustrial sources being m ore significant for groundwaters. Earlier reports ofND MA contamination of drinking water (CICADS, 2002) appear to have been parcly from direct contamination, but, with current knowledge, disinfection processes would appear to also have been parrly responsible.

NDMA Formation During Disinfection Formatio n of NOMA from the reaction of d imethylam ine with a nitrosating agent such as nitrite is well known (Mills and Alexander, 1976) (Figure 2). The fungicide chiram can also form NDMA in the presence of nitrite (G raham et al. , 1996) . Until recencly, conventional wisdom suggested that this was the mechanism by which NDMA could be formed in water, and that by identi fying and controlling the sou rce of these reactants in water, rhe formation ofND MA could b e co ntrolled. However, studies conducted recencly have shown that it is not chat simple as another reaction pathway is generally more important (M itch et al., 20036) .

Ion exchange resins The formatio n ofNDMA from anion exchange resins in the presence of chlorine has been known for so me rime. The source of the amine precurso r was shown to be the resi n itself, not unreacted monomers in the resin, and chlorine was also involved (Kimoto et a!., 1980) . Cation exchange resins were also found to give rise to NDMA, even in rhe absence of chlorine (Najm and T russel, 2001). Loveland etal. (2001 ) found NDMA to be



di methylami ne




b" nitrite


Figure 2. NOMA formation from dimethylamine an d nitrite.

Journal of the Australian Water Association

refereed paper

present in the effluent of an operating ion exchange plant used to remove nitrate from grou ndwater at concentrations of 14 32 ng/L, again in the absence of chlorine. Formation from ion exchange resins was very dependent on the type of the resin itself, and was increased by the presence of ni tri te (Naj m and Trussell, 2001 ). Resins with 2 amino groups on the quaternary nitroge n were the only ones which formed NOMA so it appears that this is the source of the dimerhylamino funcrionali cy in the NOM A. Pretreatment with GAC was the only process foun d to effectively reduce N OMA levels (Loveland et a!. , 2001 ). GAC pretreatment may remove somethi ng that is im portant in NOMA fo rmation as Najm and T russel (200 I) fou nd groundwater in contact with an ion exchange resin fo rmed more N OMA compared with deionised water. Nitrosamine fo rmation from amines and nitrite is enhanced in the presence of fu lvic acids which are com monly fou nd in natu ral waters (Weerasooriya and Oissanayake, 1989). T his may explain increased format ion in narnral waters relative to deionised or GAC treated water but the mechanism is unknown , especially in the absence of nitrite. T he use of ion

~c\ I H3C


dimethyl amine





~c\ N-NHi I H3C


H3C \-N






Figure 3. NDMA formation fro m d imethyla mine a nd monochloramine with UDMH as the intermed iate.

exchange resins in the production of potable water is not common so formation in this process is not considered a major issue. Nitrogen-containing flocculants

NOMA can also be fo rmed in water treatment processes employing cationi c nitrogen containing flocculancs such as polydiallyldimethyl ammonium chloride (polyDAD MAC) when the water is chlorinated prior to flocculation (Child et al., 1996). Monochloramine also reacted with OAOMAC polymers to fo rm NOMA with much higher levels being fo rmed com pared with chlorine (Wilczak et al., 2003). Kohut and Andrews (2003)

demo nstrated that the use of cation ic nitrogen containing polymer flocculancs such as Epi- OMA (epichlorohydrindi methylamine) also led to the formation of NOMA on chlorination. Non-ionic polymers did not fo rm NOMA. (Wilczak et al. , 2003). Wilczak et al. (2003) also fo und that in an operating plane, recycl ing of filter backwash water led to a substa ntial increase in N OMA concentrations. It is not clear whether the recycling process leads to breakdown of the polymer to fo rm ND MA precursors or the additional polymer in the recycled backwash water leads to overdosing and subsequent carryover of polymer.

••• ••• Australian

Water Quality Centre

Journal of the Australian Water Association



technical features

Anaerobic degradation of cationic polymers has been shown to produce trimerhylamine (Chang et al., 2005) and trimerhylamine can be microbially converted to dimechylamine, a recognised NDMA precursor (Ayanaba and Alexander, 1974). The overall evidence indicates char formation of NDMA from amine-based polymer flocculanc and chlorine at doses used during water treatment is insignificant (Child et al., 1996; Najm and Trussell, 2001; Wilczak et al., 2003). However with monochloramine, NDMA is formed at much higher concentrations. NDMA formation can be minimised by reducing flocculant dose and avoiding the recirculation of filter backwash water. The use of ocher than cationic type flocculants if feasible will also minimise NDMA formation. Conventional chlorination/chloramination

Formation ofNDMA during water treatment in che absence of ion exchange resins or nitrogen containing flocculancs was first reported in che early I 990s Qobb et al., 1992). More recent work has confirmed chat dimechylamine reacts with monochloramine to form NDMA via a mechanism which involves UDMH as an intermediate (Choi and Valentine, 2002, Mitch and Sedlak, 2002a) (Figure 3). However, dimechylamine does not appear co be the sole, or even major, precursor ofNDMA in chloraminaced systems and other nitrogen-containing compounds muse be involved (Gerecke and Sedlak, 2003). NDMA formation is pH dependent and at pH 8.5 may be 2 - 5 times higher than chat at pH 6.5; NDMA formation is slow and is still increasing after 20 days when a chloramine residual is maintained (M itch et al., 2003a). Hence NDMA concentrations would be expected co

Specify and use Link-SealÂŽ seals to seal pipe penetration

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


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

Projex Group Pty Limited Telephone: 02 8336 1666, Facsimile: 02 8336 1670 www.projex.com.au, email : mail @projex.com.au



Journal of the Australian Water Association

increase with increasing detention rime in water distribution systems. Choi and Valentine (2003) found chat chlorine can produce NDMA from dimethylamine in che presence of nitrite. They postulated che formation of a highly reactive nirrosacing agent such as d inicrogen cecroxide formed from che nitrite by reaction with chlorine. However, it was suggested chat th is reaction mechanism is of minor importance in real waters due co che low levels of nitrite and dimechylamine. Low levels of NDMA can be formed from dimechylamine and nitrite, even in the absence of chlorine (Choi and Valentine, 2002) bur chis route would also appear co be insignificant at normal environmental concentranons. The evidence now suggests chat, in the absence of high levels of nitrite, negligible NDMA is formed from che reaction of dimechylamine and chlorine (Najm and Trussel, 2001; Mitch and Sedlak, 2002a; Choi and Valentine, 2002); monochloramine is required for NDMA formation. The data reported by Jobb et al. (1992) are interesting in chat they show relatively high (> I 00 ng/L) NDMA levels being formed in a chlorinated system. Whether this was due co elevated high nitrite levels in che raw water is unknown. An undocumented pathway may be via UDMH wh ich is an impurity in che pesticide daminozide chat also produces UDMH on hydrolysis. UDMH contamination from chis source would produce NDMA on oxidation with disinfectants such as chlorine. The fo rmation of NDMA in ozo naced waters is unknown; on ly limited testing has been carried out (Najm and T russell, 2001). However, NDMA can be formed from chlorine dioxide in che presence of ammonia and dimerhylamine but the mechanism is unknown (Andrzejewski et al., 2005).

Levels in Drinking Water Supplies Although NDMA is a probable carcinogen, little work appears co have been done in monitoring for chis compound, as it was considered an unli kely contaminant. This appears co have changed since the discovery of NDMA as a by-product of chloraminacion disinfection. In addition, developments in analytical technology have resulted in che necessary instrumentation bei ng available at affordable prices. Several papers have reported concentrations in US (e.g. , CDHS, 2002; Barrecc et al., 2003; Valen ci ne et al., 2005) and Canadian (Andrews and Taguchi, 2000; Charrois et al., 2004) drinking water supplies. Conventional treatment without cation ic polymer flocculancs such as polyDADMAC followed by chloraminacion leads, in most cases, to che formation of low levels of NDMA (<10 ng/L). However, occasionally higher levels are formed which suggests chat, at times, precursor levels can be unusually high. A concentration of 180 ng/L has been reported from one system but this appears an exception; more typically concentrations of 20 40 ng/L might be considered high. When cationic polymer flocculancs are used, these may contribute che major proportion of che NDMA formed. Levels up co 30 ng/L can be present in systems employing ion exchange during treatment.

Wastewater Chlorination Wascewacers which generally contain ammonia will produce monochloramine on chlorination for disinfection. Organic nitrogen levels are also normally relatively high; am ines such as dimechylamine, a known precursor of NDMA, are excreted in urine. Noc surprisingly ch lo rinated wascewacers can therefore contain NDMA with levels of between 20 and l 00 ng/L reported as typical (M itch and Sedlak, 20026) . NDMA in wascewacers can

UV for NDMA Treatment also originate from root control treatment in sewers and discharges from printed circuit board manufacturing industries (Sedlak et al., 2005). All of chese sources can impact on che use of wastewater for groundwater recharge.

Removal of NDMA NOMA is highly soluble in water, has a low vapour pressure and is poorly adsorbed to organic macerial and soils (Cl CADS, 2002). Consequently, it cannot be removed by air strippi ng or convencional flocculation, and is poorly removed by accivaced carbon (Gunnison et al., 2000). It is a relatively small molecule and is therefore also poorly removed by reverse osmosis (Soroushian et al., 2001). NOMA is rapidly degraded by light; consequently UY treatment has received some attention. UY treatment wich low- or highpressure UY lamps was eq ually effective (S harpless and Linden, 2003). Pulsed-UV and pulsed-UV/peroxide processes have been investigated and both were very effective (Liang et al., 2003). However one of che major products is dimethylamine which, as an NOMA precursor, can fo rm NOMA on subsequent chloraminacion. The process using peroxide was considered superio r as produces of the phocolysis which can subsequently reform NOMA were removed. Removal was dependent on the levels of NOM in che wacer being creaced. Higher NOM levels resul ted in lower removal due co compecicion for the UV lighc. The process employi ng peroxide did, however, fo rm some T H Ms (Lia ng et al., 2003). Sunl ighc has also been found co be very effeccive (Soroushian et al., 200 I). UV systems are already in place fo r NDMA removal. For example, UV is bei ng used in ac lease chree syscems in California for NOMA control (Yamamoco and Sakaji, 2003).

Analysis for NDMA Concentracions of concern for NDMA are in che low ng/L range, seve ral orders of magnitude lower chan chose of other DB Ps. T hus che analysis icself presents a challenge. To achieve the desired dececcion Iimies, excraccion and concencracion from large sa mple volumes, and detection using sophisticated instrumencation such as gas chromatography/mass speccromecry (GC/MS) are required. Ei cher solid phase (SPE) or liquid-liquid excraction can be employed with sample volumes of 500 mL or 1 L. Due co low extraction efficiencies, isocope dilution, i.e. spiking the sample with deuterated NOMA (N OMA-d 6), is required. SPE is the excraction mechod of choice as it is less labour inte nsive and does not require the use of large vo lumes of solvent. All mechods employ carbon as the adsorbenc macerial. High resolution mass speccromecry (HRMS) co mbined with gas chromacography was inicially used co enab le mass peaks from N OMA co be resolved from ocher possible contaminancs and a dececcion limic of around I ng/L was achievable (Taguchi et al., 1994). T he inscrumentation is very expensive whi ch appears co have lim ited investigations of NOMA occurrence. A number of methods have been reporced using chemical ionisacion (Cl) racher than electron impact ionisation. Th is improves specificity and sensicivity, leading co lower resolucion, and much cheaper instruments being suitable for analysis. Most mechods use methanol or aceco ni crile as che Cl gas. Charrois et al. (2004) also employed an isocope dilu tion mechod wich mass selective dececcion (MS D) and ammonia as che Cl gas; ammonia furcher im proves che specificity of che analysis. T he dececcion limit was I ng/L. Specificity can be furcher improved by using MS/MS dececcion with quantification based on daughter ions

Trojan UV The Proven Barrier to Contaminants of Concern Regulators and water providers are increasingly recognizing N-nitrosodimethylamine (NDMA) and other nitrosamines as credible health threats. Conventional water treatment processes a re ineffective for the removal of nitrosamines, as are membranes (including reverse osmosis) and ozone. Only one technology is proven to cost-effectively remove NDMA from water: UV-oxidation. Only one company has the experience in municipal, large-scale, high-efficiency NDMA treatment systems: Trojan Technologies. Us ing a combination of UV light and hydrogen peroxide, the TrojanUVPhox™delivers microbial disinfection and a proven way to address NDMA and othe r environmental contaminants including seasonal taste & odor causing compounds, algal toxins and pharmaceuticals. Fo r cost-effective drinking water and water reuse solutions that deliver reliable NDMA treatment and s imultaneous UV disinfection, the world is turn ing to Trojan. www.trojanuv.com Trojan Technologies is represented by:

AQUATEC-MAXCON PTY LTD A. B .N . 4 5 002 250 482


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

..:• ··:..


Journal of the Australia n Water Association




(Cheng et al., 2005; Munch and Bassett, 2006).

Limits in Water The US EPA regulates NOMA bur only under acts which control discharge to the envi ronment. The US Food and Drug Administration (FDA) regulates NOMA in malt beverages (beer) at 5 µg/L, barley malt at 10 µg/L and 10 µg/kg in rubber teats fo r baby bottles (USDHHS, 2002). There is no US limit for NOMA in drinking water NOMA was recognised in Canada as a potential problem from contamination of source waters in the 1980s. Ir was subsequently identified as a disinfection by-product in Ontario and a limit (interim maximum acceptable concentration) of 9 ng/L set (Andrews and Taguchi, 2000) In California a notification level for NOMA was adopted to protect human health following the detection of wells contaminated by NOMA. The value has changed over rhe past few years and now stands ar IO ng/L. A public health goal of 3 ng/L has also been proposed (CDHS, 2006). A gu ideline of 10 ng/L has also been adopted in Massachusetts (MORS, 2006). No World Health O rgan ization (WHO) or Australian drinking water guidelines for NOMA have yet been set. However, it is being considered for inclusion in the WHO guidelines and a guidel ine may be published lacer this year. The impact on che water industry will, no doubt, depend heavi ly on the guideline value.

Summary and Conclusions NOMA is an emerging issue for the water industry due to irs identification as a disinfection by-produce fo rmed during chloramination. It is also an issue in terms of wastewater reuse. !rs main mechanism of formation is through rhe reaction of nitrogen-co ntaining precursors such as dimechylamine and monoch loramine. An important source of precursors can be catio nic flocculant polymers but un identified naturally occurring precursors are also important. Less common sources of NOMA in drinking water are ion exchange resins and contamination from industrial discharges. The latter can be very important in wastewater treatment systems. Processes involving nitrite appear to be insignificant in water supply systems due to the natu rally low nitrite levels. UV irradiation is a treatment technique which is effective in removing NOMA 50 SEPTEMBER 2006


once formed but care needs to be taken that NOMA does nor reform in the distribution system. The impact on the water industry will depend on rhe guideline value being developed by WHO. Monitoring fo r chis compound, particularly in chloraminared water supply systems, will then be a priority.

The Author Dr Brenton Nicholson is Senior Organic Research Chemist with the CRC for Water Quality and T rearment and the Appl ied Chemistry Unit at che Australian Water Quality Centre, Salisbury, SA 5108, Brenton.N icholson@sawater.com.au

References Andrews, S. A. and Taguchi, V. Y. (2000). N OMA - Canadian issues. Proc. A WWA

Water Qua/. Technol. Conf, Salt Lake City, Utah, CD ROM. Andrzejewski, P., Kasprzyk-Hordern, 8. and Nawrocki, J. (2005). The hazard of Nnitrosodimechylamine (NOMA) formacion during wacer disinfection with strong oxidancs. Desalination, 176: 37-45. Ayanaba, A. and Alexander, M. (I 974). Transformation of mechylamines and formacion of a hazardous produce, dimechylnicrosamine in samples of created sewage and lake water. j. Environ. Qua/., 3: 83-89 . Barren, S., Hwang, C., Guo, Y., Andrews, S. A. and Valencine, R. (2003) . O ccurrence of NOMA in drinking wacer: A North American survey, 200 I - 2002. Proc.

A WWA Ann. Conf, Anaheim, California, CO ROM CDHS (2002) . Scudies on the occurrence of NOMA in drinking water. California Oeparcment of H eal ch Services. www.dhs.ca.gov/ps/ddwem/chemicals/ NOMA/sn,d ies. hrm. CDHS (2006). A briefhistoryofNDMA findings in drinking water supplies. California Deparnnenr of H ealth Services. www.dhs.ca.gov/ps/ddwem/chemicals/ ndma/hisrory.hrm. Chang, J .-S., Abu-Orf, M. and Dentel, S. K. (2005). Alkylamine odors from degradation of Aocculant polymers in sludges. Water Res., 39: 3369-3375. Charrois, J. W. A., Arend, M. W., Froese, K. L. and Hrudey, S. E. (2004) . Dececring Nnitrosamines in drinking water ac nanogram per licer levels using ammonia positive chemical ionization. Environ. Sci. Technol., 38: 4835-484 I. C he ng, R. C., Andrews-Tate, C. , Hwang, C. J., Guo, Y., Pastor, S. J. , Grebel, J.E. , Suffer, I. H. , Carr, S. and Young, C. (2005). Alternative Methods for the Analysis

ofNDMA a11d Other Nitrosami11es in Water and Wastewater, WareReuse Foundation, Alexandria, Virginia. C hild, P., Kaa, G ., Benitz, 0 ., Fowlie, P. and Hong-You , R. (I 996). Reaction between chlorine and dimethylamine containing

Journal of the Australian Water Association

polyelecrrolyre leading co che formation of N-nirrosodimerhylamine. Proc. AWWA Ann. Conf., Toronto, Canada, Water Res., Vol. C. C hoi, J. H. and Valencine, R. L. (2002). Formacion of N-nicrosodimechylamine (NOMA) from reaction of monochloramine: a new disinfeccion byproducc. Water Res., 36: 8 17-824. Choi, J . and Valencine, R. L. (2003). Nnirrosodimerhylamine formation by freechlorine-e nhanced nirrosarion of dimechylamine. Environ. Sci. Teclmol., 37: 4871 -4876 . C[CAOS (2002) . N -nitrosodimethylamine. Concise International Chemical Assessment Oocumenc 38. Prepared by R. G. Liceplo, M. E. Meek and W . Windle. www.inchem . org/documents/ cicads/ cicad38 .h rm. Gerecke, A. C. and Sedlak, D. L. (2003) . Precursors of N -nicrosodimethylamine in natural waters. Environ. Sci. Technol., 37: 133 1-1336. Graham, J. E., Meresz, 0., Farquhar, G . J. and Andrews, S. A. (1996). Thiram as an N O MA precursor in drinking water treacmem. Proc. A\Y/WA A nn. Conf., Toronto, Oncario, Water Res. : 15-27. Gunnison, 0., Zappi, M. E., Teeter, C., Penningron , J. C. and Bajpai, R. (2000). Artenuarion mechan isms of Nnitrosodimethylamine at an operacing intercepc and treac groundwacer remediation system. j. Hazard. Mater., 73: 179-1 97. lARC ( 1978), Monographs 011 the Evaluation of

Carcinogenic Risk ofChemicals to Humans. Vol. 17 Some N-nitroso Compounds. Lyon: Internacional Agency for Research on Cancer, pp. 125-175. Jobb, 0. 8., H unsinger, R. B., Meresz, 0. and Taguchi, V. Y. ( 1992). A study of rhe occurrence and inhibition of formacion of N-nirrosodimerhylamine (NOMA) in the Ohsweken wacer supply. Proc. A WWA

Water Qua/. Technol. Conf, Toronto, Ontario: I 03-132. Kimoto, W . I., Dooley, C. J ., Carre, J. and Fiddler, W . ( 1980). Role of strong ion exchange resi ns in nirrosamine formation in warer. Water Res., 14: 869-876. Kohuc, K. D. and Andrews, S. A. (2003) . Polyelecrrolyces as amine precursors for Nnirrosodimechylamine in drinking water.

Proc. A WU'i'A Ann. Conf, Anaheim, California, CD ROM . Liang, S., Min, J. H., Davis, M . K., Green, J. F. and Remer, D. S. (2003) . Use of pulsedUV processes co destroy N OMA. / Am. Water Works Assoc., 95(9): 121-131. Lijinsky, W . (I 999). N-Nirroso compounds in the diet. Mutat. Res., 443: 129-138. Loveland, ) . P., Means, E. G. and Rasouli, M. (2001) . Pilot resting co inhibit N nitrosodimethylamine (N OMA) format ion d uring anion exchange treatment. Proc.

A WWA Water Qua/. Tee/mo/. Conf, Nashville, Tennessee, CD ROM. Mills, A. L. and Alexander, M. (1976). Factors affecting dimechylnicrosamine formation in samples of soil and wacer. j. Environ. Qua/., 5: 437-440. M itch, W. A. and Sedlak, 0. L. (2002a) . Formation of N-n irrosodimethylam ine

technical features


refereed paper

health-related issues (N OMA) from dimethylamine duri ng chlorination. Environ. Sci. Technol., 36: 588-595. Mitch , W. A. and Sedlak, D. L. (20026). Faccors controlling nitrosamine format ion during wastewater chlorination. Water Sci. Technol.: Water Supply, 2(3): 191-198. Mitch, W. A., Gerecke, A. C. and Sedlak, D. L. (2003a}. A N-nirrosodimerhylamine (NOMA) precursor analysis for chlorination of water and wastewater. Water Res., 37: 3733-374 I. Mitch, W . A., Sharp, J . 0., Trussell, R.R., Valentine, R. L., Alvarez-Cohen, L. and Sedlak, D. L. (20036). Nnirrosodimethylam ine (NOMA) as a drinking water contaminant: a review. Environ. Eng. Sci, 20 : 389-404. MORS (2006). Standards and Guidelines for Contaminants in Massachusecrs D rinking Waters. Massachusetts Office of Research and Standards. www.mass.gov/dep/water/dwsrand.pdf. Munch, J. W. and Bassett, M . V. (2006). Method development for the analysis ofNnitrosodimethylamine and other Nnirrosamines in drinking water at low nanogram/liter concentrations using solidphase extraction and gas ch romarography wit h chemical ionizat ion tandem mass spectrometry. j. AOAC Int. , 89: 486-497.

l)e iMAII

Najm, I. and Trussell, R.R. (2001) . N O MA format ion in water and wastewater. j. Am. Water Works Assoc., 93(2) : 92-99 . Reh, B. D . and Fajen, J.M . ( l 996). Worker exposu res t0 nirrosamines in a rubber vehicle sealing plant. Am. Indwt. Hyg. Assoc.}., 57: 918-923. Sedlak, D. L., Deeb, R. A., Hawley, E. L., Mitch, W. A., Durbin, T. D ., Mowbray, S. and Carr, S. (2005). Sou rces and fate of nirrosodimethylamine and its precursors in municipal wastewater treatment plants. Water Environ. Res., 77: 32-39. Sharpless, C. M. and Linden, K. G. (2003) . Experimental and model comparisons of low- and medium-pressure H g lamps for rhe direct and H 2 0 2 assisted UV phorodegradarion of N -n itrosod imethylamine in simulated drinking water. Environ. Sci. Technol., 37: l 933- 1940. Soroushian, F., Shen, Y., Patel, M . and Wehner, M . (2001 ). Evaluation and pilot testing of advanced treatment processes for N O MA removal and reformation prevention. Proc. A WWA Ann. Conf,

Washington, DC Taguchi, V. Y. , Jenkins, S . W . D. , Wang, D . T., Palmenrier, J.-P. F. P. and Rei ner, E. J. ( 1994). Determination of Nnitrosodimerhylamine by isotope dilu tion, high-resolution mass spectrometry. Can. j. Appl. Spect1'osc., 39(3) : 87-93.


T are, R. L. and Alexander, M. (1975) . Stability of nitrosamines in samples of lake water, soil, and sewage. j. Nat. Cancer Res. !mt., 54: 327-330. USDHHS (2002). 10th Report on Carcinogens. US D epartment of H ealth and Human Services, Public Health Service, National Toxicology Program. www.ehp.niehs. nih.gov/roc/renrh/profiles/s l 28nitr. pdf. Valentine, R. L., Choi, J., Chen, Z ., Barrett, S. E., Hwang, C., G uo, Y., Wehner, M., Fitzsimmons, S., Andrews, S. A., Werker, A. G ., Brubacher, C. and Kohut, K. (2005),

Factors Affecting the Formation ofNDMA in Water and Ocrnrrence, AwwaRF, WERF, Denver, Colorado. W eerasooriya, S. V. R. and Dissanayake, C. B. (l 989). The enhanced formation of Nnitrosamines in fulvic-acid mediated environment. Toxicol. Environ. Chem., 25: 57-62. Wilczak, A., Assadi-Rad, A., Lai, H. H., Hoover, L. L., Sm ith, J. F., Berger, R., Rodigari, F., Beland, J. W., Lazzelle, L. J ., Kincannon, E. G ., Baker, H. and Heany, C. T. (2003). Formation of NOMA in chloraminared water coagu lated with D AMAC cation ic polymer. j. Am. Water ~Vorks Assoc., 95 (9) : 94-106. Yamamoto, G . and Sakaji, R. H. (2003). A California regulatory perspective on NOMA.

Proc. A WWA Ann. Conf, Anaheim, California, CD ROM.


Metallic & Non-Metallic Sedimention Tank Equipment • • • • •

Chain Sproc kets Carri er & return wear shoe s Wear strip Bearings

In 1995, REH AU developed the fi rst pipe diffuser with a silicone rubber membrane. Today, the RAUBIOXON pipe aerator has proven to be the next generation in fine bubble diffusers, with installations in hundreds of WW TPs across the globe, including Australia.

Engineering Plastic Components for water treatment • Screw co nveyor liners • Washpacter bear ings • Step screen componen t s • Gears • Machine guards • Ta nks

• Superior mechanical properties including anti-microbial formu lation • High oxygen transfer efficiency • Inventive design

• Lower operational costs • Less maintenance • Longer service life • Lower installation costs

To find out how you can save on operational and capital expenditure on yo ur new or existing aeration system project, contact: REHAU Pty Ltd Victoria New Sou th Wales

Queensland South Austra lia West ern Australia


Ph. (03) 9558 0666 Ph. (02) 9878 5544 Ph. (07) 3423 8997 Ph. (08) 8346 0011 Ph. (08) 9456 4700

dotvic@dotmar.com.au dotnsw@dotmar.com.au dotqld@dotmar.com.au dotsa@dotm ar.com.au dotwa@dotmar.com.au


91 Derby St, Silverwater, NSW 2128 Phone: 02-97481788 Fax: 02-97481878 Email: elliott.burke@rehau.com.au

Journal of the Australian Water Association




technical features


health-related issues





from the co mpound urea (NH2lzCO, which is the major source of nitrogen in both of these substances. This is because it hydrolyses in d ilute aqueous solution (assisted by bacteria and enzymes which frequen tly also are present), to for m ammonia and carbon dioxide. (NH 2)z CO + H 2 0 ~ CO 2 + 2NH 3 (1)

i In 2001 two requests were received co .,Q. 0.8 monitor the "chlorine-like C 'C atmospheres" associated with a spa 0.6 0 :E pool and a large municipal indoor u 0.4 0 swimming pool, to check for C 0.2 compliance with "the relevant ~ u regulations". The problem was then, ~ 10 4 2 6 8 as it remains today, char no pH regulations, standards or TLV values, ere, exist for the contaminants Figure 1. Effect of p H on ch lorine species in so lution. concerned, which are principally The Mechanism of the hypochlorous acid (with or without Chlorination Process overlying swimming pool atmospheres, and hypochlorire ion) and one or more of three the effect char these would have on both When gaseous chlorine is introduced co chloramines, of which trichloramine is of swimmers and pool employees Qessen water it rap idly undergoes hydrolysis to particular concern. 1986, Massin eta/ 1998, Thicken eta/ fo rm hypochlorous acid (HOCI) and T he deleterious physiological effects of 2002, Nemery et al 2002, Berna rd et al hydrochloric acid (HCI): these contam inants in pool atmospheres, on 2003). (2) both swimmers and pool employees, have Cl 2 + H 2 0 ~ HCI + HOCI The contaminants in swimming pools been receiving considerable attention from With a hydrolysis consta nt KH of about principally are introduced by the swi mmers European researchers in recent years. Th us 4x 104 at 25°C and a time constant themselves. Those not so introduced are it may be only a matter of rime before measured in tenths of a second, this adventitious organic and inorgan ic similar attention will be required in conversion is co mplete within about a materials transferred from the surroundings Australia and New Zealand. second or so. "'-,,< by prevailing winds (such as dust, soot, Th is paper describes the fo rmation of these The H OC! is present as rhe undissociated ¡ insects, vegetable debris, ere). Those atmospheric contaminants from the pool molecule at low pH val ues, but at higher produced by the swimmers are principally sterilisation materials, which may be pH values it dissociates into rhe textile fi bres from swimwear and towels, gaseous chlorine, sodium or calcium hypochlorite ion OCl -. At a pH of about cosmetics, skin flakes, nasal and oral hypochlorite, or chlorinated cyanuraces of 7.5 the proportions are approxi mately mucous discharges with associated bacteria varying formu lations introduced 50 years equal, as indicated in Figure I . The OCI and viruses, bacteria fro m skin lesions and ago . Also described is a simple monitoring ion also has disinfection properties, bur is sores, and particularly nitrogenous procedure, devised in 2001 in response co less effective than the undissociated compounds associated with perspi ration the above- mentioned req uests, which hypochlorous acid, largely because the and urine. For convenience these determines a "hypochlorous acid latter, as an und issociated molecule, can nitrogenous compounds are subdivided into equivalent" in ppm, and the reasons fo r penetrate bacterial membranes more easily two categories; namely "ammonia nitrogen" suggesting, as a first approximation and than the charged hypochlori re ion (Weiner and "organic nitrogen" . Thus pool subject to review, chat a TLV for chis 2000(a)) . contaminants are characterised as (i) "hypochlorous acid equivalent" could be particulate matter, which is removed by The results are the same if sodium or 0.3 ppm, and a ST EL 1.0 ppm. filters; (ii) ammonia nitrogen; (iii) organic calcium hypochlorire are used instead of nitrogen, and (iv) bacteria, viruses and Introduction gaseous chlorine. T hese react with water to ocher microbiological materials, many of give hypochlorous acid and sodium or The purpose of disinfection principally is co which also fall with in the "organic calcium hydroxide respectively. Under destroy various bacteria and viruses which nitrogen" category. normal operational conditions the amounts may cause health problems. T he of HCI produced by using gaseous ch lorine, nor on ly T he "ammonia nitrogen" comes disinfection procedures, utilising either or NaOH or Ca(OH)z by using the sodium from ammonia itself, which is a constituent gaseous chlorine or various chlorineor calcium hypochlorices, are coo small in a of both perspiration and urine, but also containing compounds (with or without swimming pool or spa pool environment co ocher additives) were understood and affect the pH in practical terms. However, employed before and during WWl(ICI The formation of cases have occurred where the use of 1933). H owever, it was not until the later gaseous chlori ne by careless or 1980s chat attention was directed to the atmospheric contaminants inexperienced pool operators has prod uced so much by-produce HCI char the pool pH by chlorination of T his paper was first published in Clean Air, has falle n to below 3; (i n one case to as low Journal ofCASANZ, May 2006. as 2. 7) causing severe and potentially indoor pools.






Journal of the Australian Water Association

disfiguring erosion of dental enamel in many swimmers (Centerwell et al 1986) . Deneal enamel decalcification can commence when the water pH is below 6.0. The recommended range for pool pH is 7.2 ro 8.0, achieved by using the buffering action of added sod ium carbo nate (Na2CO3) if necessary (Certerwell et al 1986). This problem does not arise with the use of sodium or calcium hypochlorires as sterilising agents.

and hydrochloric acid (Cli fford White J 972(c)) .




Chlorine Dose

H ypochl oro us aci d, however A Monochloramlne formation. produced, reacts with am monia ro B Increasing dichloramlne formation with dectease fo rm chl oramin es. As stared above, of monochloramlne. ammonia originates principally from IIT9ducible minimum residual: formation C of trichloramlne commences. rwo so urces in troduced by the 0 DI- and trichloramlne. swimmers themsel ves; namely perspiration and urine. Nor only E Free hypoclllorous acid. does a small amount of ammonia exist as such in each of these Figure 2. Effect of chlorine dose on product materials, bur substantial addi tional composition. quantities are provided by the hydrolysis of urea, which occurs at NH 2CONH 2 + 2HOCI ~ 2NH 2CI + about 0.01 - 0. 2% in perspiration and I ro H (6) 2O + CO2 2% or thereabouts in urine, accoun ting for N H CO N H + 4HOCI ~ 2N HCl + 2 2 2 80 - 90% of the nitroge n con tent in each 3H 20 + CO 2 (7) case (Cli fford White I 972(a)). Other + NH CONH 6HOCI ~ 2NCl + 5H nitrogen-containing co mpou nds in both 2 2 2O 3 + CO 2 (8) perspiration and urine are uri c acid, creatinine and ami no acids, but these are The chloramine formation reactions occur organic nitrogen (as opposed ro amm onia seq uentially with increasing applications of nitrogen) co mpounds, and react slowly with hypochlorous acid. The initial chlora mine hypochlorous acid to prod uce different to be form ed is monochloramine. Alth ough this is a disinfecting agent in its own right, products, as discussed more full y below. it is less effective than hypochlorous acid. T he chloramines for med may be one of lrs odour is described as nor generally monochloramine, dichloramine or offensive. W ith increasing additions of tri chl oromine (or ni trogen trichloride), hypochlorous acid the fo rmation of acco rding ro the following reactions: dich loramin e commences. Dichloramine is HO C! + NH 3 ~ NH 2CI a more effective disinfectant than (3) (monochloramine) + H 2O monochlo ram ine but it possesses an HOC!+ NH 2C I ~ NHCl 2 (d ichloram ine) offensive odour. As the additions of + H 20 (4) hypochlorous acid increase, rrichloramine is HO CI + NH Cl2 ~ NCl3 (rrichlorami ne) + produced. T his compound, unlike the two H20 (5) just mentioned, is less soluble in water, so it appears largely in the gaseous phase in the These reactions are dependent on pH, atmosphere above the pool surface. 1t has a temperacure, reaction time and initial ratio quire objectionable odour, and is a of ch lorine to ammonia (Cli fford White lachrymaror and severe eye irritant. With 1972(6)). However, any urea wh ich further increases in hypochlorous acid, the remains unhydrolysed accordi ng ro chloramines are largely (but not equation (1) above may be attacked by comp letely) destroyed to fo rm nitrogen gas HO CI to form chloramines.

The processes described above may be illustrated as indicated in Figure 2, where monochlorami ne is fo rmed durin g the upwards rise of the resi duals curve, A, and both monoand d ichlormine occur during the down wards d ip B. At the trough marked C trichloramine fo rm ation co mmences, and monoch loramine almost ceases ro exist. The rise of the curve thereafter is ca used by free hypochlorous acid with residual amounts of both di - an d rrichloramine. This graph is similar ro rhar shown by several authors (Clifford Whi te 1972(d), Seli nger 1998, Weiner 2000(6)). With increas in g proportions of orga ni c nitrogen, the downwards slope of the curve B rises more towards the horizo nta l before rhe eventual rise nam ed "chlorine res idual after demand satisfied", and ex hibits higher co ncentrati ons of di- and trichloram ine than occurs with amm on ia nitrogen alone. Theoretically, if only am mon ia nitrogen constitu tes the main co ntam in ant to be destroyed, addition of sufficient hypochlorous acid should remove all chlorami nes by converting them to nitrogen gas and hydrochloric acid, as mentioned above. Hence a sayi ng quoted by C lifford White (1972(e)}; "l f the air above th e pool water smells of ch lorine, there is nor enough chlorine in the water." However, because of organ ic nitrogen contaminants, and operating facto rs preventing th eo retical conditions from applying, di- and trichloramines, along with free hypochlorous acid, occur in varying amounts in accual pool operational cycles. Ir should be poi nted ou r that gaseous chlori ne and sodium and calcium hypochlori res are not the only chemicals ava ilable for chlorinating swimming and spa pools, although they are rhe most prom inen t. In the early 1950s, chlorinated cyanurares in various forms such as

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 find it all @www.glv.com

Journal of the Australian Water Association


rJ GL V i)~~



rrichloroisocyan urare (85% available chlorine), dichloromo nosodium isocyanurare (60% available chlorine) and dichloromo no potassium isocyanurare (59% available ch lorine) were marketed under rhe trade names ACL 85, 60 and 59 or COB 85,60 and 59, depen ding on rhe manufactu rer, specifically fo r the swimm ing pool marker. T hese, and ocher related chlorinated organic compounds also produce hypochlorous acid when introduced ro water, and are further discussed by Cliffor d W hite (1972(f)). Ir was mentioned above char hypochlorous acid also reacts with organic nitrogen compounds (as opposed ro ammonia nitrogen compounds which yield the chloramines), bur rhe produces are ocher chlorinated organics such as the rrihalomechanes, of which chloroform (crichloromechane) is the most prominent. Unlike the chloramines, these chlorinated organics do nor play a pare in the disinfection process and do nor contribute significancly ro pool odours (Aiking et al 1994, Nieuwenhu ijsen et al 2000, Chu et al 2002). The three references cited refer on ly ro chloroform, and make no reference to chloramines. T hese papers describe rhe presence of chloroform within che water itself, which is in co ncen trations much higher than in tap water, and suggest char rhe ingestion thereof by swimmers nor only by dermal contact and swallowing, bur also by inhalation of air at the air/water interface, are matters of concern . Chu et al (2002) point our char although inhalation of the evolved chloroform is a major inges tion ro ute, they nevertheless did nor undertake any atmospheric measurements of its concentration therei n. However, this subject is outside of the field of chis paper, which is concerned only wirh the hypochlorous acid and chloramines, for which no TLY's or related concentratio n figu res exist. On the ocher hand, chloroform has been allocated an atmospheric TLY concentration of 10 ppm for industrial exposures, by both the ACG IH an d WorkSafe Australia. W hether chis, or some related figu re, should be applied ro swimming poo l atmospheres, would be a matter for rhe appropriate regulato ry authorities ro invesrigare.

The Development of a Monitoring Procedure Although analytical methods exist for the dererminarion of residual chlorine in water, including commercially available colorimetric kits for home swimming pool owners, no analytical procedures have appeared with regard ro hypochlorous acid




and chloramines in swimming pool atmospheres. Of rhe five principal species of concern, namely hypochlorous acid, hypochlorire ion, and mono- di- and rrichloramine, all bur the last named exist principally in solution in the water so their atmospheric presence is enhanced by misc drop lets and spray particulates. However, all are quire volatile, so they partition readily into rhe air (Nemery et al 2002). As rrichloramine is largely water insoluble, ir appears in rhe overlying atmosphere as a gaseous phase, ar rares governed by water turbulence (Massin et al 1998). However, all are liable to be present, ro varying degrees, in inhaled atmosphere, and all have undesirable physiological properties. Nore char in an actual pool, with continual changes in patron usage and warer conditions, all parts of rhe curves of Figure 2, namely A through E, would be present at any rime. In order to devise a monitoring proced ure, it was noted char all are capable of releasing elemental iodine from an iodide sol ution (such as potassium iodide) in rhe proportion of one molecule of iodine (I 2) for each chlorine atom in the reactant species. Assuming acetic acid (HAc) is present to give a required lower pH: HOCI + 2Kl + HAc ~ 12 + KCI + KAc + H20 (9) NH 2CI + 2KI + 2HAc ~ I2 + KCI + Kac + N H 4Ac (1 0) NHCl 2 + 4Kl + 3HAc ~ 21 2 + 2KCl + 2Kac + NH4Ac (1 I) NCl3 + 6Kl + 4Ac NH 4Ac


312 + 3KCI + 3Kac + (12)

The abovemenrioned reactions have been known and associated with rhe determination of chlorine residuals in water for some rime. T herefore ir appeared feas ible to investigate their applicability in rhe determination of rhe substances concerned in rhe overlyi ng atmosphere. As earlier mentioned, these compounds could be present either in misc or spray droplets, or in the gaseous phase, in the atmosphere.

Monitoring an Indoor Pool Following laboratory experiments, rhe atmosphere overlying an indoor pool at an inland NSW city was monitored in late 200 I at rhe request of the NSW Work Cover Authority. Pool users had comp lained of contracting asthma type symptoms and ocher respiratory problems on using the indoor pool. The swimming centre contained two pools; an outdoor Olympic sized pool sterilised with gaseous chlorine, and the indoor pool - together

Journal of the Australian Water Association

with a separate codd ler's pool - which had been sterilised with sodium hypochlorice until several months previously, when che then current practice of using calcium hypochlorire commenced. With no guidelines, existing procedures or precedents ro go on, rhe method applied was as fo llows. Two impingers, each containi ng 20 mL of an aqueous solution of potassium iodide and starch, (and acetic acid to bring the pH ro about 3), were attached to personal sampling pumps calibrated to draw 2 Umin through each. One was positioned about 50 cm fro m the water on one side of rhe indoor pool, and the ocher on a row of sears about 2.5 m from the wate r on rhe ocher side of rhe pool. In spire of the previous resting of rhe system in the laboratory via an artificial atmosphere produced from a hypochloriceconraining household bleach, wherein the impinger sol utions turned blue within a few minutes, there was no knowledge of how long ir would cake fo r che same type of colour change to occur in a real swimming pool monitoring situation. However, within about 12 ro 15 minutes rhe liquid in each of rhe impingers had co mme nced ro turn blue, and after 30 minutes it was decided to terminate the monitoring as a sui table depth of colour had developed in each. The amount of liberated iod ine was determined colorimecrically, although an alternative procedure could have been by cicrarion with sodi um ch iosulphace. The amount of liberated iodine in each impinger rube turned our to be the same, thereby indicating the uniformity of rhe atmospheric composition caused by the overhead fo rced draft ventilation . From rhe liberated iodine figure obtained, the atmospheric concentratio n of HOC!, by equation ( l O) was calculaced, giving rhe result of 0.18 ppm "H OCI Equivalent", (at 25°C and I arm.)

Discussion This result of0.18 ppm of"HOCl Equivalent" was reported in 2001 via the relevant City Council ro rhe NSW Work Cover Aurhoriry for what ir was worrh, because, as previously scared, there were (and still are) no regulations, TLY values or precedents with wh ich chis figure could be compared. Ir is nor known what action, if any, was taken with regard to chis information. However, a search of more recent literature revealed a paper (Bernard et al 2003) which reported on the effects of the atmosphere of chlori nated swimming pools on many

technical features

school children and adults. These auth ors cook numerous measurements of the rrichloramine content of rhe atmosphere 1.5 m above rhe pool surface, by using a method attributed co H ery et al (1995), and further described by Massin et al (1998). The method involved drawing a known volume of air at about 1 Umin through a glass fibre filter impregnated with sodium carbonate and arsenic trioxide, As 20 3. T he filte r was desorbed with distilled water, passed through an ion exchange resin co remove carbonate, and the chloride content determined by io n chromatography from which chloramine content was calculated. M assin et al (1998) utilised this method in rhe taking of 1262 samples of the atmosphere above 63 pools, and obtained the total chloramine concentrations. H owever, they point out that Hery et al (1995) state that rrichloramine occupies about 90 % of these, and mono- and dichloramine accou nt for rhe remainder. Therefore, they reported 90% of rhe coral chloramines as trichloamine, and state rhar of the 1262 samples taken, rhe mean of 860 measurements from rhe 46 public pools was 0 .24 mg/m 3 (S .D. 0 .1 7); and of the 402 measurements from rhe 17 leisure centre pools the mean NC1 3 co ncentration was 0.6 7 mg/ m 3 (S.D. 0.37). With regard co these figures, it fo llows from sratisrical theory rhar 2.28% of rhe 402 leisure centre pool res ults, (i.e. about 20), could have values of 0.67 + 2xS.D. mg/m3, and 0.127% would be 0.67 + 3xS.D., whereas at rhe lower end a number would be zero. T hus, rhe range of rhe 1262 results would be expected co range from zero to about 1.5 mg/m3. More recently, Thickett et al (2002) modified this method by drawing the sample air through a tube containing silica gel coated with sulphamic acid to trap hypochlorite and mono- and dich loramine, prior co rhe fi lter impregnated with sodium carbon ate and arsenic trioxide. Therefore they measured NCl3 directly in che study of 15 samples from 3 pools, with the samples taken about l m above the water surface. They state that the sampling effi ciency for NC13 exceeds 95%, and report results ranging from 0.1 to 0.57 mg/m 3. The figures quo ted above in mg/m 3 of NC1 3 convert to the follow ing ppm values (at 25°C and 1 arm) :


ppm@ 25°C and 1 atm



0 .24

0 .05

0. 57



0 .14



0 .31

Aiking, H .,van Acker, M.B., Scholten, R.J ., Fenscra, J.F., Valkenburg, H.A. (1994). Swimming pool chlorinatio n: a health hazard? Toxicology Letters 72: 375 - 380 Bernard, A., Carbonelle,S ., Michel, 0., Higuer, S., de Burbure, C., Bucher, J-P., H ermans, C., Dumont, X., Doyle, I. (2003). Lung hyperpermabiliry and asthma prevalence in schoolchildren : unexpected associations with rhe attendance at indoor chlorinated swimming pools. Occupational and Environmental Medicine 60: 385 - 394 Centerwell, B.S., Armstrong, C.W., Funkhouser, LS., Elzay, R.P. , (1986) . Erosion of dental enamel among compet itive swimmers at a gas-chlorinated swimming pool. American j ournal ofEpidemiology 123: 641 - 647 C hu, H ., Nieuwenhuijsen, M.J. (2002). Distribution and determinants of rrihalomerhane concentrations in indoor swimming pools. Occupational and Environmental Medicine 59: 243 - 247 Clifford White, G. ( 1972). H andbook of chlorination. Van Nostrand Reinhold, (a) 478 - 479; (b) 193; (c) 472 - 474 ; (d) 472 and 477; (e) 47 1; (f) 499 - 505 H ery, M., Hecht, G., Gerber, J.M., Gend re, J.C., Hubert, G ., Rebuffaud, J. (1995). Exposure to chloram ines in the atmosphere of indoor swimming pools. Annals of Ocettpational Hygiene 39: 427 - 439 IC!, ( 1933). Water sterilisation. Imperial C hemical Industries Limired. 50 pp. (No publication dare: it has been aJlocated 1933 because the latest of 53 references cited is 1932, and rhere are numerous references for rhat and rhe immediately preceeding years) . Jessen, H.J. ( 1986). Chloramine concentration in rhe air of indoor pools. Zeitschrift fi1r DGesamte Hygiene und ihre Grenzgebiete 32: 180 - 181 Massin, N ., Bohadana, A.B., Wild, P., H ery, M., Tomain, J .P. , Hubert, G . ( 1998) . Respiratory symptoms and bronchial responsiveness in lifeguards exposed to nitrogen trichloride in indoor swimming pools. Occupational and Environmental Medicine 55: 258 - 263 Nemery, B., Hoer, P.H.M. , Nowak, D. (2002). Indoor swim ming pools, water chlorination and respiratory health. European Respiration Journal 19: 790 - 793 Nieuwenhuijsen, M.J. , Toledano, M.B., Eaton, N.E., Fawell, J., Elliott, P. (2000). C hlorination disinfection byproducts in water and their association with adverse reproductive outcomes: a review.

If it should have so happened that t he co ntaminant measured at the NSW indoor pool in 2 001 , and reported as "0.18 ppm of HOC I Equivalent", consisted entirely of NCl 3, then its concentration would have been 0.06 ppm, or a third of the reported figure, because of the three Cl atoms in the NCl 3 molecule releasing three molecules of I 2 as indicated by equation (13). This fa lls within the range of figu res quoted above. H owever, as the contaminants at the NSW indoor pool consisted of an unknown mixture of hypochlorous acid, hypochlorous ion and mono- di- and trichloramines, it was considered that a more realistic category to report would be "hypochlorous acid equivalent", as was done at the time. With regard to possible exposure limits which could be applied co figures of "ppm of H OCI Equivalent", a suggestion may be derived fro m the paper by M assim et al (1998). These authors, in discussing their findings with regard to the 1262 samples analysed for NCl3, proposed a "long term limit value" (presumably equivalent co the ACGIH I WorkSafe Aust. "TWA") of 0.5 mg/ m3, and a "short term limit value" (presu mably equivalent to che ACGIH / WorkSafe Aust. "STEL") of 1.5 mg/m3, for NC13• T hese become 0.1 ppm and 0.3 ppm respectively, at 25°C and 1 arm. When expressed in terms of "HOCI Equivalent" they become three times chis value, or 0.3 ppm and 0.9 (say 1.0) ppm. Th erefore, as a proposal, and pending further investigatio n by competent committees, ere., if this is ever deemed to be of sufficient interest, the "HOC! Equivalent" of a swimming pool o r spa pool atmosphere could be allocated a TL V TWA of 0.3 ppm, and a STEL of 1.0 ppm. T hese are the same as for phosphine in the ACGIH /WorkSafe A ust. Tables. From the d eleterious effects experienced by nor only schoolchildren bur also adult pool employees, as reported in the references cited but not reviewed in this presentation (because of their medical nature), it is obvious that attention will have to be given to chis matter in the near future.

The Author Dr Ken Basden retired some years ago from the School of C hemical Engineering

& Industrial Chemistry, UNSW. He has since developed a consulting practice, mainly in the field of microscopy and microchemistry, with fore nsic applications.

Occupational and Environmental Medicine 57: 73 - 85 Selinger, B. (1998). C hemisrry in the marker place, 5th Ed., H arcourt Brace, 206 T hicken, K. M ., McCoach, J .S., Gerber, J .M., Sadhra, S., Bu rge, P .S. (2002). Occupational asthma caused by chloramines in indoor swimming-pool air. European Respiration Journal 19: 827 - 832 W ei ner, E.R. (2000). Applications of environmental chemistry. Lewis Publishers. (a) 172; (b) 162

Journal of the Australian Water Association



!fereed paper

Abstract In December 2004, Melbourne Water co mmissioned and obtained Class A accreditation from the V ictorian EPA and Department of Human Services (DHS) for a 60 million litres per day (MLD) state-ofthe-art chlorine and lN disinfection plane at Western Treatment Plant, Werribee, Victoria. From January 2005, C lass A recycled water has been su pplied to the nearby Werribee Irrigation D istrict (WID) for the production o f vegetables, mainly lettuces , caul ifl owers and broccoli, which can be cooked or eaten raw.

Keywords: lN d isinfection cred its, MS2 bacteriophage, lN validation, Class A recycled water, Cryptosporidiu m and G iardia inactivation.

Introduction Since December 2005, up to 55 megalitres per day of recycled water per day (up to 8,500 megalitres a year) of recycled water has been delivered from Melbourne Water's Western Treatment Plant at Werribee to rhe water retailer, Southern Rural Water, for distribution to to more than 100 vegetable growers in the Werribee I rrigation District, increasing their security of supply and contributing significantly to the V ictorian Government's Our Water Our Future action p lan. Since the water is used fo r prod uction of vegetables, mainly lettuces, cauliflowers and broccoli, which

Validated using the latest MS2 bioassay techniques 56 SEPTEMBE R 2006


Ultraviolet (UV) light disinfection technology is increasingly used across the world to disinfect drinking water and recycled water. Although the application ofthe germicidal properties of UV light was pioneered in the early 1900s, it was not until the 1990s that major technological advances in equipment design, increased understanding of the process fandamentals, development ofreliable validation and performance monitoring methods, as well as the growing need to address the often conflicting call for improved microbial water quality without forming additional disinfection byproducts (DBPs) led to the rapid adoption of UV disinfection for recycled water and drinking water that continues to this day. Pathogens such as Cryptosporidium are highly resistant to commonly used chlorine-based disinfectants, but UV light has been proven to be one ofthe most effective means for inactivating these microbes. Amazingly, scientific evidence proving the effectiveness of UV light against Cryptosporidium was not folly established until the late-l 990s. Doses of UV light as low as 3 m]lcm2 are effective in achieving more than 3-log reductions for Cryptosporidium in drinking water. The application of UV disinfection in the USA has been largely driven by the development of the United States Environmental Protection Agency's (USEPA) Long Term 2 Enhanced Surface Water Treatment Rule (LT2ESWTR) to further control microbial contamination of drinking water, in particular Cryptosporidium and Giardia without forming additional DBPs. The design, operation and maintenance needs ofultraviolet disinfection systems differ greatly from those for traditional chemical disinfectants such as chlorine. To provide a framework for the application of UV disinfection technology, the USEPA released Oune 2003) a draft Ultraviolet Disinfection Guidance Manual (UVDGM). This draft UVGM was adopted by Melbourne Water and Department ofHuman Services to apply UV disinfection for Class A recycled water production for the WID recycled water pmject. can b e cooked or eaten raw, it must be produced by treatment processes capable of providing a 7-lo g reduction in vi ruses and 6-log reduction in protozoan parasites and is ter med, in V ictoria, C lass A water. Table 1 shows the volumes of Class A recycled su pplied to WID from 200 5/06. By 2008/09, C lass A recycled water will also be supplied to the metropolitan water retailer C ity West Water for use in residen tial dual pipe systems for toilet flu shing, garden and open space wateri ng, commercial and industry use, thereby

Journal of the Australian Water Association

saving drinking water for future generations. Usage of recycled water in the nearby Werrib ee Tourist Precinct, a significan t tourist destination, will also be increased with the aim of fu rther reducing stress on the Werribee River. Figure I shows the exten t of planned recycled water projects in the Werribee region up to 20 08/09. As Victoria's first Class A recycled water scheme, the Werribee Irrigatio n District (WID) recycled water scheme has demonstrated t he application of recycled

refereed paper

Table 1. Western Treatment Plant, Class A Recycled Water Production Volumes (millions of litres). Year

2004/ 05 2005/06







183 568

155 33

357 1,354


water as a reliable and sustainable water resource for food production. The scheme has brought an immediate and substantial contribution to rhe Victorian Government's water recycling target of 20 per cent by 20 1O; it will also reduce the input of nitrogen to Port Phillip Bay by some 70 to nnes a year and will reduce stress on rhe Werribee River and underground aquifers.

Timetable H istorically, growers in the Werribee Irrigation District extracted irrigation water from the Werribee River, supplemented by groundwater. Because of on going drought, growers in the District received only 40 percent of their river water entitlement in 2003/04. Ac the same rime, declin ing gro undwater levels res ulted in a ba n on the use of groundwater to hale rhe risk of seawater intruding into the aqui fer. In January 2004, the Deputy Premier and Minister for Water, John Thwaites, brought forwa rd Melbourne Water's plans for the Werribee Irrigation District Recycled Water Scheme, calling for its completion in rime for the 2005 irrigarion season. T he works included rhe construction of a disinfection plant and five kilometres of pipelin e, rwo months of performance resting, endorsement by the Department of Human Services and final sign-off of rhe Class A recycled water scheme by Victoria's En viron ment Protectio n Authority.

Water Quality Issue Recycled water from WTP is produced by an advanced (anaerobic/facul racive) lagoon treatment process char incorporates an activated sludge treatment system and extended matu ration ponds. T he recycled water quality from the huge 55E Pond l O is high ly consistent and closely approaches Victorian EPA Class A gu idelines without additional treatment. The entire trearmenr process fro m raw sewage to final producr adopts the multiple barrier approach and H azard Analysis Critical Control Point (HACCP) principles. Up until June 2004, Melbourne Water was co nfident of being able to achieve Class A water quality with the addition of ch lorine disinfection. However, ongo ing water quality monitoring found low levels of Cryptosporidium when the sensitivity of the resting was increased from oocysrs/ 10 lie re


to oocysts/50 litre, and maximum co ncentrations were used instead of medians.

Cryptosporidium test results for water samples collected at the outlet of 55£ Lagoon system's Pond 10 from 17 March 2004 to 16 June 2004 were analysed and fo und char the maximum observed concentration of Cryptosporidium was 68 oocysrs per 50 litres with a median of 2 oocysts per 50 litres. Our of ch is sec of Cryptosporidium data, 27% (4 our of 15) were below the detection limit of 1 oocysts per 50 lirres. Based on the analysis of Cryptosporidium data, a minimum treatment target of 2-log (99% reduction) Cryptosporidium inactivation/removal, in addition to rhe chlorine disinfection, would be necessa ry to treat the observed peak concentrations of Cryptosporidium. The Department of H uman Services (OHS) reviewed the same Cryptosporidium data sec and co ncurred with Melbourne Water's assessment char an additional treatment seep capable of achievi ng a fu rther 2- log reduction in Cryptosporidium was required in order to achieve Class A recycled water criteria.

Assessment of Available Treatment Options Melbourne Wate r undertook a preliminary investigation of treatment processes capable of achieving a minimum of 2-log

Cryptosporidium inactivation/ removal. The four treatment options co nsidered were as follows: • UV disinfection; • Membrane filtration (micro fi Iera rio n/ ul rrafi Iera tio n); • Granular media fi lrrarion - direct filtration; • Granular media fil tration - slow sand fil tratio n. The preliminary investigation co ncluded chat all four options were capable of delivering at least 2-log of Cryptosporidium inacrivarion/removal. H owever, all of the fil cracion options required at lease 12 co 15 months to full y implement, wh ich was well beyo nd the Stage l target completion rime of end of Occober 2004, and therefore unacceptable. On the ocher hand, UV disinfection had the shortest delivery rime of th ree to six months with a most likely implementation ri me of three co four months, putting this option with in or near rhe end of October 2004 Stage l completion dare. UV disinfection was selected as the preferred treatment option co com plemen t the chlorine disinfection process and prod uce recycled water char meets Viccorian EPA Class A criteria. T his decision by Melbourne Water co use UV dis in fect ion proved to be the most

Werribee Recycled Water Project

___,............ ...... -


- • • hoptw,11...,.WW.,~~ -


~Mf-~Ml~,i,tod - T~lltN.llt

Figure 1. WTP Disinfection Plant locality plan. Journal of the Australian Water Association


SE PTEMBER 2006 57

straightforward of all the treatment options to design and implement, had the lowest capital cost, occupied the smallest footprin t, and remained withi n the delivery schedule, ensuring completion within the Government's eight ti meframe. Figure 2 is a diagram of the treatment train.

Detention Slora(o





CryptospoMdium parvum



MS-2 Coliphage

Rotvirus SA11 Pollovlrus (TypHVe1A)

Adenovlrus (Type 40)



Vibrio cholera• Salmonella typhl




. .

Water's Western Treatment Plant.

using MS2 as the surrogate organism, according to the validation protocols set our in UVDGM. In situ seeded challenge resting is presenrly considered the most rational and widely accepted method of validating the performance of UV disinfection reactors. For the project, the UV disinfection facility was required to achieve at least 2-log Cryptosporidium inactivation up to an ultimate design flow of 90 megalicres per day (MLD) with an initial design flow of 60 MLD for the expected range of 55£ Pond 10 water quality, in particular, the following critical water quality indicators: • UV transmittance (UVT); • Total suspended so lids (T SS); • T urbidicy; and • Calcium, iron, alkal inity, pH and hardness. Detailed analysis of the available historical water quality monitoring records for 55£ Pond 10 was undertaken to define the major design parameters for the UV disinfection facility. The historical water quality from 55£ Pond 10 was found to be


I Bactorlophago I







~=================================================::J c:::::=J Bacteri a

C=:J +--------~---~- ---------10





UV Ooso (m J/em2) for 2-Log (99'!.) Inactivation

Figure 3. UV sensitivity of various microbes (adapted from USEPA UVDGM). 58 SEPTEMBER 2006



Figure 2. The Class A recycled w ater process treatment train used at Melbourne



Chlorlnatlon Plant

Pvmptnc Station





I If


Melbourne Water presented the Department of Human Services (OHS) with the case for complementing the chlorination disinfection process with a UV disinfection facility. OHS responded positively to Melbourne Water's proposal for a UV disinfection facility and stated the following: • OHS agreed with Melbourne Water's assessmen t chat an additional treatment step capable of a fu rther 2-log reductio n in Cryptosporidium was needed; • OHS required in situ validation of the fu ll-scale UV facility via a seeded challenge study to demonstrate a minimum 2-log reduction in Cryptosporidium; • T he design of the UV validation programme must be based on the protocols outlined in the USEPA's Ultraviolet Disinfection Guidance Manual Qune 2003); and • MS2 bacteriophage (a virus of E. coli) was a suitable challenge organism as it is easily grown and co mmonly used for seeded challenge studies. Figure 3 compares the relative UV sensitivity of viruses to protozoans based on a 2-log reduction of these microbes. Thus, the primary treatment target for the UV disinfection facili ty was sec at 2-log Cryptosporidium inactivation as demonstrated by seeded challenge resting


~•" "--r--; ~,;===,=~

I UV Plant I If Pump&,: Station

UV Design Criteria and Validation

Giardia 1amblia

'.otmff'Ul,tblNClttin b'llloast.3'10.n

Journal of the Australian Water Association


remarkably consistent, and notably low in suspended solids and turbidity. Analysis of the historical data for the critical water quality indicators determined the following: • A minimum design UVT of 40% is reasonable and allows sufficient margin for unforeseen variability in the water quality; • TSS is expected to be below 20 mg/L; • Turbidity is expected to be below 10 NTU for more than 95% of the time; • The critical water quality parameters of UVT, T SS and turbidity are comfortably within the known upper limits for UV disinfection performa nce; and • T he correlation between TSS and turbidi ty is sufficient to use online turbidity monitoring as a critical control point for UV reactor control. In addition to the analysis of historical analytical data, the Australian Water Quality Centre in Adelaide performed two sets of collimated beam testing over a wide range of UVT, turbidity and T SS, on water samples collected from the 55£ and 25W lagoon systems. The collimated beam tests used standard laboratory-scale UV testing equ ipment and were conducted according to established USEPA protocols using MS2 bacteriophage. The observed MS2 inactivation rates followed expected UV dose-response curves and there was no evidence of UV shielding by particles/solids. T he collimated beam testing and literature reviews fo und chat UV disinfection could effectively function over the expected range of55E Pond 10 effluent quality. Collimated beam rests showed a typical UV dose-response fo r MS2 inactivation. A review of the literature also fou nd chat the UV dose-response remains typical for turbidity up to 10 NTU (or about 20 mg/L TSS) and algal counts up to 70,000 cells/ ml. T he 55£ Pond l O effluent has been consiscen rly within these values. On chis basis, a validation test using MS2 as a challenge organ ism was considered

technical features


li7.il ag

health-related issues appropriate for a UV disinfection facility serving the WID Recycled Water Scheme. In addition to the literature review conducted by Melbourne Water, a vendor guarantee was obtained based on the results of a MS2 validation test for the expected range of water quality from 55E Pond 10. The UV disinfection facility also needed to be fully integrated into the functional requirements for the chlorine d isinfection facility. The monthly availability target for the chlorine d isinfection facility was 95%. Investigation found that the UV disinfection facil ity would not impact on the 95% monthly availability of the ch lorination facility.

UV Reactor Control Strategies Control limits were determined and implemented according to HACCP risk management principles to guarantee the production of Class A recycled water and ensure the protection of public health. Proposed critical control limits fo r the following operating pa rameters are given in Table 2. Online turbidity m easurements were required to provide continuous mon itoring of the solids (TSS) e ntering the UV disinfection facility. Continuous


. • •

• •••

Table 2. Proposed UV operating control limits. Operating Parameter


Control Limit



Flowmeter (online)

Within validated flow range

UV dose

UV Intensity

UV sensor (online)

Within validated UV intensity range

UV dose, lamp status


UVT sensor (online)

Within validated UV intensity range and not less than 40%

UV dose, lamp fouling/ag ing, chemical spill


Turbidity meter (online)

Less than 10 NTU HO mg/L TSS) at all times. Lower control limit of 6-7 NTU operator surveillance

Elevated risk of off spec recycled water

turbidity monitoring is necessary to ensure that the UV reactor operates below the upper limits for TSS given by the UV equipment vendor, results of the collimated beam testing and findings of the literature review. UV transmittance (UVT) and turbid ity/TSS are the main critical control parameters for the UV disinfection process. Both of these were contin uously monitored using online instrumentation. The design of the UV facility was based on a m inimum UVT of 40% and a TSS of less than 20 mg/L (r l 0 NTU turbidity). These values were confirmed during MS2 validation testing. The UVT and turbidity were

adopted as control variables, and used to trigger alarms whenever these values are approached a nd potentially shutdown the plant if they are exceeded. The design of the UV disin fect ion system also included consideration of operational and redu ndancy factors to e nsure adequate robustness and reliability of the UV disinfection process. The validation testing was also used to define the UV reactor control strategy to ensure that the UV dose required for 2-log Cryptosporidium inactivation target was delivered and verifiable during the production of Class A recycled water. Three possible UV reactor control strategies were considered. T he

technical features

health-related issues

m . - . -.... -

• The UV d isinfectio n facility is expected to have a monthly availab ility of no less than 9S%; and

three control strategies differ in their degree of sophistication and complexity. H owever, all three control methods are equally robust, reliab le and capable of delivering the same level of recycled water quality. The simplest control method is the single set-point operation and calculated dose is the most sophisticated control method.

• Nl critical UV operating parameters can be measured continuously using online instrumentation, and therefore fit within a HACCP risk management framework. T he recycled water scheme is notable in that:

The single set-point con trol m ethod requires the least amount of commissioning and validation time and effort. Calculated dose is the most sophisticated control method and provides op timum energy use and lamp life. T o provide maximum control flex ibili ty and allowance for future optimisation of the UV disinfectio n faci lity, Melbourne Water adopted the calculated dose m ethod.

• A state-of-the-arr chlorination plant was design ed and validated to achieve a 2-log virus reduction credit;


• The chlorine plane was designed to operate in two chlorination modes depending on lagoon performance; and

Melbourne Water found that it is practical to design and operate a full- scale UV reactor with a m inimum treatment target of 2-log Cryptosporidium inactivation chat is reliable, robust and veri fiable on unfiltered source water from SSE Pond 10. This conclusion was based on the following findings:

• Full-scale testing of the UV treatment barrier using the latest MS2 bioassay techniques was completed to achieve a 2-log reduction credit for Cryptosporidium. The authors are unaware of any ocher UV disinfection facility in Australia that has been validated using UVDGM protocols.

• MS2 inactivation is unaffected in unfiltered water with turbidity up to 10 NTU (about 20 mg/L T SS);

T he testing, specification , purchase and inscallacion of the UV system was all accomplished in a eight timeframe of six months.

• There is no evidence suggesting that the algal counts in SSE Pond 10 approach a concentration that will impact on MS2 inactivation;

Add itionally, the low energy usage of the entire disinfection facility contributes to the Western Treatment Plant's commitment of greenhouse neutral operation.

• The demonstration of 2-log inactivation credit for Cryptosporidium was ach ieved using MS2 as challenge organism in collimated beam tests and full-sca le experiments;

The combined use of rwo disinfection barriers, i.e. UV and chlorine co ntact ti me, is consistent with the co ncept of multiple barriers to microbial contaminants. A UV and chlorine disinfection system combined with a HACCP risk management app roach provides recycled water schemes with a broad spectrum, robust and reliable microbial barrier to ensure p rotection of public health.

• An enh anced lagoon treatment system fo llowed by a chlorine and UV treatment barrier has been accredited with Class A recycled water status;

• C ritical UV design parameters of UVT, TSS and turbidity are within the known critical design limits fo r UV d isin fection;

The Authors John Poon is a sen ior member of Water Recycling, Strategy and Planning Group, Melbourne Water Corporation, john.poo n@melbournewater. com.au . Dr Alexandra Keegan and Dr Paul Monis conducted the MS2 seeded challenge studies, Australian Water Quality Centre & Cooperative Research Centre for Water Quality and Treatment. Dr Daniel Deere and Dr Annette Davison are the Principals of Water Futures which was engaged by Melbourne Water to provide independent oversight and documentation of the UV validation work. Dr Deere is also a member of the Cooperative Research Centre for Water Quality and T reatmen t.


For over 80 years we have provided critical water infrastructure and mangement solutions for our clients and communities.

North America and USEPA Guidance Manual far UV Disinfection,

Our dedicated team of professionals has special ist skills in engineering, environmental, risk, innovation and water cyc le management, and can provide integrated solutions for your business. To find out more please contact Selwyn McFaul or George Khouri in Brisbane on (07) 3244 9644 or Chris Morris in Sydney on (02) 8923 1555.

Success ,,_,, Partnership, Success ..ro.,, People



USEPA (2003), Ultraviolet Disinfection Guidance Manual (draft), EPA 8 l 5-D-03-007, http://www.epa.gov/ogwdw/lt2/guides.html Wh itby, G.E., Scheible, O.K. (2004) , The Histo1y of UV and Wastewate1; !UVA NEWS, Vol. 6, No. 3 Bolton, J.R. (2004), Fundamentals of Ultraviolet Treatment - Terms and Definitions do Matter, IUVA NEWS, Vol. 6, No. 3 Clancy, J.L., Linden, K.G., McCuin, R.M. (2004), C1yptosporidium Occurrence in Wastewaters and Control Using UV Disinfection, IUVA NEWS, Vol. 6, No. 3 Husley, R.A. , Mackey, H.E., N eemann, J.J. (2004), Application of UV in !UVA NEWS, Vol. 6, No. 4 Water Futures, Deere, D., Davison, A. (2004), Werribee Irrigation District


Journal of the Australian Water Association

Recycled Water Scheme UV Disinfection System, Independent Validation Report, Western Treatment Plant Lagoon System 55£, Final Report, I 7 December 2004 (unpublished) LeChevallier, M.W ., Kwok, K.A. (2004), Water Treatment and Pathogen Control, Process Efficiency in Achieving Safe Drinking Wate1; WHO Drinking Water Series, !WA Publishing MWH (2005), Water Treatment Principles and Design, second edition, John Wiley & Sons, Inc.

technical features


~fereed paper

health-related issues

WATER RECYCLING: ENDOCRINE DISRUPTING COMPOUNDS H Chapman, F Leusch Abstract T wo recent studies have measured the total estrogenic (feminising) and androgenic (masculinising) activity in the treated wa ter fro m 13 municipal sewage treatment plants in southeast Queensland. T he results, as indicated by in vitro molecular and cellular bioassays as well as in situ sampling of mosquito fish, are very low, mostly below detection limits.

Introduction During the last few decades, the focus on chemical pollu tion in water has been largely directed towards the well-known "priority pollutants" especially those displaying persistence in the environment. T here are well established methods fo r assessing th e risks posed by these chemicals and guidelines are usually available. At the same

The potential for significant endocrine disruption from treated municipal sewage discharges in Queensland is negligible. time there has been a transfer of hazard assessment critical co ntrol point (HACCP) approaches from the food industry to water as in the most recent Australian Drinking Water Guidelines (ADWG 2004) . This method is appropriate fo r known or betterunderstood hazards such as pathogens, but the available data for some substances fo r wh ich there are no guideli nes (e.g. co mplex mixtures and 'emerging' co ntam inants such as hormones, drugs and personal care

products) is limi ted. Conseq uently the methods fo r assess ing risk are less well developed. Endocrine dis rupting com pounds (EDCs), so metimes also kn own as hormonally active agents or endocrine modu lators, are substances that can affect rhe endocrine system in animals includi ng humans and fis h. When peo ple read or hea r reporrs about effects in fis h or other aq uatic life they may wonder whether rh ey should be co ncerned about similar effects in humans. There are two considerations here. I . No studies to dare have co ncl usively linked rhe low concentrations of EDCs in wastewater to adverse health effects in humans (Damstra et al 2002), and; 2. T he effects observed in fish and orher aquatic organisms, and attributed to endocrine disruption ca n have other causes

WS750 MODEL Automati c Wastewa ter/ Stormwater Sampler


Ru gged & lig ht wei gh t v' Outpu t provided for use w ith op tio nal datalogger

LEVELTROLL 300 Water Level Logging



v v

Ideal for long-term water level monitoring Ultra rugged, all stainless steel design Internal memory of 50,000 date points v Low cost, high quality data


1 (03)96C6 " 100

F (0310646 4195



T (02J9't1 l!i13 f (01)9-117 7600

-"''" IJRJSIJAN[ ..,. r (07)38526111



r {08) !Mn sn1

F (06)11'4n3m

Online DO Analyzer

v v v

No membranes requ ired Uses revol utionary optical sensor technology Ultra reliable Not flow dependent

Journal of the Australian Water Association




(07) 38526155



T (09)377576&

F (091377 5768

Kl.lAL.ALUMPUR T 60 (03) 7067 83&6

F 00 (03) 7967 8387 -OOQUOpoom

Sales & Rentals


technical features


B .... • •. •.

health-related issues (Sumpter and Johnson 2005). This is nor to say char the effects cannot be associated with endocrine disrupting chemicals, just that they may or may nor be.

Table 1. Estradiol equiva lents (EEq) for all 12 sewage treatment plants mo nito red determined w ith sheep uterine estrogen and rainbow trout brain androgen receptor binding assays, respectively. Raw = raw sewage; Plant Eff = pla nt effluent; Fin Eff = fi nal effl uent.

What is Endocrine Disruption?


During the last few d ecades a number of studies have implicated environmental chemicals with adverse changes in human health. Increased rares of testicular and breast cancers and possible foetal abnormalities due to exposure to certain compounds have been reported, as have a decline in quality and quantity of sperm (Nollar et al 1990; Sharpe and Skakkebaek 1993; Sharara et al 1998; Birkett and Lester 200 3) . I t has been hypothes ised chat substances in the environment that can interfere with normal endocrine function may cause these health problems. The potential for endocrine disruption has been identifi ed since the early 1900s (Dodds et al 1938) but on ly recently has this phenomenon emerged as an environmental and human health issue. During the 1950s and 1960s the pesticide DDT was shown to be estrogenic and to affect the reproductive system of birds and mammals. These find ings together with the publicatio n of "Silent Spring" (Carson 1962) and in the 1990s with the p ublication of Theo Colborn's "Our Stolen Future" (Colborn et al 1996) brought into public light the scientific evid ence chat certain wildlife were exhibiting abnormalities due to exposure to environ mental pollutants referred to as endocrine d isrupting chemicals.

SlA< SlA< S2A< S2A< S3A S3A S3A S4A SSA S6A SSA S9A Sl0A Sl lA Sl2A Fl A

STP Type"

Flow (m 3/ d)

Head (people)

Sample date

s s s s s s s s s s s s s s s

330 194 113 109 980 980 980 1,500 5,900 3,500 6,800 130,000 12,000 26,000 60,000 570

1250d 750d 430d 420d 3,800d 3,800d 3,800d 6,000 23,000d 13,000d 26,000d 750,000 45,000 100,000d 230,000d 2,200

7/03 7/03 7/03 7/03 8/00 8/02 7/03 7/03 7/03 7/ 03 7/03 7/03 7/03 7/03 7/03 6/03


EEqb (ng/L ± SE) Raw Plant Eff

50 ± 20 29 ± 12 <4 15 ± 6.8 54 ± 26 76 ± 11 93 ± 5.0 125 ± 8.5 137 ± 16 163 ± 1.5 81 ± 30 51 ± 6.9 66 ± 18







4.2 ± 1.7 <4 BDL BDL

<4 <4 6.4 ± 3.0


a . Type of secondary treatment: {F) fixed film {trickling filter); {S) suspended film {activated sludge). Some sample was also taken at the end of a constructed wetland {final effluent from plant F1A) or after advanced treatment (final effluent from plant S 1A). b. Values are the mean of assays with receptors isolated from 2 different sheep, ± SE. Accurate quantification limit was 4 ng/ L, and method detection limit was 1 ng/L. Samples below detection limit are indicated as "BDL", samples where activity was detected but was too low to be accurately quantified are marked as "< 4 ". c. Indicates this plant is located in a tourist area, and flow can vary depending on tourism season. d. Actual head not available, equivalent population (EP) estimated based on average daily wafer use per person in Australia {260L/person/d). ord er to enable quantification of risk to humans.

Clear evidence of adverse effects was shown following pharmaceutical use of the synthetic estrogen diechylsrilbesrrol given to women ro prevent miscarriage in the 1940 1970s. Children whose mothers took rhe drug have shown higher than normal rates of cancers and malformations associated with reproductive organs (Nollar et al 1990 ). These effects were associated with in utero exposure at early develop mental stages. However the weigh t of evidence for endo crine disruption in humans in relation to environmental chemicals is limited and the issue remains controversial (D egen and Bolt 2000; Safe 2000).

As water becomes scarce due to population growth and climate change, direct and indirect potable reuse o f water become an attractive means of providing a sustainable sou rce of water, however some issues need to be addressed before it can be implemented. There are public concerns over reuse of reclaimed sewage water due to a perception that there may be chemicals remaining in the water after treatment chat may have an adverse impact on human health and/or the environment.

The W orld Health Organization (WHO) concluded in 200 2 chat based on the epidemiology evidence exposure to environmental levels of endocri ne disrupting chemicals has not yet been demonstrated to cause harm in h umans (Damstra et al 2002). While there are documented impacts to wild life downstream of created sewage discharges (G agne et al 2002, Damscra et al 200 2; ] obiing and Tyler 2003), it is essential to determine if adverse effects are likely to occur in response to accual exposures in

There have been examples in Australia where community groups have taken action to prevent water reuse projects from proceeding because of perceived health risks. These included concerns regarding the possible presence o f 'gender bender' hormones, hospital and abattoir waste in the recycled water (Po et al 2 00 3) . In the case of the Caloundra-Maroochy Strategic Wastewater Management Study (Scenekes et al 2001), the project was rejected by the community partly because of concerns chat

How Did it Become a Global and a Local Issue in Queensland?

62 SEPTEMBER 2006 Water Journal of the Australia n Water Association

endocrine disrupting chemicals would not be removed by the treatment process, concerns which could not at the time be refuted by scientific data. T his led to a line of investigation chat is still continuing today.

Exposure and Effects Assessment T wo recent stud ies have measured the total escrogenic (feminising) and androgenic (masculinising) activity in sewage from 13 municipal sewage treatment planes in southeast Queensland (Leusch et al 2005, 200 6a). The activity was measured with estrogen and androgen receptor bind ing assays, which measure the ability of compou nds in the sample to interfere with the steroid receptor (the first seep of a receptor-mediated endocrine response). T he results fo r escrogenicicy from chat study are presented in Table 1. The escrogenic activity in raw sewage was, as expected , relatively high (up to 163 ng of estradiol equivalents, EEq, per L), as was the androgenic activity (up to 9330 ng of testosterone eq uivalents, TEq, per L). Overall, activated sludge sewage treatmen t was very effective at removing both

technical features

health-related issues esrrogenic and androgenic activity from sewage. Esrrogeniciry in rhe final effluent was below derecrion limit (< 1 ng/L) at 7 of the 12 STPs tested, below quantification limit(< 4 ng/L) at 3 of rhe STPs and 4.2 ng/L at another. A trickling fi lter plant (F lA) registered 6.4 ng/L in rhe filter effluent, bur this became undetectable after rhe subsequent maturation lagoon (T able 1). These levels are relatively low compared with overseas studies, where esrrogenic activity in final effluents of <1-16ng/L in the Netherlands (ER-CALUX assay, Murk et al 2 002), 4-35 ng/L in Japan (yeast assay, Onda et al 2002), 6 ng/L in Germany (EScreen, Korner et al 2000), and <3- 13 ng/L in the U nited Kingdom (yeast assay, Kirk et al 2002) have been reported. T rickling filters are the dominant type of secondary treatment technology in use in the United Kingdom (Angus et al 2002), which m ay explain rhe comparatively higher levels of these chemicals in treated sewage there Qobling et al 1998).

Conclusion From these studies, we concluded rhar the estrogenic and androgenic activity of treated municipal sewage at rhe Queensland sewage rrearmen r plants res red is very low, as indicated by in vitro molecular (Leusch et al 2006a) and cellular bioassays (Leusch et al 2005) as well as in situ sampling of mosquito fish (Leusch et al 20066). T his suggests rhar rhe potential for signifi cant endocrine disruption associated with treated municipal sewage discharges in Queensland is negligible, as exposure is very small. Secondary crearment using suspended sol ids p rocesses has been shown ro be very effective at reducing the esrrogenic activity in sewage treatment plants in southeast Queensland. The treatment processes remove most of the activity in primarycreated sewage, in most cases to levels below the quantification limit (< 4 ng/L EEq, Table I) (Leusch et al. 2005; 2006a) . Ir is essential ro ensure public acceptance of alternative sources of water such as wastewater reuse. Therefore there is a need for a mechanism rhar will fac ilitate discussions between various in terested grou ps so rhar concerns can be properly add ressed. Such a process is paramount if the best solutions available are ro be accepted. Public perceptions and acceptance of water reuse as well as proper communication of scientific find ings are now recognised as rhe main ingredients of success for any reuse project.

The Authors Dr Heather Chapman is program leader fo r the Sustainable W ater Sou rces p rogram

in the CRC Water Quality and Treatment and also project leader for rhe Global Water Research Co alition project on development of a biological toolbox ro d etect escrogeniciry in water, heather_chapman@health.qld.gov.au. Dr Fred Leusch is currently the principal research fellow on a Global Water Research C oalition/WERF project ro develop a toolbox of bioassays to detect esrrogenicity in environmental waters (http://www.edcroolbox.info). He has experience in adapting and validating biological assays and molecular biology techniques, as well as applying those tests ro monitor the efficacy of water treatment processes, f.leusch@griffith.edu.au

References ADWG (2004) Australian Drinking Water Guidelines 6. National Water Quality Management Strategy, N H M RC & NRMMC. Angus R, Weaver S, Grizzle J, and Warson R (2002) . Reproductive characteristics of male mosquico fish (Gambusia affinis) inhabiting a small south-eastern US river receiving treated domestic sewage effluent. Environ. Toxicol. Chem. 21, 1404. Birkett JW, and Lester JN (2003). Endocrine

disruptors in wastewater and sludge treatment processes. Lewis Publishers, London, U niced Kingdom Carson, R. (1962) Silent Spring, London (England), H amish H amilcon C hapman, H eather F. (2003) Removal of endocrine disrupcors by tertiary treatment processes in subtropical Queensland. Water Science and Technology 47(9) : 151-156 Colborn T, D umanowski, D & j. Myers (1996) Our stolen future Dutton, Penguin Books (NY) in 1996 (ISBN 0- 525-93982-2). Damscra, T., Barlow, S, Bergman, AR and van der Kraak, G . (2002) Global assessment ofthe

stau-ofthe-science ofendocrine disruptors, World H ealth Organisation/Internacional Program on Chemical Safery. Degen, G. H . and H . M . Bolt (2000). Endocrine Disrupcors: Update on Xenoestrogens.

International Archive ofOccupational and Environmental Health 73(7) : 433-441. Dodds EC, Goldberg L, Lawson W, and Robinson R (1938). Oesrrogenic activity of certain synthetic compounds. Nature 141,247 Gagne, F., Blaise, C., Aoyama, I., Luo, R.,Gagnon, C., Couillard, Y., Campbell, P. and Salazar, M. (2002) Biomarker study of a municipal effluent dispersal plume in nvo species of freshwater mussels. Environ. Toxicol. 17:149-159 ] obiing, S., Nolan, M. , Tyler, C.R., Brighry, G ., and Sumpter, J.P. (1998). Widespread sexual disruption in wild fish. Environ. Sci. Technol. 32(17): 24 98-2506 ]obiing, S., and Tyler, C.R. (2003). Endocrine disruption in wild freshwater fish. Pure Appl Chem, 75( 11- 12):2219-2234. Kirk L, Tyler C, Lye C, Sumpter J (2002). Changes in estrogenic and and rogenic activities at different stages of t reatment in


........ .

wastewater treatment works. Environ. Toxicol. Chem. 21, 972. Korner W, Bolz U, SiiBmuch W , Hiller G, Schuller W, Hanf V, and H agen maier H (2000). Input/out put balance of escrogenic active compounds in a major municipal sewage plant in Germany. Chemosphere 40, 1131. Leusch, F.D.L., Chapman, H.F. , Koerner, W., Goonerame, S.R., and T remblay, L.A. (2005) Effi cacy of an advanced biological nutrient removal plant in Queensland (Australia) to remove escrogenic chemicals. Environ. Sci. Tech. 39(15):5781-5786 Leusch, F.D.L., Chapman, H.F., van den H euvel, M .R., Tan, B.L.L., Goonerame, S.R., and Tremblay, L.A.(2006a). Bioassayderived androgenic and est rogenic activity in municipal sewage in Australia and New Zealand. Ecotox. Environ. Saf doi: 10.1016/j.ecoenv.2005.07.020 Leusch, F.D.L., C hapman, H.F., Kay, G.W., Gooneram e, S.R., and Tremblay, L.A. (20066). Anal fin morphology and gonadal hiscopathology in mosquico fish (Gambusia holbrooki) exposed co created mun icipal sewage effluent. Arch. Environ. Contam. Toxicol. doi: 10.1007/s00244-005- 1040-5 Murk A, Legler J, van Lipzig M, Meerman J, Belfroid A, Spenkelink A, van der Burg B, Rijs G, and Vethaak D (2002). Detection of escrogenic potency in wastewater and surface water with three in vitro bioassays.Environ. Toxicol. and Chem. 21, 16. Nollar, K. L., O'Brien, T., Colcon, R., Kaufman, R. and Melcon, L. J. (1990) Medical and surgical diseases associated with in utero exposure co diechylscilboesrrol. In: Clinical practice of Gynaecowgy. (No liar, K.edi cor) . Elsevier Science Publishing Co. Inc., New York. pp 1-7. Onda K, N akamura Y, Takatoh C, Miya A, and Kacsu Y (2003) . T he behavior of escrogenic substances in the biological treatment process of sewage. Water Sci. Technol. 47, I 09. Po M, Kaercher JD, and Nancarrow BE (2003). Literature Review of Faccors Influencing Public Perceptions of Water Reuse. CSIRO

Land and Water Technical Report 54103. Safe, S. H. (2000). Endocrine disruptors and human health - is there a problem? An update. Environmental Health Pmpectives I 08(6): 487-493 Sharara, F. I. , D . B. Seifer, etal. (1998). Environmental coxicants and female reproduction . Fertility and Sterility 70: 613622. Sharpe, R. M . and N. F. Skakkebaek (1993). Are estrogens involved in falling sperm counts and disorders of the make reproductive tract. Lancet 341: 1392- 1395. Stenekes N, Schaefer AI, and Ashbolc N (2001). Community involvement in water recyclingIssues and needs. In AI Schaefer, TD Waite & P Sherman (Eds.), Recent Advances in Water Recycling Technologies. Workshop proceedings, 26 November 200 1, Brisbane, Qld, Sumpter, J .P. and Johnson, A.C. (2005) Lessons from Endocrine disruption and their application to other issues concerning trace organics in the aquatic environment. Environ. Health Perspectives. 38:4321-4332.

Journal of the Australian Water Association



TOOWOOMBA WATER SUPPLY OPTIONS P Selmes, R Bain, A Domanti, W Zillmann, I Cameron Abstract 35,000

T oowoomba is a growing city of over 125,000 resid ents, including surround ing shires serviced by the Council, on top of the ranges 100 km west of Brisbane. I t has been overdrawing its water supply for some yea rs, exacerbated by a long period of red uced rain fa ll , and despi te Level 4 restrictio ns a crisis is develop ing. On behalf of the South East Queen sland Regio nal W ater Supply Strategy (SEQ RWSS) process, the D epartment of N atural Resources, M ines and W ater (NRMW) co mmissioned Parsons Brinckerhoff Australia Pry Ltd (PB) to undertake an independent review o f recently identifi ed water supply options for T oowoomba and surrounding shires. On both economic and sustainability grounds, their recommendation was for indirect po table re-use, subject to comm unity p reparedness to accep t ir.

Introduction To owoomba C ity and its surrounding customer shi res have both immediate and lo ng-te rm water supply sho rtfalls. T he area has been under water restrictions for some time, with T CC - Level 4 rest rictio ns in force since A ugust 2005. T he situation has not im proved in 2006, with T oowoom ba receiving only 200 mm of rain to dare, which is less than half o f the average fo r this period . As part of a South East Q ueensland (SEQ) wide review co nducted by NRMW, the yield of T oowoo mba's rh ree water supply storages (Lake Coo by, Lake Perseveran ce and Lake C ressb roo k) was reduced from 17, 180 ML/a to 12,720 ML/a. Fu rther, with the current extractio n of 1,800 ML/a available fro m t he established T oowoomba basalt bo res, the total safe extraction fro m all sou rces is I 4,520 ML/a. As a resul t of th is review, Toowo omba's current d em and of 17,5 10 ML/a (the estimated 2006 non-drought demand, T his is an edited versio n of the Executive Summary of the fu ll report, Future Water Supply Options fo r Toowoomba City a11d Customer Shires: Pre-feasibility Study, } 11ly 2006 ( I 30 pages), which is available o n t he NRMW website, www.nrmw.qld .gov.au.



+ 6,500 MUa Options Under Consideration = 22,720 MUa


201 012011 (approx) New permanent water source required. 5,000 •

2028 (approx)

- -E111pe,cted Demand Med Growth· No DM'1 & Rr s (TCC J uM 2005) - -Ex~ted ~




,ind u.d Growth- With DM's & RT• (TCC June 2005)

•Proposed f uture water demand '#Ith OM's & Rr1


l~ 201 1


202 1






Figure 1. De mand projections and tim ing of future o ption imple me ntati on.

The preferred option was identified as indirect potable recycle, but this was rejected by a plebiscite conducted in July 2006. exceeds its safe su pply and the need to identify sustainable new water supply options to meet short- term needs is pressing. T he D epartment of N atural Resources, Mines and Water (N RMW) commissioned Parso ns Brinckerhoff Australia Pry Ltd (PB) to undertake an independ en t review o f recently identified water supply opti ons for Toowoo mba and surrounding shires .

Options T he options reviewed were: • Op tion l: Indirect potable re-use (IPR): indi rect po table re use of recycled water fro m Wetalla W astewater T reatment Plant through advanced water treatment (involving m em b ran e processes) wi th release co C ooby Dam ,one of the surface storages w hich supplies T o owo omba. • Option 2: I m porting 6500 ML/a water fr om the O akey C reek G ro undwater

Journal of the Australian Water Association

M anagem ent Area (OCGMA), replacing it with C lass A recycled water fo r irrigation . • Optio n 3a: I mporting 6500 ML/a water from the Condam ine Groundwater Management Area (CGMA) . Replacing it with C lass A recycled water fo r irrigation. • Option 36 : Importi ng more water from CGMA - nominally 20,000 ML/a. Similar to Opt ion 3a, but boosting the supply to the irrigacors through the proposal to p ipe effl uent fro m Brisbane to Darling Downs. • Option 4 : Collecting by-product water from coal seam gas p roduction , treating and co nveying it for urban use.

The Review Process The review process involved a pre-feasibility level investigation o f the fi ve o pt ions using the foll owing m ethod ology: • establish common planning assump tions and baselines • verify technical op tions • cost op tio ns • review non-co st factors incl uding sustainability, risk, social and environmental factors • assess options •

make co ncl usions and recommendations.

The existing water sup ply and wastewater planning co nducted by T oowoomba C ity Council (TCC) was reviewed to establish

technical features

common planning assumptions and baselines. Future demands were projected based on population projections for TCC and customer shires. T he impact of longer term demand reductions arising from water conservation measures and the instaJlarion of rainwater tanks under current TCC policy were also included in planning assumptions. Included in rhe review was a preliminary investigation using a Level of Service (LOS) approach based on the work undertaken for the SEQRWSS. Existing supply side options were reviewed based upon the latest analys is of yield for surface water storage systems and long-term ava ilable supply from groundwater systems. Figu re 1 provides a summary of demand and supply source assumptions used as the context for this review. Figure 2 is a map of T oowoomba and its surrounds, involved in the study.

Detailed Review of the Options

,.. ,






.! /









- -- .*,.o,·._-. .,. .,II -__--------(

L...... •

~ N







10 ...

Figure 2. Map of the area.

Option 1 This option was proposed by T CC and was named the Water Furures Toowoomba initiative. This initiative involved a number of measures bur principally indi rect potab le reuse (IPR) of water from T oowoomba's Weta lla Sewage Treatment Plane by co nstruction of an advanced water treatment plane that would treat efflu ent to a suitable level for pumping in to the storage reservoir, Cooby Dam. Implementation of this option will have rhe benefit of recycling a total of 6,500 ML/a comprised as fo llows: • 5,000 ML/a of IPR fed into Cooby Dam from the advanced water treatment plant • 1,000 ML/a of recycled water sent to local Hampton lrrigators and replaced with 1,000 ML/a of bore water (extracted from Geham bore) • the remaining 500 ML/a will be subsrirured by dual reticulation use at a green field development at the High fields estate Key issues associated with chis option include: • ga ini ng public support for rhe scheme • esrablishing the sustainabil ity of the scheme with its added diversity, renewabi licy . and independence from climatic variab ility • managing public health risks (real and perceived).

Option 2 T his option involves sourcing water from bore licence holders in the Oakey Creek Grou ndwater Management area (O CGMA) which is approximately 40km west of Toowoomba, in rerurn fo r treated effluent from the Weralla Sewage Treatment Plant.

The expectation is char up to 6,500 ML/a is available from this source. An extraction of 8,528 ML/a from OCG MA raw bore water would be req ui red to provide 6,5 00 ML/a of potable water based on a 75% recovery rare from the water treatment process. An estimated 22 bores would be required to service this demand. Key issues associated with ch is option incl ude: • high capital cost of this option • Ii mi red avai lability of the water resource • difficulties in removal/transfer of existing water entitlements • potential sustainabi lity of rhe proposed scheme.

Option 3a Th is op tion involves partial sourci ng of future demands (6,500 ML/a) from bore licence holders in the Condamine G roundwater Manage ment Area (CGMA), which is some 70km south west of Toowoomba, in return for treated effluent from Wetalla Sewage Treatment Plant. Ir is expected char an extraction of 7,871 ML/a of bore water wou ld be requi red to provide 6,500 ML/a of drinking water allowing for an 80% recovery rare from rhe water treatment process. Ir is anticipated that approximately 11 bores wou ld be required to service rhe needs of this option. Key issues associated with chis option include: • high capital cost • limited ava ilability of rhe water resource given the over-allocated nature of the existing groundwater area

• difficu lties in removal/transfer of existing water entitlements • potential sustain abili ty of rhe proposed scheme.

Option 3b This option involves sourcing considerably more water fro m bore licence holders in the CGMA in return for created effluent from Weralla Sewage Treatment Plane boosted by supply from the proposed Brisbane to Darling Downs effluent pipeline. The estimated long-te rm extraction from the CG MA is 23,100 ML/a including an existing 3,661 ML/a urban entitlement. Therefore the remaining volu me of water char can be sustainably extracted from the CGMA for supply to Toowoomba is assumed to be I 9,439 ML/a. Allowing fo r losses during the treatment process, 16,053 ML/a of water can be retu rned to Mount Kynoch Water T reatmenc Plant fo r drin king water purposes. Key issues associa ted with this option incl ude: • high cap ital cost (note: th is analysis assumes no capi ta l conrribucion to the proposed Brisbane to Daring Downs effluent pipeli ne) • dependence on rhe Brisbane to Darling Downs effluent pipeline project proceeding • difficulties in removal/transfer of existi ng water entitlements.

Option 4 This option is to import rhe by-product water extracted during coal seam gas production from rhe gas fields southwest of Chinch illa, approximately 190km from T oowoomba. Ir has been claimed char up to

Journal of the Australian Water Association



technical features

Table 1. Cost Comparisons Comparator Avai lable supply (ML/a) Capital cost Cost to TCC after subsidies Annualised capital cost (50 yr) Annualised cost to TCC after subsidies Annualised unit cost Annualised unit cost to TCC after subsidy

Option 1 IPR via Cooby Dam

Option 2 Oakey Creek CMA

Option 3 Condamine GMA

Option 3b Condamine GMA +

Option 4 Coal Seam Gas





73.2 m 19.4 m 6.0 m l.6 m l .68/kl

148.0 m n/a 12.1 m

275.4 m

6,500 224. l m

n/a 22.5 m

n/a 18.3 m

n/a 2.64 /kl

196.8 m n/a 16.l m n/a 3.31 / kl

1.06/ kl



n/a 3.67 / kl n/ a

n/a 3.8 1 /kl n/a

n/a = not available 6,500 ML/a is available from chis source and the gas fields south-west of Dalby. Assessment of this option was limited to a desktop appraisal. Sustainability of the water resource depends on a range of factors including the number of wells in the area, the thickness of the coal seam, permeability of the coal seam and the gas content. W ith an assumed long-term water production rate of 5 L/min/well about 2,500 wells would be required to supply 6,500 ML/a. Given the level of development of CSG fields elsewhere in the world, chis level of development may not be unreasonable within the next 10 years. Key issues associated with this option include: • h igh capital cost of this option • questionable long-term sustainability of the water resource owing to high variability of flows and linkage to commercial gas production activities • full dependence on commercial gas production activities.

Options Assessment The options assessment methodology used was undertaken by comparing the options on cost, sustainability/reliability issues, and also through a qualitative triple bottom line assessment approach (social, environmental and economic consideratio ns.) Assessment of the costs was based on the fo llowing parameters: • capital cost • annualised capital cost (for 20-year and 50 -year life cycles) • operating and maintenance costs (p/a) • annualised unit cost.

• does the option provide a renewable water resource? • is the option dependent on climatic conditions?


the community's preparedness to accept indirect potable re-use of recycled water as a source of drinking water.

• does the option require community acceptance/transfer of entitlements to ensure long-term sustainability?

Other options considered under this review were considered to be highly limited, O p tions 2 and 3a were considered to be unsustainable from a water resource availability perspective without signifi cant relinquishment of existing groundwater entitlements. It is also noted that the cost penalties associated with Options 2 and 3a were significant when compared to the preferred option.

The TBL approach considered in this investigation includes the fo llowing parameters: • social - issues affecting society as a result of the option proposed • environmental - issues affecting the general environment built and the natural environment as a result of the option proposed • economic - broader economic issues chat may arise owing to the option proposed (i.e. those above the economic viability of the individual options). The cost estimates included consideration of the likely subsidies from Federal and State sources. • For Option 1: It is u nderstood chat government subsidies for Option 1 were to be 33.8% federal fu nding, 33.8% state fun di ng, and 26.5% Toowoomba City Council fu nding. It is also to be noted chat Acland Coal M ine was to p rovide a capital contribution to the project. • Options 2 to 4: Subsidy level for these options is not known at the time of this report. Consequently, no furthe r cost assessment has been undertaken to consider applicable subsidies for these options. A final decision on subsidy level would be at the discretion of the fede ral and state governments. A summary of the cost assessments is p resented in Table 1.


Assessment of the sustainability/reliability of each water supply option included the following considerations:


• are proposed extractions (from groundwater sources) sustainable given current allocations?

PB 's review concluded that while all options may be theoretically achievable, Option l, Indirect Potable Reuse was the preferred option. It has the lowest capital and operating costs and provides a diverse and renewable water resource. The most signifi cant factor affecting this option was

Journal of the Australian Water Association

Options 36 and 4, as assessed, rely heavily on external projects proceeding to ensure sustainability of the water source. In both op tions there are no guarantees that any of these sources will become available either now or in the futu re. Whilst medium-term options for securing future water resources for Toowoomba have been investigated under chis study, investigation of the longer water supply options to meet ultimate demands beyond 2030 will need to be undertaken in the context of the outcomes o f this study and the public vote on indirect potable re-use in T oowoomba.

The Authors Phil Selmes is a Principal Water Engineer with Parsons Brinckerhoff who lead s Queensland's integrated water cycle management group. Roderick Bain is a Senior Hydrogeologist with Parsons Brinckerhoff with a diverse background in groundwater, p ipeline design and mine water impact assessments. Anthony Domanti is a Senior Water Engineer with Parsons Brinckerhoff with extensive experience in water and wastewater system planning and design. Will Zillmann is a Senior Engineer with Parsons Brinckerhoff with an extensive background in pipeline manufacture, design and construction. Ian Cameron is the Queensland Water Executive for Parsons Brinckerhoff with a diverse background across all areas of water engineering. Email: I Cameron@pb.com.au

fereed paper


An Overview of Modelling

T his paper discusses models and their application to the study of groundwater movement in irrigation districts. Math ematical models are the most prevalent in groundwater investigations and these can be separated into funct ional and physical models. T he discussion of groundwater models is based on how such models can be used to represent salinity risk. Key aspects of fo ur studies have been compared in relation to the types of models chosen and che requirements for the models.

A number of authors (van der Lelij 1989; Hoxley 1997; Shaw 1997) have attempted to define a model. Hoxley's defin ition is: "A model is a numerical approximation of real world processes which we use to evaluate the Likely response or behaviour ofa real system"(Hoxley 1997 p.l 14). Van der Lelij described modelling as "an attempt to duplicate the behaviour ofnatural systems by using the laws ofscience and mathematics" (van der Lelij 1989 p.56) . Bell amy (1996; cited in Shaw 1997) stated char a model is "considered an abstraction ofreality used to provide conceptual clarity about a complex real world situation and to provide prediction ofoutcomes with a change in inputs. "

Introduction One of the major adverse outco mes of irrigation in Australia has been the onset of land salinisacion, resulting in substantial environmental degradation, as well as significant losses in agricultural prod uction and damage to infrastructure. Faced with an expensive and insidious problem char is difficult to map and pred ict, researchers and practitioners have developed a diverse range of approaches to estimating and predicting salini ty risk. These include groundwater modelling, rule-based classifi ers using expert knowledge, the use of surrogate variables and datasets such as satelli te imagery, and GIS and spatial data analysis combined with expert knowledge. Groundwater modelling provides the context fo r rhis discussion. The aim of chis paper is to compare fou r cried and true groundwater models to assist project managers in the determin ation of the key characteristics requi red fo r such modelling. First is a general overview of modelling through a discussion of model defi nitions, purposes, and development. Second , modelling is discussed in more detail, with a description of different types of models. T hird is a discussion on how groundwater models have been used to represent salinity risk.

Comparing four tried and true groundwater models to assist project managers.

Models and the modell ing process are useful throughout an investigation of a problem. Models help put various aspects of a problem together and estab lish a co ntext (Prachapar 19976). Th ey also facilitate rhe co mbining of expert and general knowledge from differem backgrounds (Poulton 1997; Prachapar 19976). Models, and rhe model development process assist in the collation, processing an d structu red analysis of data, as well as identifyi ng data gaps, and facilitating monitoring program design Qoseph 1997; Prachapar 19976; Shaw 1997). Models also provide a framework to examine flaws in the logic of concepts and model construction through eval uation by ochers. This can lead to improved understanding of nacural processes (Shaw 1997) . ln che specific context of grou ndwater, models, and the model development process are important because models enable the nature of groundwater systems to be defined, and to understand how they work in relation co the landscapes in which they occur (Coram et al., 2001 ). Th is then assists in the selection of management options (ibid). Models can also be used to cry out management options more cheaply than can be done in the real world (Hoxley 1997). This arises because natural systems are inherencly complex, exhibiting a high degree of heterogeneity, and a long response time to changes, making shortterm experimems impossible (ib id). Furthermore, with intervention in natural

systems there is a high potential for largescale error chat would cause great environmental and social disruption (ibid). A study by Fraser (2002) noted several considerations when developi ng or choosing models. T hese include: the co mplexity of the world being simulated, the purpose of the model, the model's level of deta il and applicability to the real world, the specificity of outcomes created by the model, and co nstraints imposed by rhe geographical area. Furthe rmore, different models can be used to cackle the same problem, with some models bei ng more useful or accurate than ochers (Fraser 2002). Results from different models should be compared, and it is possible to develop a new approach by comb ining approaches used in differe11t models (Fraser 2002).

Types of models Environ mental simulation models can be divided in to three gro ups (Sceyaerc 1993) : I) Scale - an exact replication of the object or system bur at a d ifferent size e.g. model plane 2) Conceptual - where major systems and processes are shown as a flow chart and 3) Mathematical - where existing knowledge is integrated into a logical framework of rules and relationships (Moore et al., 1993). Within the realm of groundwater modelling, mathematical models are most prevalent. These mathematical models can be subdivided imo different groups based on how they deal with randomness, spacial variation , and time (Maid ment 1993). Furthermore, mathematical models can be categorised based on thei r co mplexity and how they link variables together (Moore et al., 1993) . Mathematical models ca n range from simp le empirical regression equations (functional models) through to a number of secs of complex differemial equations derived from fu ndamental physical concepts (physical models) (Moore et al., 1993) . Fu nctional models give a simplified treatment of fundamental processes with modest inputs (Shaw 1997). Examples of functional models include: mass balance approaches (e.g. sale balance= sale at scare +

Journal of the Australian Water Association



technical features

salt imported - salt exported ) an d derived regression equations describing a system's response to changes in certain variables. Because of their relative simplicity, functional models have low data requirements. H owever, fu nctional models have limitations in determining processes operating under transien t conditions (Shaw 1997). Physical models are based on che physical laws of mass and energy conservation or empirical equations derived from independent research (Bath urst and O'Connell 1993). These models are deterministic and contain parameters that have physical meaning as well as being able to be measured in the field, for example, soil conductivity (ib id). Physical models can be applied to a wide range of data as the constraint is the applicability of the physical laws contained in them . Calibration, or adj ustment of model parameters, can occur on the resu lts of fieldwork (Bath urst and O'Connell 1993). To determine what fi eldwork is required in the context o f groundwater modelling in irrigation districts, an understanding is needed of the factors th at affect the change in dep th to groundwater. T hese factors govern how the models described below are constructed.

Groundwater Modelling T h is section discusses how groundwater models have been used to represent salinity risk. Firstly, four papers (Chiew and McMahon 1991; Storm and Punthakey 1995; Prathapar et al., 1996; Demetriou and Pu nthakey 1999) are of particular interest and will be heavily drawn upon. Fo llowing a summary, key aspects of the scudies will be compared. The comparison will focus on the types of models chosen, the data required, the processes represented by the models, the relevan ce of predicted outcomes to salinity risk and methodological issues that arise.

1. Modelling environmental change in the Wakool Irrigation District - Storm and Punthakey ( 1995) Storm and Pun thakey (1995) present the development and application of a surface and grou ndwater model in the Wakool Irrigat ion D istrict to predict the environmental impact of different management options. M IKE-SHE, an integrated model was chosen to simulate surface and groundwater movement within the d istrict. MIKE-SHE's ability to represent interaction between su bsurface and surface water systems was a crucial facto r in its selection (Storm and Punthakey 1995). Spatial variation was considered through the division of the study area into grid cells 2 km x 2 km in size. Processes modelled by MIKE-SHE included groundwater p umping, channel seepage, infiltration and evapotranspiration o n different soils, run-off, cap illary rise, and groundwater flow both vertically and h orizontally. Information requ ired by the model included rainfall and water usage, potential evapotranspiratio n, soil d istribution, land use (from satellite imagery), hydrogeological characteristics, topography, supply channel and d rainage networks, groundwater pumping and groundwater table observations. Calibratio n of the model was undertaken and parameters were modified to ach ieve good agreem ent between simulated groundwater levels and those observed . Insufficient data on stream flows meant calibration of the surface water component could not be undertaken (Storm and Punthakey 1995). The outcomes of the model focussed on representing salinity risk as the area with a depth to groundwater of less than 2 metres. Various management scenarios were tested and the model omcom es reported.

A summary of groundwater modelling studies

2. Evaluating sustainable groundwater management options using the MIKESHE integrated hydrogeological modelling package - Demetriou and Punthakey ( 1999)

Four papers are summarised in th is section. Two papers are reporcs of groundwater modell ing studies undertaken within the Wakool Irrigation District which is located near Deniliquin, NSW. Another paper is a groundwater modelling scudy undertaken in the Campaspe River Basin. Part of the Campaspe Basin is intensively irrigated, and salinity prob lems are present. The last paper is a discussion of a groundwater model that has been developed specifically for irrigatio n areas and has been used in several irrigation districts in New South Wales and Victoria.

T his scudy is an enhancement of the model developed in study (1) (Storm and Punthakey 1995) with more layers and improved calibration (Demetriou and Punthakey 1999). Almost identical datasets were used in both cases, except fo r different satellite images to deter mine land use. The processes modelled are identical to tho se outlined in the section above. The MIKESHE model uses equations based on p hysics, as well as emp irically derived equations to estim ate water flows for key components such as evapotransp iration, overland and channel fl ow, as well as water

68 SEPTEMBER 2006 Water

Journal of the Australian Water Association

movement in the unsaturated and saturated zones (Demetriou and Punthakey 1999). Improved calibration was obtained by checking predicted stream flows against those recorded because sufficien t data was available. Model calibration was undertaken to assess the performance of the m odel. Sensitivity analysis was also conducted to ascertain which variab les stro ngly in fluenced model simulations of watertable levels. In certain areas, the model did not simulate grou ndwater levels well due to the effects of localised flooding which were not considered in the study (Demetriou and Punthakey I 999). T he sensitivity analysis showed that model results were strongly affected by evapotranspiration at a regional level, as well as groundwater pumpi ng and channel seepage at a local level. Hydraulic cond uctivity and aquifer storage coefficient val ues were found to have little effect on watertable levels. Different management scenarios were reseed in the study using the calib rated M I KESH E m odel. Outcomes were reported as percentage of area with a watertable depth of less than the desired thresholds ( 1.5, 2.0 and 2 .5 metres). Salinity risk has been ind icated as a function o f area or percentage area with a shallow watertable, and thus the factors affecti ng the rise and fall of watertables o n sal inity risk were id entified.

3. Groundwater recharge from rainfall and irrigation in the Campaspe River Basin - Chiew and McMahon ( 199 1) To investigate the effect of rainfall and irrigation recharge to gro undwater in th e Campaspe Basin , a surface water model, HYDROLO G, and a grou ndwater model, AQUIFEM-N, were integrated . Recharge in both d ryland and intensively irrigated areas were investigated and predictions were made of recharge u nder different fu ture irrigation intensities. T h ese predictions of recharge were then used to estimare rises in warertables an d the fl ow of groundwater into the Campaspe and M u rray Rivers. The HYDROLOG m odel required daily data o n rainfall, irrigation usage and potential evapotransp iration, whilst the AQUI FEM-N model req uired data on aquifer p roperties for the shallow, deep and bedrock aquifers. This aquifer data included aqu ife r level and thickness, storage coefficient, hydrologic conductivity, specifi c yield, vertical hydraulic conductivity, as well as inform ation on groundwater extractio ns, river stage heights and in itial distribution of potenriometric (or p iezom ecric) heads.

technical features

By integrating the surface and groundwater models, the key processes of infiltration, run-off, evapotranspiration, vertical and horizontal groundwater flow and interactions of gro undwater with river systems were modelled (Chiew and McMahon 199 1). However, capillary rise, a particularly important mechanism in shallow warerrable areas, was igno red. The model was calibrated by adjusting model parameters to minimise the difference between stream flows and piezometric heads created by the model and those acrually recorded (Chiew and McMahon 1991) . Spatial va riation was considered in the modelli ng approach, with the surface model applied independently to each sub-a rea withi n the region. Within irrigation subareas, the surface model was applied to each groundwater element to further improve model ling of spatial va riation in recharge. Once the model had been calibrated, recharge rares, groundwater flows to and from the shallow aquifer and annual warertable rises were assessed under different irrigation scenarios.

4. A soil water and groundwater simulation (SWAGSIM) model Prathapar et al. ( 1996) SWAGSIM was developed for use in shallow warerrab le areas to assess warerrable management options (Prarh apar et al., 1996) . SWAGSIM is a physical model that uses laws of physics to model numerous processes involvi ng water movement ro and from the shallow warertab le. These processes include evaporranspiration from the root zone, recharge to the watercable vi a the soil matrix and cracks, recharge from evaporatio n basins, discharge via cap illary actio n, discharge via mole and tile drai ns and p umps, as well as lateral groundwater flow, leakage to and from the underlying confined aquifer and the interaction betwee n surface channels and rivers and the shallow aquifer (Prachapar et al., 1996). T hus, SWAGSIM captures the major processes causing rapid changes in wacercables driven by rainfall, irrigation and evapocranspiracion (Poulton 1997). In order to model these processes, SWAGSIM requires the following datasets: compo nents of water balance (e.g. rainfall, irrigation , evapotranspirarion, ru n-off), and basic estimates of aq uifer and so il parameters (Poulton 1997) . The SWAGSIM model does have so me drawbacks. First, it assumes fi xed run-off leading to either an over-estimation or under-estimation of water input (Poul ton 1997). Second, SWAGSIM does not model lateral flow in che deeper confi ned aquifer and does not incorpo rate an infiltration

Table 1. Processes represe nted by g roundwater models. PROCESS

MIKE-SHE (Storm, Demetriou)


SWAGSIM (Prathapar)

Surface water Land Use (i ncluding irrigation intensity)

Rainfall Evopotranspiration

• •


Overland Flow

• •

• • •

• •

Groundwater Capi llary Rise

Shallow Aquifer - Lateral Flow

• •

Deep Aquifer(s) - Lateral Flow Vertical Flow between Shallow & Deep Aquifers Surface-Groundwater Interactions

• •


• •

• •

Channel Seepage Pumping Groundwater Discharge to drains Groundwater Discharge to streams

model, assuming instead that water input to the so il profile occurs through micro and macropores (Poulton 1997). SWAGSIM outputs include water balance and groundwater flows (Poulton 1997), identification of recharge and discharge areas (Poul ton 1997; Prathapar 1997a) and piezomecric heads (Prarhapar 1997a) . SWAGSIM was used to model warerrable fl uctuations, identify recharge and discharge zones and evaluate drainage options in the Mead Ridge Drainage Catchment of Victoria (Poulton 1997; Prathapar 1997a). Ir was also used to assess the impact of rice production in the Murrumbidgee Irrigation Area (MIA) of New South Wales (Prathapar et al., 1996) and to evaluate rhe effectiveness of shallow groundwater pumping in rhe same region (i bid).

Salinity risk and groundwater modelling T he four studies outl ined above have so me common features. They all foc us on groundwater depth. Hence, they attempt to simulate, and estimate the effects of various factors such as rainfall, irrigation, managemen t practices, and grou ndwater pumping on warertable levels. Outcomes are generally repo rted as recharge rates or areas with shallow grou ndwater. SWAGSIM also ind icates recharge and discharge areas. All the models summarised are physical models, i.e. they use laws of physics and derived empirical relationships that have physical meaning to predict grou ndwater behaviour. As such, groundwater modell ing is very derailed, with a number of processes in groundwater movement being simulated by che different models, as shown in Table 1.

• •

• •

T he Murray-Darling Basin Commission (M iddlemis 2000) identifies groundwater processes in terms of hydrological stresses. In cl uded are the effect on flows and aquife r levels from sources and sinks; rhe rares, patterns and durations of water movement from recharge to discharge areas; and the interaction berween streams and aquife r. Table 1 shows char the MIKE-SHE model covers all the processes in grou ndwater and surface water movement. However, rhe Campaspe model does nor consider rwo very important variables, namely capillary rise, and channel seepage. Boch capillary rise and channel seepage play an important part in grou ndwater movement in irrigation districts. SWAGSIM, on the other hand, considers these variables but does not rake inro account lateral flow in deep aqu ife rs or overland flow. SWAG SIM also only considers run-off as a fixed percentage of rainfall and irrigation. After ch is point, runoff is no longer co nsidered. Th is raises two key questions: Where does chis run-off water end up? And what is the effect of not incl uding ic? The exclusi on of overland flow from che model is also noceworchy as overland flow is a linking factor as water moves across the landscape surface to rivers and streams. By testing model resu lts against scream flows, the model simulation of run-off and surface water movement can be verified. SWAGSIM's omission of overland flow elimin ates ch is method of verification. However, Storm an d Punthakey (1995) also did not have scream flows for verification and noted chat chis shou ld not affect che si mulation of groundwater. This co nclusion was justified on che basis chat

Journal of the Australian Water Association



an over-esti mation of irrigation water w ill lead to com pensating over-est imations of stream flows or evaporation, nor recharge (ib id). T he D em etriou and Punthakey (1999) and Storm and Punthakey (1995) studies noted a paucity of deep aqu ifer data with in the Wakool Irrigation D istrict. T h is deficit req uires further data acq uisition and recalibration of their models. A second common feature of these groundwater models is their high data req uirements. T he data required by each mod el is summarised in Table 2 . Table 2 shows that each model requires n u merous datasets span ning landform, land use, cl imate, aquifer properties for che different aquifers present in che study area, and human featu res such as drainage lines. The SWAGSI M model requires additional climatic data because it uses a modified Penman equation (a physical model for evaporative dem and) to determine evapocranspiracion (Prathapar et al., 1996) rather than pan-evaporatio n measurements . As well as intensive data requirem ents, phys ical models also require the estimation of param eters co ntained with in th e model. The Campaspe and MIKE-SHE models, in part icular, require the estimation of parameters affecting infiltration and soil mo iscure (Chiew an d McMahon 199 1; Dem etriou and Punthakey 1999). Fo r the Campasp e model, param eter estimatio n also affects horizontal and vertical groundwater flow calculations (Chiew and McMahon 1991). T h e process of calibrating models is used to refine parameter estimates, and is a key part of the reported grou ndwater studies. T he calibration process involves the op tim ising of model param eters to mi nim ise the d iffe rence between simulated val ues of scream flows and/or groundwater, and those recorded (e.g. streamflows and p iezomerric heads) (Schofield 1990; C hiew and McMahon 1991; G rism er and Tod 1991; Salama eta/., 1993; Pavel ic eta/., 1997; Prathapar 1997a; Storm and P unch akey 1997; Clarke et al., 1998a; Clarke et al., 19986). H owever, the calib ration p rocess has a number of drawbacks. First, ic only provid es co nfidence in the mod el over the range o f values tested (H oxley 1997). Second, cali bration restricts the application of the calibrated model to the areas for which ic h as been calibrated (D eller-Sm ith 1997) . Third, calibration on ly p rovides a guide for the model's ability to simulate observed behaviour, not its predictive ability (Bathurst and O'Co nnell 1993). Fourth, any model or set of parameters chat p red icts the variable or variables of interest must be



Table 2. Data used by groundwater models. DATA ITEM

MIKE-SHE MIKE-SHE (Stormi (Demetriou I


SWAGSIM (Prathapar)

• •

• •

Irrigation or La nd Use


• • •

• •

Drainage Lines

• •

• •

• •


Supply Channel Network Groundwater Pumpi ng

• •

• •

Stream Flows & heights

• •

Soi l Type & Distribution

• •

Deep Aquifer thickness

• •

Evapotronspiration Watertable Observations (Shallow aqu ifer) Shallow aquifer thickness

Deep Aquifer Piezometric Head

• •

Wind Speed

• •

Dew Point Temperatu re

Average, Maximum, Minimum Temperatures

considered equally likely as a simu lator of the system (Beven and Binley 1993). T h is arises because all mod el scruccures must be in error to some extent, as must the measurements chat provid e che basis for calibration (ib id). H ence, within any one particular m odel st ructure, no o ne sec of parameter values can be considered as defini tive (ib id) . Th is is particularly true in hydrological scud ies at a catchment level, where a number of differen t mechanisms of hydrological response (e.g. infi ltratio n, overland flow} can occur (ibid). Beven and Binley (1993) ill ustrate this point particularly well using the example of storms in a catchmen t. Ac any one rime, d ifferent hydrological mechanisms can occu r in d ifferent pares of che catchment, fo r example during che same storm (ib id) . Alcernacively, d ifferent mechan ism s can occur ac th e same point ac d ifferent times, for example different storms (ibid). A sim ilar situation ap pl ies to groundwater models where chis issue has not yet been considered. Within che liceracure, all char is reported is chat che groundwater model has been calib rated co o ne particular sec of parameter values. T hese reports do nor indicate che likelihood of the selected parameter values or any other viable alternatives being the defi nitive set co describe the system under consideration . The groundwater models themselves are extremely detailed, requiring a qualified hydrologist co undertake che modelling and interp ret the res ults. The level of derail also dem ands, as a req uirement, a large vol ume of spatially distribu ted data, which may nor be readily available.

Journal of the Australian Water Association

Concluding Comments Predicting sal in ity r isk requires th e modelli ng of groundwater depth and/o r groun dwater sali n ity. The scudies reviewed, d emonstrate t he impo rtance of m o delli ng both su r face an d groundwater m ovem en t and the interactions b etween chem . Thus, var iables su ch as land use or land cover, rainfall , evapo ration, soil typ es, ch an nel seepage, aquifer properties, piezometric heads, groundwater pu m pin g and surface drainage and run-off need co be considered. For modelling groundwater salinity, the movement of salt (via water moveme n t) is the primary co nsideratio n. There is a variety o f modelling app roaches th at can be used to mod el both gro u ndwate r b eh avio ur and salt movement, ranging fro m sim ple fu nct ional models co complex physical m odels. T h us, d ifferent modelling approaches can be used with in the same r isk framework. T h is gives great flex ibi lity, which allows modell ing to be adjusted according to the level o f detail req uired or th e data available. It must be said that grou ndwater modell ing is a dynamic and inexact scien ce and therefore it is important to consider u sing th e outpu t from a n umber of models in any decision making related to groundwater. The Murray-darlin g Basin Comm ission has a publication "Groundwater F low M o delli ng Gu ideline" (Middlem iss 2000 ) wh ich is a valuable resource for those wanting more in for m atio n relating to selecting groundwate r models in a range of hyd rogeo logical settin gs.

technical features refereed paper

The Authors Dr David Fraser is Senior Lecturer in Environmental Dara Analysis, RMIT University, email: david.fraser@rmic.edu.au and Dr Paul Lambie is Scientist & Economist - Sustainability, Land & Water Resources, with GHD Pry Led, Newcastle, 2300, email: nrlm ail@ghd.com.au

References Bathurst, J. C. and O 'Connell, P. E. 1993, 'Future of distributed modelling: che Sysceme Hydrologique Europeen', In Terrain Analysis and Distributed Modelling in Hydrology Eds. K. J. Beven and 1. 0. Moore. John Wiley & Sons Led, Chichester, England, pp. 213-226. Beven, K. J. and Binley, A. 1993, 'The future of distributed models: model calibration and uncertainty prediction', ln Terrain Analysis and Distributed Modelling in Hydrology, Eds. K. J. Beven and I. D . Moore. John Wiley & Sons Led, Chichester, England, pp. 227-246. Chiew, F. H. S. and McMahon, T. A. I 991, 'Groundwater recharge from rainfall and irrigation in the Campaspe River Basin ', Australian journal ofSoil Research, Vol. 29, pp. 651-70. Clarke, C. J. , Mauger, G. W., Bell, R. W. and Hobbs, R. J. l 998a, 'Computer modelling of che effect of revegecacion strategies on salinity in the western wheacbelc of Western Australia I. The impact of revegecation strategies', Australian journal ofSoil Research, Vol. 36, pp. I09-129. Clarke, C. J., Mauger, G. W., Bell , R. W. and Hobbs, R. J. I9986, 'Computer modelli ng of che effect of revegecacion strategies on saliniry in the western wheacbelc of Western Australia 2. T he interaction between revegecacion strategies and major fau lt zones', Australian journal ofSoil Research, Vol. 36, pp. 13 1-42. Coram , J. E., Dyson, P.R. and Evans, W. R. 200 I, An Evaluation Framework far Dry/and Salinity, Bureau of Rural Sciences, Canberra. Deller-Smith, J. I997, 'Salinity predictions using a geographic information system', In Role of Computer Modelling in the Development and implementation ofland and Water Management Plans far Irrigated Catchments. Drainage Program Technical Report No 5, Murray Darling Basin Commission, Canberra, pp. 152- 156. Demetriou, C. and Punchakey, J. F. 1999, 'Evaluating sustainable groundwater management options using the MIKE SHE integrated hydrogeological modelling package', Environmental Modelling & Software, Vol. 14, No. 2, pp. 129-140. Fraser, 0. 0., 2002, Development of a spatially based decision-making construct with application ro rural land management. PhD thesis, RMIT University, Melbourne. Grismer, M. E. and Tod, I. C. 1991, ' Drainage of clay overlying artesian aquifer. I: Hydrologic assessment', journal ofIrrigation and Drainage Engineering, Vol. 117, No. 2, pp. 255-270. Hoxley, G. 1997, 'Groundwater models for salin ity planning', In Role of Computer Modelling in the Development ofland and Water Management Plans far Irrigated

Catchments. Drainage Program Technical Report No 5, Murray Darling Basin Joseph, S. 1997, 'Computer models and land and warer management plans in che Murray Valley - a regional perspective', In Role of Computer Modelling in the Development and i mplementation ofLand and Water Management Plans far Irrigated Catchments. Drainage Program Technical Report No 5, Murray Darling Basin Com mission, Canberra, pp. 71-73. Maidment, 0. R. 1993, 'GIS and hydrologic modelling', In Environmental Modelling with GIS, Eds. M. F. Goodchild, B. 0 . Parks and L. T. Steyaerc. Oxford University Press, New York, pp. 147-167. Middlemis H, 2000, 'Groundwater Flow Modelling Guideli ne', Murray-Darling Basin Commission, Final Guideline - Issue 1, Project No. 125, Aquacerra Consul ting Pry Led Moore, I. 0., Turner, A. K., Wilson , J.P., Jenson, S. K. and Band, L. E. 1993, 'GIS and land-surface-subsurface process modeling', In Environmental Modelling with G!S, Eds. M. F. Goodchild, B. 0. Parks and L. T. Steyaerc. Oxford University Press, Pavelic, P., Narayan, K. A. and Di llon, P. J. 1997, 'G roundwater fl ow modell ing co assist dryland salini ty management of a coastal plain of southern Australia', Australian journal ofSoil Research, Vol. 3 5, pp. 669-86. Poulton, D. C. 1997, 'SWAGSIM application co Mead Ridge project', In Role ofComputer Modelling in the Development and Implementation ofLand and Water Management Plans far Irrigated Catchments. Drainage Program Technical Report No 5, Murray Darling Basi n Commission, Canberra, pp. 80-84. Prarhapar, S. A. 1997a, 'Application of a soil water and groundwater simulation (SWAGS IM) model to the Mead Ridge drainage catchment', ln Role ofComputer Modelling in the Development and Implementation ofland and Water Management Plans far Irrigated Catchme11ts. Drainage Program Technical Report No 5. Murray Darling Basin Commission, Canberra, pp. 77-79. Prathapar, S. A., Meyer, W. S., Bailey, M.A. and Poulton, D. C. I996, 'A soil water and groundwater S!Mularion model: SWAGS IM ', Environmental Software, Vol. 11 , No. 1-3, pp. 151- 158. Salama, R. B. , Laslerr, D. and Farrington, P. 1993, 'Predictive modelling of management options for the control of dryland salini ry in a first order catchment in rhe wheatbelt of Western Auscralia',jo11rnal of Hydrology, Vol. I45, pp. 19-40. Schofield, N. J. 1990, ' Determining reforestation area and disrriburion of salinity control', Hydrological Sciences journal, Vol. 35, pp. 1-19. Shaw, R. 1997, 'The Role of Models in Decision Making for Natural Resource Use and Managemenr', In Role ofComputer Modelling in the Development and Implementation ofland and Water Management Plans far irrigated Catchments. Drainage Program Technical Report No 5, Murray Darling Basin Commission, Canberra, pp. I 3-49.

Sreyaerc, L. T . 1993, 'A perspective on the state of environ mental simulation modeling', In Environmental Modelling with GJS, Eds. M. F. Goodchild, B. 0 . Parks and L. T. Steyaerc. Oxford University Press, New York, pp. 16-30. Storm, B. and Punchakey, J. F. 1995, 'Modelling Environmenral Change in the Wakool Irrigation Oisrricc', MODSIM 95. International Congress 011 Modelling and Simulation, Newcastle, pp. 47-52. Storm, B. and Punrhakey, J. F. I 997, 'Application of MIKE SHE on the Wakool Irrigation Oiscricc', In Role of Computer Modelling in the Development and implementation ofland and Water Management Plans far irrigated Catchments. Drainage Program Technical Report No 5, Murray-Darling Basin Comm ission, Canberra, pp. 66-69. van der Lelij , A. 1989, 'Salt and water movemenr', Management ofSoil Salinity in South East Australia Symposium, Albury, NS ~'(/, 18-20 September 1989, Eds. E. Humpreys, W. A. Muirhead and A. van der Lelij. Australian Society of Soil Science, Riverina Branch, pp. 49-59.

water Editorial Submissions Technical Papers Water journal welcomes the submission of papers equivalent to 3,000-4,000 words (allowing for graphics) relating to all areas of the water cycle and water business co be publ ished in the journal. Topical stories of up to 2,000 words may also be accepted. All submissions of papers intended for the main body of the journal should be emailed to the Technical Editor, bswinron@bigpond.ner.au and hrrps://zeus.econ.umd.edu/wj (Editorial Express). Shorter news items should be emailed to jsage@awa.asn .au. A submitted paper will be tabled at a monthly Journal Committee meeting where, if appropriate, it will be assigned co referees. T heir comments will be passed back co the principal author. If accepted and after any comments have been dealt with, rhe final paper can be emailed with che rexr in MS Word bur with high resolution graphics (300 dpi tiff, jpg or eps files) as separate fi les. Authors should be mindful char Water joumal is published in a 3 col umn 'magazine' format rather than the full-page format of Word documenrs. Graphics should be sec up so char they will still be clearly legible when reduced to two-column size (about 12cm wide) . Tables and figures need to be numbered with the appropriate reference in the text e.g. see Figure 1, not just placed in the text with a (see below) reference as they may end up anywhere on the page when typeset. See index page 2 for more derails on chis and other editorial submissions.

Journal of the Australian Water Association



.fereed paper

TUBULAR POLYMERIC AERATORS FOR WASTEWATER TREATMENT R Galich, Y Meshengisser, V Los, P Spiridonov Abstract This article summarises theoretical methods fo r che design and calculation of aerators and aeration systems, on which bases tubular polymeric aerators have been developed, wh ich combine che functions of che air d istributor and diffuser. Recent cescs by an independent German universiry have shown chat these aerators have higher oxygen transfer races, more stable size of air bubbles, lower hydraulic resistance and work with a wider range of airflow races. Since 1999 they have been employed in l5 countries around the world, serving 120 million p eople.

6 C

E 0)-

.::, 'o

A system of equation s for the behaviour of an air bubble at stages of its formatio n and growth has been developed and solved by standard numerical methods (Meshengisser, 2002).


:0 3 .0



2 1






Po re diameter, mm

Figure 1. Dependence of bubble s ize {D,ep) on the pore diameter {dp) at different speeds of air effusion

(w9 ): 1 - 6 m/s; 2 - 12 m/s; 3 - 20 m/s.

Introduction I n wastewater purification aeration is crucial and is one of the most energy consuming processes, accounting for more than 50% of all coses Th erefore the analysis of theoretical methods for the design of aerators and aeration systems, co guide the development of advanced, effect ive and reliable devices, is extremely imp ortant. Aerator design should meet a combination of contradictory and so metimes mutually exclusive requirements: • air bubble diameter should be minimal to ensure h igh mass exchange rate, but simultaneously be large enough to provide energy for water agitation; • d iffuser resistance should ideally be equal co zero; • the resistance of all aeration systems should be sufficient to maintain even air distribution along air rank length; • aerators sho uld be designed in such a way char in rhe case o f air supply sto ppage and filling of diffusers with water, hydraulic impacts on resumption of air supply would be eliminated or reduced


Theoretical Aspects of Aerator Design

E (IJ 4

system, oxygen transfer efficiency

• diffusers should not be blocked by airborne particles from the inside or be exposed co biofouling from the outside;



Key words: diffuser, wastewater, aeration


design and calculation of the aeration systems, on which bases effective and reliable tubular polymeric diffusers have been developed.


Tubular polymeric aerators, developed in Ukraine, work in a wide range ofair flow rates and are easy to install. • aerators should be interchangeable, easily mounted and adapted co the existing systems of air supply and air tank designs; Analysis of existing aerators and theoretical investigation of rhe aeration process enabled the authors co develop a new type of aerator, which combines the functions of che air distributor and diffuser. This article describes the theoretical aspects of the

--E._.. 1/)

"C Q) Q)

0. Cf)

The speed of the gas effusion has been correlated with the angle of pore edges by modification of the above equations. As shown in Figure 2 the bubble diameter decreases with che increase of che angle of pore edges, which remains typical for different airflow races. These calculations have shown chat with an identical pore d iameter, a tubular aerator produces, on average, smaller bubbles than a fla t o ne.

Oxygen transfer rate The parameter used in aerator calculations is oxygen transfer rate (OTR): =

KlaVR(C* - C)

where KLa is a vol umetric mass-transfer coefficient, V R - water volume in reactor, m 3, C * - dissolved oxygen saturation concentration , C - dissolved oxygen concentration.

14 'iii 12


These calculations are valid for the case where the pore edges are horizontal. Pores in tubular aerators are located at an angle from horizontal and at different distances from the water surface.



Figure l shows chat the diameter of the formed bubbles depends not only on the pore diameter, but also on che speed of the air effusion through the pores.


8 6

4 -90





Angle of pore edges, degree Figure 2. Dependence of speed of gas effusion o n the a ng le of pore edges to horizontal at different airflow rate: 1 - 5 m 3 /(h*m); 2 - 7.5 m3 /(h*m);

3 - 10 m3/(h*m) .

Journal of the Australian Water Association

If the members of equ ation are determined at normal pressure and temperature, and the dissolved oxygen concentration is zero , the value obtained is termed che standard oxyge n transfer rate (SOTR) . A common criterion for comparison and design of aeration systems is the standard oxygen transfer efficiency (SOT E). The physical meaning of chis parameter is percentage of oxygen absorbed by water

when air passes through water char does nor contain oxygen.

smoother, and the saturation efficiency of porous polymeric diffusers becomes higher than chat of perforated membranes.

SOTE, %, according to ASCE (1992), can be expressed as follows: SOTE

SOTR · 100 0,2765 · Qs



where SOT R - standard oxygen transfer race, kg/s; Q, - airflow race at standard conditions (temperature 20°C, pressure 101 ,325 Pa, relative humidity 36%),


0,01 ~ - - - - - ~- - - - - ~ 0,01 0,1 Experimental SOTR, kg/h

Meshengisser and Marchenko (2000) have compared experimental values of oxygen transfer races and values obtai ned by calculations based on bubble size, floating speed and paccern of con race between the rising bubbles and accompanying water flow

Figure 3. Compari son of experimental and ca lcu lated va lues of oxygen transfer rates.

::R 0

The results are shown in Figure 3. T he deviation between calculated and experimental data of oxygen transfer rate was no more th an 8%, which is withi n experimental error.




b [:


\ ~

~ 4





Airflow rate, m /h Figure 4. Dependence of SOTE on aerator airflow rate: 1 - elastic perforated membrane; 2 - rigid perforated membrane; 3 - porous polymeric diffuser.

Analysis of modelling results T o check the developed mathematical model the authors performed calculations for three different types of diffusers: rigid perforated membrane, elastic perforated membrane and porous po lymeric diffuser. The calculation res ults are given in Figure

structure. T hey include ceramic and porous polymeric diffusers . Ac low air flow, air passes only through large pores and fine pores are flooded and closed for air. T he increase of ai r flow race resul ts in the increase of air pressure in the diffuse r which leads to opening of fin er pores. Thus, che average pore diameter is reduced when ai rflow race is increased; accordingly, the SOT E dependence on ai rflow race is less.

4. Despite che face that the average pore diameter was stipulated to be equal for all diffuse rs, their SOTE characteristics are different. T he calculations have confirmed the known fact that SOTE depends on ai rflow rate and reduces when airflow rate increases. T his is explained, firs t of all, by the increase of the diameter of air bubbles upon increase of speed of gas effusion from the pores. Such phenomena are also relevant for membranes with ch e same pore diameter, which do not depend on the airflow rate.

2) Diffusers with polydisperse pore structure provide a constan c oxygen transfe r rate within a wide range of air flow races. Characteristics of the Designed Tubular Polymeric Aerators




Hence, their theoretical model of oxygen transfer qu ite precisely describes the physical process and can be used for the comparative analysis of aerators of various designs.

The opposite picture is observed for diffusers with non-uniform pore



m 3/s.

ln practice such di ffusers are very rare and possibly may include rigid metal places with apertures of equal diameter. Elastic perforated membranes screech with an increase of airflow race, their pore size is increased and, accordingly, SOTE is drastically reduced.


Thus, che theoretical analysis allows statement of the followi ng: 1) Tubular aerators are preferable to flat ones: with similar pore diameters they provide smaller bubbles and therefore larger specific area of liquid-gas interface and hence oxygen transfer efficiency;

Diffusers with polydisperse pore scrucrnre (such as porous polymeric diffuse rs) have lower saturation efficiency at smaller flow rates. However, due to opening of finer pores with increase of the airflow rate, the SOTE dependence on the air flow is

Design features O n the bases of their resea rch, tubular polymeric aerators (Figure 5) have been des igned by the authors and manufactured by Ecopolymer in Ukraine. T he aerators have a number of attributes, which give chem adva ntages over other aerators: • Innovative technology allows the production of a polyd isperse diffuse r which provides fine bubble aeration and minimises hydraulic resistance, down to 200-300 mm of water column. • A disti nctive featu re of che des igned aerators is the gap between the casing and che di ffuser, which provides steady and reliab le operation of aerators due to the redistribution and levell ing of airflow along the aerator ax is, reduction of pressure losses at che dispersing element, and increase of dust holding capacity. • Due to the change of che aeration zone width and opening of pores of di fferent diameters, the tubular aerator is a selfadjusting system, which minimises energy co nsumption used for forcing air through the pores.

• The aerators have an extremely wi de range of steady operation - from 2 to 30 m3/h of airflow per metre of aerator length. • The combination of air pipe and diffuser funct ions allows simpler and faster installation of aeration system and provides increased reliabi lity. They can be manufacrnred as long aeration beams (up to 50 meters). • The aerator structure is demountable. Ir allows regeneration (washing) or replacement of the Figure 5. Tubular aerator: 1 - perforated casing; diffusers without disman tling che 2 - diffuser; 3 - clamping nut; 4 - sealing gasket; whole system. 5 - retaining rin g; 6 - sealing rings. Journal of the Australian Water Association



technical features

.€~ 0

Table 1. The aerators' test conditions a nd results.


G -A

8 6


~ 4

b (f)


tl ~

Tests conducted at



water temperature, °C

2 0

diameter of a column, D, m







volume of water in column, VR, m3 depth of aerator immersion, h, m

Qs , m:?/(h*m)

ratio of geometrical dimensions, h/D

Figure 6. Dependence of SOTEH on the specific airflow rate: 1 - ISAH; 2 - RLE .

diffuser length, I, m

Determination of mass-exchange parameters of the designed aerators was carried out in che Inscicuce of Water Quality and Waste Management of Hannover University, Germany (ISAH, 2003) and in che Research Laboratory of Ecopolymer (RLE). Tests in ISAH were conducted in tap water in a column reactor by che desorption method, (German Standards ATV-M209). Water was saturated by technically pure oxygen up co a concentration exceeding 40 mg/ I. Then, after a pause, air was passed



1.67 3.0 3.75 0.2 4.3 - 15.3

0.69- 2.06

0.51- 1.84

oxygen transfer efficiency per 1 m of depth of aerator immersion, SOTEH, % /M

8.64- 4.86

8.09- 4.77

1.3- 2.0

hydraulic resistance of aerator, kPa

through the aerator until the concentration of dissolved oxygen (CDO) reached equilib rium. To measure CDO an oxygen measuring device COS 3S was used, with th ree sensors placed at depths of 0.5 , 2. 14 and 3.28 m. Three aerators were investigated at three air flow races; each experiment was repeated. The cesc certificates were given for a batch of ae rators.

. • '



.. •


Sand, Gravel, Coal, Garnet, Manganese Greensand

Invest in accurately graded, durable media from your complete filter media professionals.

We are pleased to offer you: • Decades of mineral processing experience • Media produced to the AWWA B100-89 Standard • An extensive product range at competitive prices • Manufacturing in accordance with Quality System AS/NZS ISO 9001:2000 • Packaging alternatives to suit any requi rement • Proficient technical assistance and support • Prompt delivery Austra lia-wide and Overseas So, tap into our extensive experience and helpful service when next you require filter media.



683 Beenleigh-Redland Bay Road, Carbrook Queensland 4130

International +61 7 3287 6444 Freecall 1800 077 744 Facsmile 07 3287 6445


19.7-22 .5

4. 1 - 12.2

Filtration Media


12.5-19.5 1.2 5.09 4.28 3.57

air flow ra te per volume of water in column, Q/VR, m3/(h*m 3)




number of aero tors in column air flow rate per diffuser length, Q/I, m3/ (h*m)

Technical characteristics Mass-exchange characteristics


Journal of the Australian Water Association





. .. ..... ..


. . .•.t


Tests in RLE were conducted in cap water in a column reactor by a desorption method according co che ASCE Standard (1992). Deoxygenation of water was carried out by purging techn ically pure nitrogen through the aerator, fixed near the bottom of the column. C OO was measured with an oxygen measuring device YSI-57, whose sensors were placed at a depth of 0 .6m in che middle of the column. As can be seen from Table 1, the tests conditions in ISAH and RLE can be considered si milar for the following reasons . First, the equali ty of mixing intensities (Q/VR) and geometrical similarity (h/0), and consequently the hydrodynamic conditio ns of mass-transfer p rocess were equal. Secondly, equal air flow race per diffuser len gth (Q/1) was p rovided and therefore equal conditions of air dispersion - which influences the size of air bubbles and the distribution of bu bble diameters the major factors determining mass-transfer efficiency. The results of ISAH and RLE tests presented in Figure 6 agree well. le was established that unlike the majority of modern fine bubble aerators the designed tubular aerators have a constant SOTE mode within a wide range of airflow rate . This confirms the theoretical results.

Airflow rate per 1 m of aeration system An airflow rate within 14 - 20 m 3 /h is recommended for the aerators. H owever these races can d iffer slightly depending on tank design, configuration of aeration system, worki ng d epth , temperature of wastewater, etc. A one metre long cubular aerator aerates approximately 3 m 2 of air tank.

Hydraulic characteristics The pressure losses on aerators d epend on the air flow race. The pressure losses on the

refereed paper

designed aerators va ry from 1.5 co 3.0 kPa. This value is for a new aerator immersed in pure water.

people. T hey are used co treat 25.000.000 cubic meters of wastewater a day.

While in service, the resistance of the dispersing element grows (which is typ ical for all porous aerators). The suspended substances penetrate into a porous layer and partially increase its resistance. The design of the tubular aerators allows minimisation of the Figure 7. A typical tubular aeration system. influence of these processes. T he blockage of the pores usually leads co the increase of pressure losses, increase of co ncentration of disso lved however in the designed aerators it results in oxygen from 0.5 co 1.5-2.0 mg/I; the time the red uction of air bubble diameter and, of solid particles retention decreased co 18 therefore, increase of oxygen transfer rate. days; and the percentage of the removed Hence, the increase of pressure losses is organ ic substances increased up co 56%. compe nsated by red uction of the required The aeration system provided good air flow an d the operation cost does not agitation of sediments, and co nsiderably increase. improved the perfo rmance of the entire facility. The cost of electric power was Case Studies and Economic Benefits reduced by 34%. The possibili ty of assembling long beams of Conclusions aeration units reduces ai r piping network T heoretical and experimental studies have insrallarion co a min imum, requires a enabled the formula tion of the basic smaller quantity of air intake pipes and principles fo r a new generation of aerators: reduces the quantity of stop valves and fittings (see Figure 7) and assembly takes • a tubular des ign which combines air pipe mini mal time. and diffuser; For example, in a regional wastewater facili ty in O hio, USA, an 8000 111 3 air tank was equipped with a tubular aeration system with in rwo days. Ir rook 10 days co re-equip 2 air ranks (19,000 111 3 each) with rhe tubu lar aerators in Kharkov and Kiev, Ukrai ne. For co mparison, re-equipment of 3 aeration basins (3,800 m3 each) with ceramic disc diffusers req uired 4 labour days per tank, which is as twice as long in relation co the volume. Usually, after retrofit of co nventional aerato rs by the tubular system , one in three operating blowers can be discon nected witho ut deterioration of water quality. In some cases, when rhe preceding aeration system consisted of perforated pipes, the installation of the tubular aerators allowed savings of up to 50% of electric power costs.

• aerators with a gap between construction casing and diffuser element; • diffusers made of porous polymeric materials with a polydisperse pore structure; T hus a new tubul ar polymeric aera tor has been developed and parented. (Galich, 1999, Meshengisser, 1996). The aerators are highly efficient and work in a wide range of air flow rates. Since 1999 the tubular aerators have been put into operation in 15 countries around rhe world and currently serve 120 million

The Authors Rostyslav Galich is Vice-President of Ecopolymer, Academician of the Ukrainian Engineering Academy, coauthor of many international parents of aeration systems, and has managed more than 120 national and international projects. Yuriy Meshengisser is President of Ecopolymer, Academician of the Ukrainian Engineering Academy, created theoretical basis of tubular aerato rs investigations, holds many international patents of aeration systems, realised more than 200 aeration projects around the wo rld. Victor Los is Dep uty Chief of Process Eq uipment Division and Quality Manager of Ecopolymer. Pavel Spiridonov is T echnical Directo r of lnnovEco Australia which rep resents Ecopolymer in Australia. Email: concact@innoveco.com.au

References ASCE ( I 992) Standard : Measurement of Oxygen Transfer in Clean Wate1; Second Edi tion. Am erican Society of Civil Engineers, ISAH (2003), Institute of Water Quality and

Waste Management ofHannover University Test Report on Oxygen Mass Transfer Characteristics H annover, 52p. Galich R.A. et al. 1999 Aerating Device, US Patent 5.868.972. Meshengisscr J.M et al 1996 Aerating Device., US Patenr 5.560.875. Meshengisser Yu M, Marchenko Yu G, 2000.

Modelling ofmass-transfer process in water aeration, Vodosnabjeni San . T ekhnika, No 6, Meshengisser Yu.M .. 2002 Dynamic model of gas bubbles fo rmation in barborage through a liquid. Chemical Technology, 12, p. 39-42.

Another example was published in Operations Forum, No I, January 1998 . In 1996 an aerobic mineraliser at a wastewater treatment plant in Ohio, USA, was re-equipped with rhe designed aerators. T he reconstruction resulted in Journal of the Australian Water Association



Ranges east of Melbourne. The 90km 2 region contains 700 0 properties to be sewered as part of the project.

PRESSURE SEWER PROGRAM AT YARRA VALLEY WATER By Paul Edwards Yarra Valley Water (YVW) provides water and sewerage services to 1.5 million people in the Yarra River catchment of Melbourne Australia. YVW casked Dalton Consulting Engi neers (DCE) w ith developing fu nctional designs for its pressu re sewerage p rogram. D CE developed a 'world first' dynamic pressu re sewer model using Info Works CS to optimise system design and analyse critical hydraulic conditions likely to occu r in a pressure sewerage system.

Pressure sewers

A pressure sewer system (PSS) is a network of fu lly sealed pipes, fed by a grinder pump located ac each property. The grinder pump red uces any solids into a slurry, which is then pum ped th rough a small diameter p ipeline from the house to che reticulation p ressure sewer in the street. The pipes are sized to accomm odate a li mited nu m ber of pu m ps o perati ng ac one time. As a result the pipelines needed fo r pressu re sewers are considerably smaller than chose required for crad icional gravity sewers, and pipe gradient 1s n ot a concern. Pressure systems are suited to locations where the terrain is rocky or steep or, conversely, very flar. As these sewers are water-eight to a level nor often achieved in gravity sewer systems, they are also effect ive fo r locatio ns where in fl ow and in fi ltration are an issue - for instance, areas where there are high gro undwater levels, or that are fl ood -prone. Pressure sewers are effective in coastal locatio ns, where emergency overflows could cause environmental damage, and areas w here ic is not practical to install a conven tional system. As the majority of the system mains are located in the road reserve, pressu re sewers are well suited to provide a retrofi tted reticulation sewerage system in areas with fai ling on-site treatment faci li ties (such as septic tanks).

Water Business aims to keep readers alert to business news and new product releases within rhe water sector. Media releases should be emailed co Brian Raulr ar b rian. raulr@halledir.com.au or Tel (03)

8534 5014. AWA wishes co advise readers rhar Water Business information is supplied by thi rd parries and as such, A WA is not responsible fo r che accuracy, o r otherwise, of the information submitted.

Investigations in the US DCE has been involved in all aspects of PSS design over the past three years, fro m feasib ility studies to derailed design and const ruction management. Its p revious independent investigations into p ressure sewer systems in the US - in Hawaii, H ancock Cou nty, Mississ ippi, and Lake Patoka, Ind iana - indicated char there were no standard arrangements fo r p ump ownership, siting or maintenance responsibil ities. T here was also m inimal concern expressed by system operators involving ongoing maintenance. DCE's studies also fo und that pu mps discharging to these systems operated, on average, less than 20 min utes a day and chat system flows were less than 50% of those of a similar conventional sewerage system in a high groundwater ingress area. There was agreement by system operators that a dynamic model would enhance understanding and future control of the system. In the US, design of the p ressure systems is normally undertaken by rhe pump manufacturer using either a p robab ili ty or rational method to determine peak fl ows and static modell ing to determine pipe sizing.

Yarra Valley model The p ressure sewer model developed by DCE for Yarra Valley Water's backlog sewerage program covered the Emerald Cockatoo area, located in the Dandenong

The area is d istinguished by its dramatic topography, which is covered with mountain ash trees and rem nant pockets of temperate rainforest that had to be p rotected d uring con struction. Each p roperty was modelled as an individual sub-catchment d isch arging into its own grinder pump wet wel l. The wet well was modelled, and the manufactu rer's p um p curve linked each p roperty to the reticulation p ressure sewer. Sanitary flows of 700 litres p er day were assigned co each p roperty, and the model was run for a m ini mum of 10 to 14 days to assess the lo ng-term syn chrono us operation of the system. The model was then used to refine the design of the pressure reticulatio n network by analysing the pressures and velocities with in the system so ch at opti mal p ipe sizes could be chosen to reduce retenti on times and o btain self-cleansing velocities. The model was also used to analyse critical hydraulic conditions fo llowing a power outage enabling the impact on downstream gravity sewers and pumping stations to b e analysed and emergency storage volumes to be accurately calculated. DCE's modelli ng and desig n process using Info W orks, was verified with data fro m an existing pressu re sewer system . South East Water provided network and operational data for the town ofTooradin which has a p ressure sewerage system serving over 230 p roperties. A model was b uilt of the exist ing system and verified against fl ow d ata from the inlet works of the wastewater treatment works. A go od correlation was observed between flows p redicted by the model and those recorded at the inlet works.

Conclusions T he use of a dynamic model allowed op timal pipe sizes to be selected with confidence an d system pressures/velocities to be monitored and analysed. Other benefits included: • rhe interact io n between the p ressu re system and the receiving gravity system

WaterGEMSÂŽV8 MULTI-PLATFORM WATER MODELLING Bentley WaterGEMS VS brings the only platform-independent solution to the world of water distribution modelling, allowing engineers to build, analyse, and map water distribution models without being anchored to a specific engineering design or geospatial platform. For more information, see the inside front cover of the September issue of Water Journal, visit www.bentley.com/A WA, e-mail anz.marketing@bentley.com, or call +61(0)39699 8699.

76 SEPTEMBER 2006 Water

Journal of the Australian Water Association