GUARANTEED With over 70 years experience in the water treatment industry and backed by the world w ide resources of the Thames Water Group, you can rely on Thames Water Asia/Pacific to deliver a w ide range of project objectives. We provide total design and construction services and guarantee t he su itability of our processes and the performance of our projects. In fact, you cou ld say we offer a wat er t ight guara ntee. Thames Water Asia/ Pacific provides comprehensive water and wastewater treatment and management services in the Asia-Pacific region. A key area of our expertise is the design and construction of municipal water and w astewater treatment facilities, both within Australia and overseas. We also have extensive knowledge in water and wastewater for the petrochemical, mining and metal, power and pulp and paper industries.
For further information contact: Thames Water Asia/Pacific Pty. Ltd. 28 Clayton Rd, Clayton North Victoria 3168 tel +61 -3-9239 41 00 fax +61-3- 9239 4200 ema il: thameswater@thameswater.com.au website: www.thameswater.com.au
WATER MANAGEMENT SOLUTIONS
Volume 27 No 2 March/April 2000 Jou rnal Australian Wate r Association
Editorial Board F R Bishop, Chairman l3 N Ande,~on. D Deere. I' Drnaym, W J Dulfer. G Finlayson, GA Ho lder, M Kirk, M M u ntisov, N Orr, P Nadebau m. J D Parker, M Pascoe, A J Priestley, J R issman . F R o ddick. E A Swinton •, Water is a refereed j o urnal. T h is sym bol indicates t hat a paper has been refereed.
CONTENTS
Submissions Su bmissions sho uld be made to E A (13ob) Swinton. Features Editor (sec bdow for details).
General Editor Peter Stirling PO 13ox 8.\, Hampton Vic 3 188 Tel (03) 9555 7377 Fax (03) 9555 7599
From the Federal President ...... .......................... ... ................................... ..... 2 From the Executive Director ................. ..... ... .................................. .... .... .... .. 7
Features Editor
MY
E A (Bob) Swin ton 4 Pleasant View C res, Wheclm Hill Vic 3 150 Tel/Fax (03) 9560 4752 Email: swintonb@c03 I .aonc.neuu
Branch Correspondents ACT - Ian Bergman Tel (02) 6230 I 039 Fax (02) 6230 6 265 New SoutJ1 \Vales - Leonie H uxedurp Tel (02) 9895 5927 Fax (02) 989 5 5967 Northern Territory - Mike Lawton Te l (08) 892.\ 64 11 Fax (0 8) 8924 64 I0 Queensland - Tom Belgrove T cl (07) 38 IO 7967 Fax (07) 38 10 7964 South Australia - Angela Colliver T cl (08) 8227 I I I I Fax (08) 8227 I I00 Tasmania - E d K leywcgt T el (03) 6238 284 1 Fax (03) 6234 7 I 09 Victoria - Mike Muntisov Tel (03) 9278 2200 Fax (03) 9600 1300 Western Australia - Jane O liver Te l (08) 9380 7454 Fax (08) 9388 1908
Water Advertising & Production Hallmark Editions PO l3ox 84, Hampton, Vic 3188 Suite I, 350 Sou th lload, M oorabbin Te l (03) 95 55 7377 Fax (03) 9555 7599 Email: halln1ark@h allcdit.co111.au
Advertising coordination: Fio na Second Graphic design: Mitz i Mann
Water (ISSN 0310 · 0367) is publish ed in J anuary, March, May, July, Sep tem ber an d N ovc1n bcr.
Australian Water Association Inc AltBN 0 5.\ 2 53 066
Federal President Alle n Gale
Executive Director
AWA Iii...~~
~
AUSTRALIAN WATER ASSOCIATION
C hris Davis Ausn~lian Water Asso ciation (A WA) assumes no responsibility for opinio ns o r scatcnients of fa cts expressed by co,mibutors or advertisers. Editorials do not necessarily represent official AW A policy. Advertisements arc included as an infon m cio n service to readers and arc reviewed before publication to ensure relevance to the water environmen t and objectives of AW A. All material in Water is copyright and sho uld not be reproduced wholly or in part without the written pen n ission of the General Editor.
Subscriptions
POINT
OF
VIEW
A Public Health Issue, or a Public Issue? .. .. ............... ...... ..... ........ ...... ..... ... 4
J La ngfo rd WATER
QUALITY
AND
TREATMENT
Overview of the CRC for Water Quality and Treatment ................................... 11 D Bursill and T Pri estl ey Disinfection Byproducts and Reproductive Effects: Where to from here? ............................................................................................................................................ 16 M Sinclair Blue-Green Algal Toxins and their Relationship to Cancer ... ..................... 18 I Falcon er, A H u m page, E M oore, S Froscio Mixing It Up with Blue-Green Algae ............................................................................ 21 M D Burch , JD B rookes, P T arran t, P D c llave rde NOM Treatability and Catchment Characteristics ....... ........................................ 25 KM Spark Tastes, Odours and Algal Toxins, Which PAC is Best? .................................... 28 D Cook, G N e wcom be, J M orrison Biofilms - A Sticky Situation for Drinking Water? ............................................... 33 I Fisher, M Angles, J C handy, P C ox, M Warn ecke, G I( asci, V J egacheesan WASTEWATER , Indicator Organism Levels in Effluent from Queensland Coastal STPs .. ... . .. . .. . .. ... . .. . .. . .. . ... ... ... ... ... .. . .. . .. ... . .. . .. . .. . .. . .. ... ... ... .. .. ... ... .. ... . 38 RJ T ho mas, EA Gardn er, GA Be rry, H N C hi ni vasagam , PE G ree n, AV Kl ie ve , PJ Blackall , GW Blight, BJ Blaney BUSINESS ·, Managing Water Quality Risks from Catchment to Tap ........ .. ...... .... 46 R Ban nister, N O ' Conno r, C Sti ve rs ENVIRONMENT ·, Energy Efficiency - A Management Approach ..... .............. ... .. ..... ... ...... 51 A M cC leery, G W e iss DEPARTMENTS Aquaphemera . .. ....... .. .. .. ..... .. .. .. .. .. ....... .. ....... .. .. ..... .. .. .. .. ......... .. .. .. .. .. .. ............... . 7 International Affiliates .... .. ........ ....... .. ........... ... ... ............ .. ...... .............. .... ..... . 8 Membership .............. .. ... ..... ..... ..... ... ...... ... ... ..................... .. ... ..... ..................... . 55 Meetings . .. .. .. .. ......... .. .. ..... .. .. .. .. ......... .. ....... .. ....... .. .. .. ... ...... .. .. .. .. .. ..... .. ..... . ... . .. . 56
J Vnlcr is sent Lo all AW A members six times a year. It is also available via subscri ption for S50 a year.
Visit the Australlan Water
Auoclatlon
HOME PAGE
and access our calendar. bookshop, membership fonns and 50 pages of Information at
OUR COVER: Tl,e quality ~f tl,e 111ater supplied ro 011 r pop11/atio11 is steadily bei11g i111pro11ed, bot/1 by be/fer 111miage111e11t as well as thro11g!, beller 1111dersta11di11,rz of the 1111derlyi11,Q scierrce a11d tec/1110/ogy. This iss11ef eat11res tl,e 111ork c!f tl,e Coopcrath;e Research Ce1t1re f or ltVa tcr Q11ality a11 rl Treat111e11t, 111/,ich has added a11orher di111e11sio11 lo researcl, , by 111easmi11g tl,e act11al i111pact (of 111ater q11ality) 011 public /,ca/ti, . (Photo co11rtesy of tl,e C R C WQT).
FROM
THE
PRESIDENT
A CHALLENGE TO SENIOR MANAGERS AW A has a long and proud history, starting in 1962, when a diverse group of professio nals with a shared interest in water decided th at they wanted a mu ltid isc iplinary assoc iatio n to serve their commo n interest u nde r o ne ro of. At that tim e, oth er professio nal so cieties were a bit stuffy abo u t allowing in 'foreign ' pro fess ionals, so our fo unding fa th ers (most of them were men, l have to confess) were ah ead o f their tim e . T he ir foresight has proved to be most successfu l, as A WA has grow n to be the peak body in Australia's water in dustry. Where o ur asso ciation has no t been uni versally successfu l, th ough , is in the area of engaging top m anageme nt. In the early yea rs AW A had m an y se nior executives, fro m both the pub lic and priva te sec tors, takin g very active roles in d eve loping th e orga nizati o n. Fo r example, m ost of our early Presid ents were leaders of o rgan izati ons. W ith the changes in th e wate r in du stry over the last few years, the direct in puts from top management appea r to have wa ned. T his is understandable in view of the additio nal pressures on seni or managem ent in this ra pidly changing indu stry. On e o f my ambitio ns whil e in office (and everyon e wi ll be aware o f it by the tim e I'm finish ed!) is to get the industry's leaders involved m o re than th ey are at prese nt. H o w can we clai m to represent the total water industry if th e leaders of our industry do not have an involvem ent' In ge neral l do no t believe thi s can b e by r eprese n t at io n on committees, w hich has been th e means in the past (although l wo ul d never discourage anyone from w o rking fo r their organ ization in th is manner). AW A o ffers so methi ng fo r all p eople involved in the water industry: rangin g from scho ol age activities (the C D R OM, the Sto ckholm Ju nior Wa ter P rize), to people recen tly in the industry (Summer Scho ol) and through o u r mainstrea m conferences, Ozwater and th e lo cal bra nch activities . Perhaps we have ten ded to lean too m uch towa rds technolo gy , w h ich to p managers ha ve had to leave behind them, as they fo cus o n th e business and regul atory climate fo r their w ater organisations. Increasingly, the Internet is in terfacing between us, our m em bers and the industry stakeholders gen erally. W e 've hit a resonant note in th e weekly e- m ail news , which now has a fo llowing of 2
WATER MARCH / APRI L 2000
Allen Gale
over 1,700 direct recipi ents and perhaps a few tim es that on forwa rded copies (emai l news@awa.asn.au to get o n the list). C learly, th is mediu m is relevant in the current climate o f intense pressure at work and the need for bi te-sized chunks of in fo rmatio n that ca n be absorbed and expanded on if necessary . I think we n eed a sim ilar news veh icl e, bu t aimed at C EOs, for th eir m ilieu o f strategic issues, long range trends and immediate threats and o ppo rtunities. Also, why ca n ' t we provide a mea ns fo r all top management thro u ghou t the industry to netwo rk with their counterparts in o ther sec tors of th e ind ustry in bo th th e pu blic and the pri vate secto r;, Th e differences between pu bli c and pri vate are ever dec reasi ng and both ca n ben efi t by closer com mu nicatio ns and transfe r of in fo rm at io n on the d ri vi n g fo rces beh ind success. l plan to have o u r m arketing staff condu ct a survey am o ngst top m anagem ent to determine d irectly fr om them what it is they need to kn ow and how AW A can b est service th eir organisations - o ur corp orate m em bers. It wou ld be excellent to have som e spontaneous id eas em erging from CEOs directly short-circu iting the survey process . O f everyon e in the fi eld , C EOs probably ru n th e greatest risk of in for mation o verload , so we can help th ere, with filter steps to cu t o ut the extraneous no ise and to p ass on j ust the grist o r the good oil. Th is info rmat io n d isse m i na ti on fu nction is all the more important in our
u niqu e Australi an co ntext, w here the water industry resides w ith the states, w hich bu ilds artifi cial w alls betwee n jurisdictions. A few lu cky people straddle the bou ndari es and keep in touch , bu t m ost don 't and they can be in the dark a b o ut so m e t h in g i m p o rta n t happ ening in another state or territo ry. Th e o ther bo undary is between the priva te secto r and the public. W hile there has been a lot o f h ype in the 1n edia about priva tisa tion, it is no t happenin g to an y great exten t in th e w ate r industry and th ere is still a lot o f m isunderstanding between the two worlds. I don 't believe that govern m ent p eople u nd erstand busin ess very well (that's a generalisatio n, but it see ms to hold true . I ha ve cro ssed the divide and have seen both sides now) . I susp ect too that private sector p eopl e are no t full y aware of the for ces th at can p rop el a state agency in unexpected directions. AWA can help in all these areas, w ith for mal an d informal ways o f developing b etter comm u n icatio ns and alerting everyone to cha nges, trends and impedim en ts. W e are reso u rcin g our natio nal secretariat better no w, so expect to see staff mo re frequ ently , and to hear from o ur vo lu ntee r lead ers too, as they play a m o re active role in probing m emb ers' needs and ge ttin g th ose key, sen ior peopl e into th e loop . So , my chall enge to top man agem ent is: help us to help yo u - initially by a positive response to ou r su rvey to know how AW A ca n serve you better. Our orga nizatio n, and the water industry can only be stron ger from positi ve contrib u tions by top managem ent.
A llen Gale
EFFLUENT IRRIGATION The September issue of Water will focus on this topical subject. Submissions are invited from practitioners, equipment suppliers, authoritie s and government agencies. We would like to hear of both successes and difficulties in this area. Please contact the Features Editor , Bob Swinton (see page 1 for contact details).
In t he year 2000, the C R.C fo r Water Qua lity and T reatment (CR.C WQT) w ill co mple te its fifth yea r of operation as an unin corporated j oint venture o f research and ind ustry partners committed to the p rovisio n o f hi gh qua li ty w ate r t o c o n su m e rs a t a n afford abl e p r ice . M e mbe rshi p has risen from an initial group of 17 to a curre nt total o f 23, w ith all secto rs o f the industry being represented, while a vibra nt researc h program is beginni ng to produ ce signi fi cant resu lts. It's initial budge t fo r the seve n years to ta lled more th an M $55, of whi ch currently some M $36 has been spent. Some 40 indi vidual proj ects are eith e r in progress, comple ted, or just beginnin g . l n 2000, th e C R C WQT has two major additiona l challenges to deal w ith . le wilJ have its fi fth year review under the Co m mo nw ealth 's accountability requi rem e nts fo r the C oope rati ve R.esea rch Centres Program . In addi tion , it will need to submit a re-bid application to ensure that it continues in th e C R C Program beyond its curre nt seven year term. As the C R C WQT approaches th ese two significant activities, it is interesting to refl ect on its achieve ments and on the changing industry environment over the past fo ur or fi ve years. Th e C R C WQT ha s on e n o ve l program , a public health and risk assess-
Director, Professor Don Bursill
Deputy Director, Dr Tony Priest ley
menc program w hi ch has a number o f proj ects looking at the fu ndamental issues associated w ith health o utcomes from drinking w ater, in addition to the three
~
-
_)
(R\ fo r \X1,ttcr Q 11,1li1y ,rnd Treatment
programs of research chat cover water technology issues from catchment to cap . For the first tim e in Australia, medical researche rs have cooperated directl y w ith w ate r tec hno logists o n wate r q ua li ty problems. T h is cooperatio n is having a very benefi cial impact on the progressio n of t h e Au stralia n Dri nk in g W a t e r G u idelines and on industry appreciation o f the C R C W Q T - but more of that lacer. Other im portant aspects include a very w ort h w hil e e d u cati o n a nd t ra in i ng program , the core o f w hich is the curre nt list of 36 PhD students, and an expand ing technology transfer program , w hich is atte m pting to co m mun icate the rapidly in creasing output o f know ledge and in fo rm ation eman ati ng fro m the research program . It is hoped chat the gradua ting Ph D st udents w ill play a future key role in chis kn owledge transfer process, as they take up em ploymenc in a progressive and growing industry . le is t he intent ofche C R C WQT co be a truly national R &D orga nisati on , w ith a broad representation fro m aro und the country. In chis light, it has been encouraging to see mem bershi p grow from 17 to 23 over th e lase fi ve years. Also, in an attempt to broade n coverage to include the needs of sma lle r, region al w ater authorities, an Associates Program was established in 1999. Mem bership of this program has grown rapidly to seven in as li ttle as nine m onths, acco mpanied by a WATER MARCH/ APRIL 2000
11
WATER
QUALITY
From the catchment . ..
series of workshops held throughout the country, from Townsville to Perth. T hese wo r kshop s have p roved ex tre mel y valuable in clearly establishing both good communication li nkages and the major water quali ty issues in regional Australia. Although it is not the intention of this short article to outl in e all the specifi c achievements of the CR.C WQT , it is sufficient to mention a few significant outcomes that will impact o n the water
AND TREATMENT
. .. to the t ap
industry in Australia and elsewhere. They include: • The Water Quali ty Study. T his study seeks to establish whether or not the existing wate r qua li ty in M e lbourne co ntri bu tes in any way to e ndemic gastroenteritis levels in that community. We believe the study protocol that has been developed and used in th is project is the best that has been developed to date. The information gained is important for a
better understanding and management of the microbiological quality of public water supply systems and will b e very usefu l in refinin g the application of the relevant part of t h e Au stra lian D ri n k in g Wa t e r G uidelines. T wo overseas studies have indica ted that high qua lity filtered water supplies may sti ll contribute up to 30% of en demic gastroen teritis, however th e study protocols have not faired well in peer review. T he results of the Melbourn e
The ZElllocc Process Ior Wastewater Treatment The ZELflocc Process provides significant operational and environmental benefits including:
•
ZEL
• • •
• • • • •
Effective Nitrogen Removal Increases Plant Capacity Reduces Capital Expenditure Better Effluent Quality Operational Stability Reduction in Chemical Dosage Improves Sludge Characteristics Dramatic Reduction of Odours Municipal and Industrial Applications
ZEDLITE AUSTRAUA llMITED Phone: 03 9699 4599 Facsimile: 03 9699 4522 Website: www.zeolite.com.au Email: info@zeolite.com.au Level 3/ 100 Dorcas Street, South Melbourne, VIC. 3205
lEDllTE AUSTRAllA. ALEAD/NS EDSE TECHNDlDSY DEVElDPER
- -
-=-=-
-=~~~~=
ZEOLITE
AUS TR AL I A
WATER
QUALITY
AND
TREATMENT
CRC Director, Professor Don Bursill, presenting a plaque to the Chairman of the Townsville Thuringowa Water Supply Board, Mr Ian Hamilton, to mark their joining the CRC's Associates Program.
based study will be published in early
2000. • T he understand ing o f pu blic h ea lth outc o me s of algal tox in s and th e man agem e nt o f algal bloo ms has bee n ad va nce d conside rably. Th is k n o w ledge has contri bu ted to the re ce nt WHO publicati o n "Toxic C ya11obacteria in Water. A Guide to t/1eir Public H ealth Co11seque11ces, !Vlo11itori11g a11d 1vfa11age111e11t". • It has been established that th ree of the fo ur probl em species o f cya nobacteria (tox ic algae) have a genetic basis for toxin production and a co mple te sequ e nce of the mi crocysti n synthetase gene cluster has been determ.in ed . Th is will a!Jow th e id entificatio n of toxogenic strains in water suppl ies and provides a basis fo r the possible development of commercial mo lecular probes fo r this purpose. • A comprehensive set of studies o n water trea tment processes has added significantly to the knowledge of these techniques fo r improving water quality. So me of these outcomes are already improving the petformance ofsystems in th.is counny, both from a water quality point of view and from a cost and effi ciency perspective . Two examples of this progress include the development of technology to study the levels of polym ers used as flocculants and flo cculant aids that cany over in to the product water. T h.is is the fi rst time that this has been achieved. T he second example is the development of a low cost, convenient m ethod of regenerating granular acti vated carbon used in water treatment fo r the removal of algal toxins or taste and odour compounds. Agai n, this is a significant advance of inter-
national importance w hich is in the process of comm ercialisation. • A great deal of progress has been made in understanding the conditions that lead to
bio film grow th in wa ter d istribu tio n syste ms. lnfo1111ation developed in the laboratoty has been used to refi ne computer based models that are now being used to enhance the management of water disttibution fo r better water quality outcomes. A related developm e nt is th e acc ura te modelling of chloiine decay. C hlorination is still the most com monly used method of disinfec tion and the C R C WQT developments provide th e means to manage chloti nation to achieve better microbiological control w hile mi ni mising problems associated with inapprop1iate chlorine residuals. W he n the C R C WQT was established in 1995, th e industry was fu!Jy foc used on th e C OAG refo rm agen da and the Hillme r competition agenda . As a consequ en ce, at that time w ate r q u ality management issues were no t often a key feature o f the age ndas of boards o f management or sen ior executive tea m meetings. On e m ight have assum ed that restru cturing and economic reform were core business for the water industry and th at wa ter quality manageme nt was no t. Today, wate r q uality managem e nt issues see m to have assu med greate r impo rtan ce and the value of goo d scie n-
Super solutions f0r: the water industry Join many of Australia's leading utilities already enjoying •
t op level hassle free service
•
excellent investment performance
•
com petit ive fees (no commissions)
•
defined benefit & accumulat ion divisions
•
a full range of products and services
Call Brian Towers on
(03) 9248 5911
WATER MARCH/ APRIL 2000 CRC for Water Quality .ind Tn:.i tmcnt
13
WATER
QUALITY
tific and w ater tec hnology kn owledge is better appreciated. The Sydney water quality incident is clearly the main reason fo r this c hange, but there are other reasons as we ll. T he re is an e mergin g view among the leadership o f the w ater in dustry that th ere is a need to foc us less on inte rnal compe titio n and more on the com petitiven ess o f the industry in an international sense . This view requ ires a grea ter focus on areas w here the industry mu st pe rfo rm w e ll and have some co mpetitive adva ntage. Water quality is clearly one of those and understanding the technology in volved in de liveri ng th e service custo mers w ant prese nts opportun ities for the Australian wa te r in dustry. The C R.C WQT se nses a greate r level of appreciation for its skills among its industry partners and this is beli eved to be due, in part, to recogniti on of the ac hievements made since co m mencement. The new er, em erging visio n for t he industry also requires an inh ere ntly lon ger term strntegic vie w , w here w ell structured and targeted research and deve lopm e nt ca n pro vide signifi cant competitive advantage. As we move in to 2000 with a re-bid and a major review o n o ur plate, we w ill ta ke som e stre ngth from th e support and en couragement we arc receiving from the wa ter industry. The artic les and papers w hich fo llow in this feature have bee n selected from the fo ur R esearch Programs w hich are be ing operated by parties in Victo ria , NSW, W A and ACT, incl uding un iversiti es, wate r authoriti es, CS I R. O and co nsultanc ies.
AND TREATMENT
CRC PARTNERS ACTEW Corporation Australian Water Quality Centre Australian Water Services Australian Water Technologies CSIRO Egis Consulting Australia Melbo urne Water Corporati on Monash University Orica Australia RM IT University South Australian Wat er Corporation South East Water Sydney Water Corporati on The Australian National University The University of Ade laide The University of New South Wa les United Water International University of South Australia Victoria n Department of Human Services Victorian Department of Natural Reso urces and Environment Water Corporat ion of Weste rn Australia Water Services Association of Australia Yarra Valley Water
CRC ASSOCIATES Gippsland Water Gold Coast City Cou ncil Grampians Water Hunter Water Corporati on NSW Department of Public Works and Services South East Queensland Water Board Townsville Thuringowa Water Supply Board
Dr Dennis Steffensen addressing a Workshop on 'Protecting Water Supplies' hosted by the Sydney Catchment Authority.
14
WATER MARCH/APRIL 2000 CRC f()r W'at<:r Qua li ty :111tl Trc.:itmcm
FOR FAST, SIMPLE, EFFECTIVE RELIEF FROM STORMWATER POLLUTION, CALL ROCLA. f the slower, more expensive and less
The cost-saving of this is o utstanding to say
effective m eans of stormwater pollution
the least.
control give you a headache, take hea rt.
Feeling better now?
I
The Roda CleansAll'" Gross Pollutant Trap is the perfect long term solution to keeping pollutants out of our water ways. Designed to treat large Hows in-line or end-
To take the pain out of your stormwater pollution problems call Roe/a Pipeline Products on 131 004 or visit our informative web site at www.pipe.rocla.com.au
of-line, the Roda CleansAll'" GPT is suitable for new developments or fo r retrofitting to existing pipes and culverts. The Roda Clean sAll'" GPT is much easier to install than anything else and most models
RocJa Pipeline Products
can be installed in less than a day without the need for speciali sed installation crews.
THERE
IS
A
BETTER
Rocla Limited ACN 000 032 191. Trading as Rocla Pipeline Products. A member of the Amatek Group. TM and ÂŽ are trademarks of Rocla Limited.
FUTURE
I N
THE
PIPELINE
WATER QUALITY AND TREATMENT
DISINFECTION BYPRODUCTS AND REPRODUCTIVE EFFECTS: WHERE TO FROM HERE? M Sinclair Research Progm111 One, Public H ealth Assessrnent, has a number ef i111porta11t projects. This paper s1-1111rnarises a recent pilot st11dy. Abstract
D isinfection of drinking water with chlorine or cbJoram.i ne remains the most common m ethod of water treatment to protect public health. In the 1970s the discovery that disin fection produced a range of byproducts raised questions about the health etfects of such co mpounds, with attention being primarily focused on the possibility of cancer risks associated with long term exposure. M ore recently, the issue of reproductive efJccts has been raised w ith several studies reporting associations between adverse pregnancy outcomes and high DBP exposures. If these risks are real , the water industry will face a massive c hange in th e way t h at DBPs are monitored and regulated. Whi le the evidence is not convincing, more investigation is warranted because of the public health implicatio ns of these observations. In particular, th ere is a pressing need for the d evelop m en t of b et ter m et hods for exposure assessment. Keywords: c hl orin ation , disinfection byproducts, r e produ ct ive e ffec ts, pregnancy outcomes, exposure assessment Introduction
The introduction of di sinfec tion is recognised as the single greatest advance in the provision of safe dri nking water for human communities, and is credited with dramatic reductions in the prevalence of waterborne disease . Chlorinati on and chloramination remain the most w idely practised methods of disinfection because of their relatively low cost, simpl.icity and prove n effectiveness against common microbial pathogens. Ho weve r since th e discovery in the 1970s that chlorine reacts with natural organic matter in water to fo rm disinfection byproducts (DBPs), questions have been raised abou t the possible effect of these compounds on health. M ost research has foc used on trihalomethanes (THMs), the m ost abundant and easily measured class of volatile DBPs in chlorinated surface waters, and the first to be characterised.
16
Over J35 individual DBPs have since been identified and grouped into 35 classes according to their stru ctural and chemical properties (eg volatile vs non-volatile, polar vs non-polar). Many remain rather poorly characterised and are difficult to measure in water supplies. Th e o bse rvatio n that c h loroform (generally the predominant THM) was associated with an increased rate of some cancers in rats and mice drew attention to the prospect of cancer risks in humans, and led to a number of epidem.iological studies investigating this possibility. Ironically it was later w idely recognised that the methodology of the rodent cancer tests was flawed, but the issue of cancer 1isk contin ues to be an area of active research and controversy, and a major influence on drinking water regulation. More recently, concerns have been raised about the possibility of adverse reproductive effects associated w ith exposure to DBPs. Sin ce 1992, six studies from the US, one fro m Canada and one from Norway have been published in this area. The range of reproductive outcomes examined has bee n dive rse, including sponta neous abortion, premature delivery, low birth weight, stillbirth and birth defects. In some studies an apparent association was fo und between exposure to DBPs and increased risk of adverse reproductive outcomes. Implications for the water industry
The possibility that exposure to DBPs may increase risks of adverse reproductive o utcomes is an important and highly
WATER MARCH/APRIL 2000 CRC for Water Qual;ry
.ind Tn:,11mem
em otive issue. ff health risks in th.is area can be substantiated then a massive change will be requi red in the way that DBPs in drinking water are monitored and regulated. Focus will shi ft from the long time frames of exposure related to cancer risk assessment to much sho rter time frames and day to day peak exposures. It will become necessary for water utilities to m easure and control DBP levels on a continuous basis through entire distribution systems, rather than conducting quarterly measurements at a few sites. It is even possible that this issue may ultimately fo rce the abandonment of c hlorinati o n and c hloram in at ion as m ethods of disinfection. How strong is the evidence?
Exam.ination of the evidence to date shows that in most cases the observed increases in risk were smaLl and all of the studies had significant weakn esses in m ethodology w hich limit the interpretation of their findings. Seven of the eight of the studies were retrospective - m eaning that th e iden tifi catio n o f p regnancy outcomes relied o n the use of data sources such as hospital records, birth registries and m edical practitio ners. Thus the strength o f the analysis depends on th e comprehensiveness and precision of these data sources in recording the pregnancy o utcom e(s) under study. T his is less desirable than a prospective design where women are identified early in pregnancy and followed until the outcome is known. Of more significance is the lack of accuracy in the methods used to classify exposure to DBPs. The exposure levels of pregnant women have been estimated on the basis of residential address and only two retrospective studies assessed individual water consumption. Furthermore, DBP levels in water have been estimated from water utility sampling programs (usually at the treatment p lant) rath er than by measurements at the consumers taps. It is known that the reactions which result in DBP formation continue to occur as water moves through the distribution system away from the chlotinatio n point, with
WATER
som e compou nds con tinuing to fo rm w hile others decompose and decrease in concentration over time and distance. Significant chan ges may also occur due to seasonal variations in individual raw water supplies and blending of water sources for operatio nal reaso ns. Th us, in fre que nt measurement ofDB P levels at on e or a few points in a distributio n system w ill not adequately represent the concentrations chat are delivered co individ ual consumers. In some studies DBPs w ere not actually measured - instead levels of exposure were infetTed from the water source (ground vs surface), the type of treatment (chlmi nated/ chloram.inated vs untreated) or water colour (as a measure of dissolved organic carbon). M ost studies have not considered other sources of exp osure (eg dermal absorption of volatile DBPs during showering/bathing or swimming) or other important influ ences on pregnancy outcome (eg histoty of prior pregnancies, maternal smoking, diet, alcohol consum pt ion , soc ioe cono mi c level) . Overall, these li mitations mean that the results of these studies can only be regarded as inconclusive. T he single prospective study conducted to da te in ves tiga ted th e re latio n ship betw een spontaneou s abortio n and drinking w ater in th ree regions of C aliforni a. T he resul ts o f th is stu dy were published in tw o papers: the first examining the amo unt and source of drin king water, an d the seco nd examini ng total TH M levels (Swa n et al. 1998, Waller et al. 1998). The prospecti ve design ove rcame ma ny of the shortcomin gs o f retrospective studies (eg verifi catio n of pregnancy outcomes, assessme nt of smoking, water consumption) but exposure assessment was still limi ted. Total T H M exposure (TTHM) w as estimated from quarterly readin gs or an nual ave rages re p o rted by w ater u tilities, w ith no allo w ance made for variations w ithin th e distribution syste m . T he results o f the first anal ysis suggested tha t high consumption of ground water in one regio n was assoc ia ted w ith an increased risk of spontaneous abortion , however this e ffect was not related to wa ter chl orin ati on a nd t h e a uth o rs co ncluded that an unidentified substance in the raw ground water was the most li kely cause . The second analysis showed an increased risk o f spo ntaneous abortio n in wom en w ith high personal exp osure to T T H M s (classifi ed as 5 or more glasses per day of cold water w ith >75 ni.icrograms / L). For women in th is category (onl y 2% of wome n in the study), the risk of miscarriage was about 1.8 times that fo r w o men d rin king less wate r w ith high T THM leve ls, or w ater w ith lower TTH M levels.
QUALITY
AND
TREATMENT
H owever the risk was not affected by swimmin g o r pr olo n ge d sho we rin g (although both would be expected co in crease T T H M exposure), or w he n adjustm en t was made for smo king or alcohol consumption (bo th sign ificant 1i sk factors fo r spontaneous abortion). T hese discrepancies, coupled w ith th e substantial sho rtco m ings in individual exposure assessment, throw considerable doubt on the findi ngs w hich implicate T H M s.
ical half li fe of TCAA was measured by providing participants w ith bottled w ater (which docs not contain T CAA) fo r a period o f two weeks and tracking the declini ng levels in their mine. Finally, the rate of increase of T CAA concentration was estimated afte r participants resumed drinking their usual tap w ater. Similar studies w il.l be undertaken in one or more loca tions in Canada to provide a larger data set fo r analysis. As an adj unct to direct human measurements, a proj ect is also be ing undertaken co c harac te rise th e relatio nships bet w een maj or classes o f DB Ps in distribution system s. Existing datasets on DBP occurre nce i n a nu mber of distribution systems in Australia and Canada are being collated , w ith th e aim of developing a mode l describing the relationship between maj or classes of DBP. T his w ill provide an additio nal mea ns to es tim ate h um an exposure, particularly w here biological measurem ent o f an excreted D BP o r metaboLte is not possible. T he ultimate aim is to develop a suite of m eth ods w hich ca n be e mplo yed by epidemiologists to generate accurate individual exposure indices, which take into account the complexities of both water distribution systems and water usage patterns. lt is envisaged that data from a range of biological samples, and water consumption/ water use questionnaires will be combined w ith wa te r sa mplin g, d istr ibuti o n sys te m modelling and verification samples to provide a com posite picture of ind.ividual exposure. O nly when such tools are available will we be in a position to undcttake epidemiological stud.ies of improved design to help resolve the contentious issue of reproductive health effects.
Where to from here? In contrast to cancer risk assessment w hich requires estimation of all relevant exposures o ver several decades, the short tim e frame in w hich reproductive e ffects occur means that it is feasible co study these effects in a prospective manner. H ow ever it is clear that for future epidemiological studies to provide any valuable information concerni ng these health risks, th ey need to incorporate better exposure assessment at the indi vidual level, w hich takes into acco unt the diverse chemical nature of DBP s, the complexity o f water distributio n systems and va riatio ns in individual water consumpti on and water use patterns. As part of its Public H ealth R isk Assessme nt program , the CR C fo r W ater Quality an d T reatment is wo rking in co!Jaboration w ith researche rs from the Environmental H ealth Sciences Program at the U niversity of Alberta on a series of proj ects to develop better w ays of assessing huma n exposure co DB Ps. T he first proj ect to be undertaken was a smaU pilot study in Adelaide to investigate the m easure ment of urinary concentrations of trichloroaceti c acid (T C AA) as a marker of exposure. T C AA is on e o f several haloacetic acids fo rmed during chlorinatio n , and these compounds are generally the m ost abundant of the nonReferences vo latile DB Ps in water suppl ies. The levels Swan , S. 1-1., K. W alle r, et al. (1998) . A Prospective Study of Spontaneous Abortion: of haloacetic acids in Adelaide tap w ater ls.elation to Amo unt and Source of D1i nking are readi ly detectable by standard assay Water C ons um ed in Earl y Pregna ncy . m ethods, and provide a natural exposure Epidemiology 9(2): 126-33. setting fo r investigatio n. Wa ll er, K. , S . 1-1 . Swan, et al. (1998) . R ece n tly p ubli shed wo rk suggests T rihalomethanes in Drin king Water an d uri naty T C AA excretion shows a good Spontaneous Abortion . Epidemiology 9(2): 134--W. level of correlation with T C AA intake when correction is made fo r factors suc h as Weisel, C. P. , 1-1. Kim, et al. ( 1999). Exposure estimates to disinfoction by-products of chloriheating of beverages prior to consumption nated drinking water. Environ Health Perspecr (W eisel et al. 1999). H owever in this study I 07(2): I 03- 11 0 . only female subjects were tested, and only two urine samples w ere measured for each Author Dr Martha Sinclair is a senior research subj ect. In the Adelaide study both male fe llow in the D epartm ent of Epide miology and fe male participants w ere included, and a series of urine samples fro m each person a nd P reven ti ve M e di c i ne, M o n as h w ere tested to establish the range of day to Univ e rsit y M e di ca l Sc h o ol , Alfr ed day variability. Participants kept a fl uid H ospi tal, Pra hran VIC 3 181. She w orks in th e P ubli c H ea lth Ri sk Assess m e nt intake dia1y to allow esti mation of boiled and unboiled tap w ater intake. The biolog- program of the C R.C for W ater Q uality and Treatm ent, and is editor o f the Healt/1 Stream newsletter. E- mail martha.sinclair@ med.mo nash .edu .au WATER MARCH/ APRIL 2 0 0 0 CR\ for \'q;ucr Q11ali 1y .i nd Treatment
17
WATER QUALITY AND TREATMENT
BLUE-GREEN ALGAL TOXINS AND THEIR RELATIONSHIP TO CANCER I Falconer, A Humpage, E Moore, S Froscio Introduction
As we li ve to o lder ages th e cha nce of o ur d eve loping cance r increases and it is li ke ly that a substantial proportion of these cancers are ca used by factors in our e n vironment. A ir , fo od and wate r co ntinua lly enter and leave our bodies carrying ch e m icals that ca n initiate or stimulate cancer growth. Ultravio let light a nd ambie nt radioa c ti vi t y ca u se mutations and som e of these will result in potentia lly cancero us ce lls. Th ere are two main p rocesses invo lved in a cell developin g into a ca ncer, the first be ing the mutation o r defect in th e DNA that ca uses chan ges hav ing cancer fo rming potential and th e second the grow th o f these pre-cancerous ce lls into malignant can cer. T o explore wh e ther blu e-green algal toxin s in o ur envi ronment have any role in human can ce r is not easy but there are several quite dif:ferent approac hes that can be used. O ne is to swdy the rates of particu lar cancers in the popu la tio n and to correlate these with human exposure to blue-green algae o r the ir toxins. H ere the difficu lty is to de te rmin e the acwal exposure of people to blue-green algal toxins (P ilotto, e t al. , 1999). Th e presence o f algae in a water supply reservoir do es not mean that people are drinking the toxins. Th e water treatment may re move the toxins, or the taste and odour of the water may cause people to drink rain water or buy bottled water instead. Until a human ' marker' for algal or toxin exposure can be id enti fied, it wi ll be very d iffi cu lt to use ep id e miol og ic al approaches to answer these qu estions. Animal models-Microcystis toxins
For th e determ ination of safe 'guidelin e va lu es ' for drinking w ater , th e W H O and similar age ncies use experim ental data fo r toxicity and carcinogenicity in an imals and apply safety (also called uncertainty) facto rs. For c hemical toxicants t hat do not cause cancer the maximum experimental dose that does not cause an ill effec t after continued consumption is the basis fo r the calc ula tion (Falcone r et al. , 1999).
18
Pictured are (from left to right) Emma Moore, Suzanne Frascio, Prof Ian Falconer and Dr Andrew Humpage.
An ot her app roach is to use animal experi m ents to detect the actio n of bluegreen algal toxins on ca ncer growth . This is the basis of the CRC-fund ed experim ents that have bee n un derway in o ur la borat o ry for th e last fi ve y ea rs. Microcystis is the predo minant toxic alga in drinking water reservoirs wo rldwide and it's toxins have received th e greatest study. Our first series of expe rim ents were based o n ea rl ie r studies that showed that the toxins from 1'\llicrocystis can stimulate th e growth of precancerous ce lls in skin and the li ve r , see (Kui perGoodman , e t al. , 1999) . H owever in the co n text of cance r in the Australia n population, we considere d that ca ncers of th e gastro intestinal tract and the skin arc of grea ter clinical signifi cance than cancers of t h e live r, so our experi ments have aimed at thi s area. Duodenal cancer growth. Th e first series of trials in mi ce used a known carcinogen, methyl- nitroso- urea (wh ich is si mi la r t o so m e ca r c ino gen ic compou nds in the die t) , to ac tivate ce lls in the upper part of the intestine into precancerous growth. T he mice were then provided w ith a range of dilutio ns
WATER MARCH / APR IL 2000 C.RC. for \\1/:itcr Quality and Trealmcnt
of Microcysris extract in the ir drinking wa ter over periods up to 22 wee ks. After euthanasia the animals were autopsied and the gut and organs examined for ca ncers. Th e mice showed ca ncerous growths in t he li ning of the upper small intestine and of w hite ce ll invasion in the liver. H owever there were no increases in the s.ize of these growt hs in respo nse to th e ran ge o f co n cen tra tion s of Microcystis, indi catin g that these cells were not growin g fas ter as a result of the toxins the m ice were drinking (Falconer and Humpage, 1996). Colon cancer growth . T he seco nd series of e xperiments looke d at t he possibility t hat colon cancer might be stimu lated by Microcystis toxins, since the main excre to ry ro u te for t h ese toxins is through the faeces. T hu s the cells lining th e co lo n are likely to be co n tinuously exposed to the toxins, w hen the an im.als are drinking a water supply contaminated by Microcystis. Simila rly to the previo us experim ents a kn own carcin ogen was used to initiate preca n ce rou s g r owths, this ti m e azoxymethane, wh ich causes carcinogen ic muta ti ons 111 cells of the colon
li n ing and in the liver. Mice were th en given 3 dilutions of Microcystis extract for 30 weeks. At autopsy the presence of raised hypertrophic crypts in the colon linings were recorded and thei r areas meas ured. Data from these measurements of precancerous growths in the co lon linings are currently being prepared fo r publication. In the li vers of these mice a co mpl ex pattern of ca n cerous nodules was also o bserved, and there were so m e similar nodules in the lungs. From data such as these WHO decide w hether there is evidence of carcinogen icity and the risk to huma n populations. Cyli11drosper111opsis toxin T his toxic blue-green alga has come into recent promi nen ce in Australia as a res ult of it's appearance in the reservoirs supplyin g Brisbane. It previously was responsible for an o utbreak of gast roenteritis o n Pal m Island , in which some 140 c hildren and 10 adults rece ived hospi tal treatm ent (see Kuiper-Goodman et al. , 1999) . T he o rgan ism is predom inantl y a tropi cal species though it occurs widely in temperate waters at low cell numbers. T he c he mical nature of the identified tox ic mol ecul e synthesised by the alga has been shown to be an alkaloid with a side chain w hi ch is a nucleic acid base (u raci l) . This immediately suggests tha t it might be a carc inogen. T here are several possibil ities for DNA damage by th is type o f molecul e, leadin g to mutation s which can initiate cancer. Cancer evalu ation. At th is stage of researc h there is no data on the indu ction of can cer by Cyli11drosper111opsis toxin, either in humans or in experimenta l an imals. Und erway in our la boratory is a trial to eva lu ate the cance r- initiating capac ity of Cyli11drosprn11opsis extract give n by m o u th to mi ce. In thi s experim en t we are focussing on t he ea rl y stage of ca ncer formation and afte r the toxic dose we are suppl yin g a known promoter of ca nce r growth in the food of the mi ce. In this way a ny ce lls initiated into a potentially cancerous state by the toxin arc more likely to show up as tumours during the time period of the experiment. We wi ll have completed th is trial by th e e nd of the year and hop e to ha ve a clea rer understanding of the potentia l for cancer initiati o n by Cyli11drosprn11opsis as a resul t. Toxicity. As part of o ur research into this organism we are also studying the long-te rm toxic effects to mice of drinking water co ntaining low co nce ntration s of Cyli11 drospcm1opsis extract . Acute poison ing by th is toxin has been stud ied earlier, but the effects of continued exposure arc unkn ow n (Falconer ct al. , 1999; Seawright ct al. , 1999). To determin e a sa fe conce ntration for this tox in in d rinking water for human consum ption over a li fe tim e req u ires exte nsive data for lon g-te rm toxicity in animals . T h e W H O has rece ntly determin ed a safe 'Guideli ne Value' for the toxin m icrocystin - LR in drinking wate r, from oral toxicity experime nts in mice and pi gs (Falconer ct al. , 1999). For a sim ilar 'G uid elin e Va lue' to be determi ned for Cyli11drospen11opsis roxi n a range of low co ncentrations dosed by m outh o r in drink ing water to mice o r rats will need ro be eva luate d for adverse effects on the animals over man y weeks. Th e expe rim ents that we have u nderway at present will provide data cowards th is evaluation. Cell-based techniques
We are partic ipating in the international t rend away fro m ro xici ty assessment using whole an imals and have develo ped a series of m etho ds usi ng isolated cells and cell cu ltures for measurement of cyanobacte rial toxicity. Suspensions and monolayer cu lt ures of hepatocytes have been used to explore the m ec han ism by w hich mic rocystin-LR. promotes ce ll mu ltiplication in liver ce lls. Our data suggest that very low concentrations of microcystin stimulate cell divisio n in these ce lls ;rncl also pre vent the no rmal m ec hanism for ce ll death of damaged cells (Humpagc and Falconer, 1999). H epacocyte cultu res are also in use for assessment of t he m echanism of action o f cylindrospcrmopsin and for compariso n of purified toxin w ith cell extracts containing the toxin. The
WATER MARCH/ APRIL 2000 C-RC for \'(l,1tc:r Qu,,lit)' ,incl Trcatmcnt
19
WATER
presence of more than one toxin in Cylindrospermopsis is a possibility and can be investigated by this m ethod. Genetic damage to cultured human cells from cylindrospennopsin poisoning is also under inves tigation , in collaboration with the CS IRO D ivision of Human Nutritio n. T he mechanism by which protein synthesis inhibition by the toxin occurs is being studied as part of the PhD program of Suzanne Froscio, w ho is using cell extracts to sepa rate out the components of the reaction . Conclusions
The study of potential human health risks from cyanobacterial toxins can only be done at present by an im al experiments of the type described, supplemented with information from accidental huma n poisoning. Th e researc h in our laboratory is exploring whether these cyanobacterial toxins can cause cancer ini tiati on and or stim ulate ca ncer growth in experimental an imals. On com pletion th e data wi ll be evaluated by WHO as part of the on-going revisio n of drinking water guidelines and ultimately used by national authorities for determi ni ng local
QUALITY
AND
TREATMENT
guidelines or standards for safe water supply. Acknowledgments
T he authors are m embers of Program 1 (H ealth Risk Assessment) of the CRC for Water Quality and Treatment and wish to thank th e University of Adelaide for the Hono rary Visiting Fellowsh ip awarded to Em.Professor Ian Falconer and the o ngoing use of facil ities in the Medi cal School for this research. References Falconer, I. , Bartram,
J. , C horus, I., Kuipe rG oodman , T., Utkilen , 1-1., Bu rch, M. and Codd, G.A. (1999) Safe Levels and Safe Practices. In: Toxic Cya11obacteria i11 Water. A C11ide To Their P11blic I-lea/ti, Co11scq11ences, Mo11itorit1,{! a11d Mmiage111e11t (Chorns, I. and
Bartram, J., eds.), pp. 155-178, E & FN Spon on behalf of WHO, London. Falconer, l. R.., H ardy, S.J., Hu mpage, A .R.. , Froscio, S.M. , Tozer, G.J. and Hawkins, P.R . (1999) Hepatic and renal toxicity of the blue- gree n alga (cya noba cterium ) Cyli11drosper111opsis mciborskii in male Swiss Albino mice. E11viro11111e11tal Toxicolo,~y 14, 143-150 . Falconer, l.R.. and H umpage, A. ls.. (1996) Tumour promotion by cya nobacte rial toxins. Phycolo,~ia 35, 74-79. Humpage, A.R . and Falcone r, l.R.. (1999) Mi crocystin-LR. and liver tumour promo-
tion: Effects on cytokinesis, ploidy and apoptos is 111 cul t ure d h e pato cy t es . E11viro11111ental Toxicology 14, 61 -75 Kuiper-Goodman, T. , Falc oner, 1. a nd Fitzgerald, ). (1999) Human Health Aspects. In: Toxic Cymiobacteria Ill Water. A Guide To
Their P11blic Health Co11seq11e11ces, Mo11itori11g a11d Ma1iage111e11t (C horus, I. and B artram, J. , eds.), pp. 113-153, E & FN Spon o n behalf ofW H O, London. Pilotto, L.S., I<Jiewt:r, E.V., Davie::s, R..D. , Burch, M .D. and Attewell , 11...G. (1999) C yanobacterial (blue-green algae) contamination in dri nking water and pe rinatal outcomes. A11stralia11 and New Zeala11djo11rnal of Public Healt/1 23, 154-158. Seawright, A.A., Nolan, C.C., Shaw, C.R ., Chiswell, R .K. , Norris, R..L., Moore, M.R.. and Smith, M.J. (1999) The oral toxicity for mi ce of t h e tropi ca l cya nobac terium Cylindrosper111opsis raciborskii (Woloszynska). E11vim11111e11tal Ti,xicology 14, 135-142.
Authors
Ian Falconer, Andrew Humpage, Emma Moore and Suzanne Froscio are with the Department of Clinical and Experim ental Pharmacology, University of Adelaide M edical Sc hool and the Cooperative R esearch Centre for Water Quality and Treatment.
Waste water management is a job best left to the experienced.
And few companies have as much experience as Henry
Walker. ln the last 17 years we
have successfully completed over 300
turnkey waste
wate r
trcann ent plants.
Call as It yaar iumt HIiary Walker olflcl for farther lnfarmatloa.
Over that time we have evolved
s,.,
simple, robust technologies, including our patented gaslocked syphon decanter system .
We've applied what we've leamt not only in Australia, but also in
Malaysia, Thailand, Indonesia, Fiji, Western Samoa, Papua New Guinea, Soucl1 Afiica and China. Wherever they arc found, waste
water treatment plants from Henry Walker are known for producing consiste ntly high quality cffiucnc. Even more importantly, they deliver the lowest
HWE
whole-of-life
cost
compared with altcmativcs. If you're contemplating a waste water management project, you should talk to Henry Walker
Henry Walker Eltin
Envir011111cntal.
Uait 3, 1 The Crescent
Klapgm1 2208, New South Wales Telephaal 81 1012 9554 4555 facsimile 81 1012 9554 4827
.......
Level 1, 147 Frome Stnet,
...
Adal1hle 5000, South Aatnlll
Telephoae 81 1018 8232 4949
55 Broadway Netlllllds 8909, Westem Aastnll1 Telephaal 81 1018 938S 4777
....
858 Stuart Highway, Berrtaall 0828, Northern Territory TeleplleM 81 1018 8943 1000
WATER QUALITY AND TREATMENT
MIXING IT UP WITH BLUE-GREEN ALGAE M D Burch, J D Brookes, P Tarrant, P Dellaverde Research Progrn111 T1110 ro11ce11trates 011 Catr/1111e11t a11rf S0 11rce Water J\lln11agellle11t. O11e of its Projects i11vesrigares rfestratificntio11. The excessive growth of cyanobacteria (blue-green algae) in rese rvoirs is an ongo in g probl em for the wate r industry. T h ey arc recognised as a source of tastes, odours, and toxins in drinking wate r. The presen ce o f these contami nants, even at quite low concentratio ns, represe nts a source of co mplaint and custome r dissatisfacti on . T he occurre nce of toxins in drin ki ng water w ill also soo n be a regulatory and compliance issue, with the imm in ent re lease of drinking water gu ideli nes fo r so me toxins durin g the year 2000. T oxins, tastes and odours ca n be re m oved by treatme nt but not w ith out som e difficu lty and co nside rable cost. Given th is ex pe nse, it is sensible to investigate a balanced range of o ptio ns for catc hme nt mana geme nt, so urce water m anageme nt and water treatment as part of a comprehe nsive risk- redu ction strategy . Strategies in vo lvin g impro ve d catc hme nt managem ent leading to nutrie nt reduction arc no t likely to provide an i mmediate or uni ve r sa l sol u t io n. Th erefore so urce water strategies should b e co nsi d ered in co nj unctio n w ith o p tio ns for water treatm e nt. In tailoring a so lution there w ill always be degrees of sui tabil ity, va rying timeframes for implem e ntation and effect, and of course cost. Enhanced artificia l mixing used in co njun ct io n w ith de stra tifi ca ti o n is potentially a cost-effective opti on to consider as a short-term stra tegy for the control of cya nobacteria in reservo irs. D es tratifica tion w ith bubbl e- plum e ae rators has a long hi story of use and has h ad variable success w he n used fo r algal control. The techn iqu e is used wide ly for oxyge natio n of hypo li mnetic \vater to co ntrol th e remobil isa tion of nutrients and metals, such as iron and manganese, fro m rese rvoir sediment~. The relati ve su ccess of bubble plumes to destratify r ese r vo irs w ill depen d up on t h e ba thym etry and size o f the sto rage, the local climate, an d the aerator d esign . Relative ly little aeration takes place in th e bubbl e plumes th emselves. l nstea d,
Figure 1. Location map th e large-scale circulatio n set up by the d es tratifi e r weakens th e temp e rature grad ient and the refore t he res istance of the wate r column to m ixin g by wi nd and s urfa ce coo lin g . H oweve r , d u rin g pe riods o f hot ca lm wea th e r, it is beli eved that su rface heating, particu larly in rese rvo irs w ith signi fi cant shall ow areas, can still promote and sustain t he growth of cyan obacteria in rese rvoirs that are equ ipped with aerators. This is the case, o n occasions, in M yponga and Little Para R eservo irs in Sou th Australia, both of w hich have o th erwise effecti ve aerators. Moni toring has shown that periods of surface hea ting have been associated w ith blooms o f A11abae11a circi11alis w hile th e aerators arc operating (S A W a t e r , unpubli sh e d) . Quite sm a ll te mpe rature gradie nts (<1.0°C), w hi c h
previously wou ld not have been conside red ecologically significa nt, may provide suffici ent stability to allow cya nobacte ria to become do minant. T he c hall enge is to fin d practical ways of preventing the establishment of a buoyan t surface layer. T he CR C for W ater Qua li ty and Tre atm e nt in co nju n ctio n with SA Water is evaluati ng a nove l techni que to control th e gro wth of cya nobacte ria in drin kin g water reservoirs. Th e tec hnique in corporates rese rvoir mixing usin g raftmounted mecha nical mi xers in combination with an aerato r, ie a hybrid system. T he aim of the proj ect is to evalu ate this dest ratifi cation tec hnique at M yponga and H appy Vall ey R ese rvoirs in Sou th Au stral ia (Fi gure 1), as a m ea ns of controlling th e abundance of cyanobacteria in these two reservo irs of diffe re nt size and average depth. T he aerator will be used to brea k down t he primary th ermoc lin e w hil e t he m ixers w ill be di rected at entraining t he warm surface layer into th e main circulation patte rn o f th e reservo ir (Figure 2). T h is co mbination n o t only h o lds prom ise for contro llin g cya nobac teria bu t also has prospects for redu cing th e cost of destratification. Th e use of th e mec hani cal mixers may allow the inte rmi tte nt use of aerators, and th e re by achi eve savings in th e runnin g costs of the compressors used for aeration. M ypo nga R ese rvoir was chose n as a
Figure 2. Destratifieation by Hybrid Mixing
WATE R MARCH/ AP RIL 2000 CRC for W;1tcr Quality ;:ind TrcJtmt:nt
21
WATER
QUALITY
AND
TREATMENT
req uireme nts of cya nobacte ri a and other plankton ic microalgae, as they arc circu lated th rou g h t h e wate r column . This is achieved by both laboratory and field based studies that involve probing the pho tosynthe tic systems of the algae w ith Pulse Amplitude Modulated (PAM) Fluorometry to determine the amount of light received and utilised by individual cells and co lonies of Figure 4 . The mixer is mounted on a floating ph ytoplankton . raft which is anchored in the body of the Th e proj ec t is a reservoir multi-d isc iplinary effort, reservoir (Figure 4). The with a team of 13 people, combining mixers hav e pr ov ided expertise in engineering, ecology, instruflow-del ivery rates of 4,500 mentation and hydrodynamic modelling to 6,000 L sec- 1, dependin g stu di es. Partne r organ isations invo lved in upon the length of t he draft the work are SA Water, the Water cube, and each unit is Corporation (WA), the Austra lian Wate r powered by a 3 k W Quality Ce nt re, Th e University of e lectric motor. The mixers Adelaide, CS IR O Land & Water, and were install ed ove r the the C R C for Freshwater Eco logy. The Figure 3. A mechanical mixer for algal control being summer of 1998/ 1999, and project runs over the period 1998-2001. installed w ill have their first full References primary site to eva luate th is techn ique as season of fie ld tria ls in '1999/2000 . T he Velzeboer, R., Cugley. J. A. and t>atterson, J. C. it has a hi story of cya n obacterial mixers were designed and supplied to SA ( 199 1). Modelling optimum conditions for problems, a good historical data set Wate r by an Australian Company, Water reservoir destratification using mechanical mixers. Research Report No. 24. Urban Water including physica l, chemical and biolog- Engineering and R esearch Solutions l~esearch A'5ociarion of Ausa.ilia, Melbourne. ical data and has previously been the site (WEARS Pty. Led.) of artificial destratification studies. Th ese A major assumption of dcstratification Authors studies have in cluded the use of bubble as a manage ment tool to control Michael Burch is a project leader plume aerators and submersible mechan- cyanobacteria is that the cyanobacte ria and senior research biologist at the C R.C ical mixers moun ted on the dam wall will be ci rculated throughout th e water WQT and the Australian Water Quality (Velzeboer et al. , 199 1). Myponga is a column and will expe ri ence light limita- Ce ntre. H e has a long-standing interest relatively deep and high ly coloured tion. T h e buoyancy afforded to i n res earc h and manag e m e nt of reservoir, which should make mixing an cyanobacteria, by gas vesicles, is clearly cya noba cteria in lakes and rivers. Mike is effective method for red ucing cyanobac- advantageous in stable water colu mns also currently the AR.MCANZ National terial growth provided that the cells are because it allows the cyanobactcria to Al gal M anager. Ph.: (08) 8259 0352, emixed down below the leve l of light out-compete so me other algal groups, mail: m ikc.burch@sawater.sa.gov.au penetration. Two mechanical mixers are w hich sin k out of the illumin ated surface Dr Justin Brookes is a research being used here in conjunction with the zone. However, the cyanobacteria h ave sc ientist specialising in reservoir mix ing small light harvesting antennae relative to existi ng aerator. research at the CR.C WQT. Justin has Although Happy Va lley R eservoir is a o ther microalgae and therefore arc more research interests in cya noba cterial sma lle r and shall ower storage, it was susceptible to light limitation duri ng ecology, photo-physiology and buoyancy chosen as an additio nal study site beca use mixing. The challenge for this project is control. e -mail: ju stin.brookcs@ of its strategic importan ce to the to determine what degree of mixing is sa water. Sa .gov .au . Adelaide water supply system and the necessary to promote light-limitation of Peter Tarrant is Man ager Water regular occurrence of algal blooms. the cyanobacteria and whethe r the Quality and Paul Dell a verde is The mixers under trial entrain surface proposed mixing systems are able to Manager Operations, Bu lk Water water into a draft tube, which then exits accomplish this. Division of th e South Australian Water at a depth of 13111 in Myponga The experimenta l studies being carried Corporation . Both have ex t e nsive R eservoir, and at 7m in H appy Valley out in this project are investigating the experience in wa ter supply e ngineering, R eservoir (see Fi gure 3). Each mixer influence of the mixers o n deepening the water treatme nt , water quality and reserconsists of an imp ellor, which is five surface layer and redu cing the access of voi r management with SA Water, em etres in diameter, a nd wh ich is the cyanobacteria to light. This is being mail: pcter.tarrant@sawatcr.sa.gov .au, positioned two metres below the surface. done by exa mining the li ght dose pa u l. de Ila verde@sawa tc r .sa. gov .a u. The entire unit is m ounted on a freefloating raft anchored in the body of the
22
WATER MARCH/APRIL 2000 CRC for Water Quality Jnd Trc.11mcm
•
Ultraviolet Technology of Australasia
How much is UV Disinfection of your Secondary Treatment Effluent costing? a) In$ Terms b) in Environmental Terms The answer is: UV Operating costs would be lower if you use virtually non-scaling, non-wettable, Tough, AFP Tubing through which the effluent flows. (Ask UVTA to demonstrate how tough AFP is)
Advanced Fluoropolymer Tubing (AFP)
~
A Effluent/ Water
UV Lamps ~
UVTA Design offers: a) Less Fouling b) UV Lamp Temperature Control UV disinfection is good for the environment - no chemicals/oxidants to worry about. So think of the future, pay a little more up front when purchasing and save countless $ for many operating years by buying a UVTA UV system.
0
• •
• • •
Just fax:
• •
Water Industry Alliance
(08) 8337 9465
Or phone: (08) 8337 0079 P.O. Box 222 MARDEN SA 5070 Email: uvta@uvta.com.au Web site: www.uvta.com.au
I
Quality
Endorsed Company
\SO Sl002 Uc 2712 Slandards Australa
Easy maintenance and high kill rates of disease causing virus and bacteria.
For more information call (03) 9728 5855, Toll Free 1800 335 571 or visit our Internet Site at www.merck.com.au
WATER QUALITY AND TREATMENT
NOM TREATABILITY AND CATCHMENT CHARACTERISTICS KM Spark of th e DOM fra ctio n depe nds o n the four major types of processes (Spark 1999a) ; che mi cal and ph ysical processes Introduction in th e so il and wa te r e nv iron m e nt The remova l of NOM from rese rvoir (Fabris a nd Spark 199 8; Page et al. , wate rs is a sign ificant cost in the trea t1998 ; Spark l 998 ; Spark e t al. , 1998 ; ment o f water for drinkin g purposes in Spa rk 19996; ), mi c robial processes m any areas. Conside rable research has (An sti s 1 99 9), a n d ph o to c h e mi cal been carried out to improve the methods p ro cesses Qa blonska s 199 9 p ri vat e of NOM re moval. In co ntrast, li ttle commun icati o n ; Spark 1 999a) . Th e investi gatio n has take n place to dete rc haracte ristics of DOM impa cti ng on min e the major catchme nt param ete rs the treatability of th e reservo ir w ate r has asso ciated with hi gh and recal citrant also been studi ed (Chow e t al. , 1999). NOM in reservoir wate rs. Catch me nt W hat is not kn o wn , the m iss in g lin k m anage me nt as a means of reducing the be tween these two bodi es of knowledge, amo unt ofNOM in reservo ir wate r may is th e relative signifi ca nce within a be a more economic alternative to th e catchm ent o f th ese fo u r maj o r pro cesses, indu stry , bo th direc tly in te rm s of and he nce th eir impa c t on th e fin al r e d u ced tr eatm e nt c o s ts , as w e ll character of the DOM transported to in directly by reducing costs asso ciated rese rvoirs. Th e relati ve impo rtance of w ith sludge disposal and high nutrie nts in eac h of th ese processes wi ll de pend stora ge waters. partially on th e path by w hi ch th e DOM The Project is transported; surface , stream or su bNOM co nsists of two fract ions, su rfa ce w ate r flo w. particu late organic matter (PO M ) and T o ga in a be tte r understanding of the dissolved organic matte r (D OM) , of relative significance o f each of th ese which th e latter is th e most diffi c ult to processes it is useful to take a " topremove from wate rs using conventio nal down" approach , whi ch in th is situ ati o n w at e r trea tm e nt m e th ods. Previous would be to compare th e ca tch me nt research in this proj ec t has investi gated characteristi cs o f a number of catchm en ts how th e co ncentration and composition w hi ch d iffer in th e amou nt and natu re of th e DOM in th e raw as so c iat e d waters. T he foc us o f th e co mpari son wou ld be to determi ne wh ich one or more o f th e fo u r pr oc e sses m aj or influ ence the DOM fra c ti o n , are m o st lik e ly c ontrolling th e d iffer e n ce in quantity and character of the DOM 111 the reservo irs. T h is t yp e o f AWQC's Dr Kaye Spark with colleague Dr John van Leeuwen (centre ), and CRC for Water Qua lity and approac h was part-
Research Progra111 T 1110 dea ls with Catch111ent arid S011rce Water Ma11a,(!e111e11t.
Treatment post graduate student, Declan Page
ially applied in two prev io us stu dies co mparin g cat c hm e nts in southe rn Au stra li a. fn b oth cases th e co mparison was made be tween a pair of ca tchm ents in a similar cl imatic region , on e in the Mt Lo fty Ran ges in South Australia (N elson et al 1990) and o ne in th e Otwa y R an ges in sou th ern Vi ctoria (N elso n e t al 1993). In both studies it was found that th e catchment with th e hi ghe r proportion o f clay to sand pa rticles was most likely th e reason for th e lower amo unt o f DOM transported to strea ms in the catchment. H owever, in th e seco nd study it was fou nd that th e natu re of th e DOM in the soils diffe red to that in th e streams. As the soils did not app ear to be selective as to the nature o f th e DOM adsorbed , it was concluded th at mi crobial pro cesses ma y also be signi fica nt with respect to the nature of th e tran sported DOM . A more recent study (S tevens e t al. , 199 8) has sho wn th at th e a ve rage concentration of DOM transported via eith er surface or along the A/ B ho rizon interface in a sandy loa m soil catc hment was consiste ntly about double that transpo rted in a silty-cla y loam soil catc hm e nt over th e sam e July-September per.iod . Th e two sites, both covered with wi nte r g rass, we re in t he Ad elaid e re gion (Myponga and M t Bold reservo ir ca tchments, respecti vely). H oweve r, th e total amo unt o f DOM transported per hectare (surface, A- and B- horizon flow) from the catchments was at least 15% m ore in the silty- clay loam so il. Clearly, information about the hydrology of catch m ents is important in understandi ng the signi ficance of the individual processes affec ting th e nature and transpo rtation o f th e DOM on a catchment scale. T akin g a " top- do wn " approac h can identify possible catchme nt characte ristics that coul d produ ce differe nces in transported DOM. Consider the possible effec ts of not just the so il type, but the soil e nvironme nt. T he conce ntratio n
WATER MARCH /APRIL 2000 \RC for \ '(/.itt:r Qualit)'
and Treatment
25
WATER
of DOM in the reservoir wate rs servin g th e Ad e laid e re g ion is ge n e ra ll y relatively high compared with that in r ese r voirs servi n g mo st of th e Melbourne region. Th e average "Tru e colo u r" valu es fo r untrea ted/treated water from M ypo nga (-55 km S of Adelaide) reservoir over th e years 19921996 were 80/4 H U (data suppli ed by SAWat e r), whereas th e " Appare nt colour" valu es for the Sugarloaf (-95 km NE of Melbourne) and Yan Yean (-35 km NE of Me lbo urne) reservoirs were 20/3 and 18/4 HU, resp ectively (data supplied by M elbourn e Water). Most of Adelaide raw water to be treated for drinking purposes co mes from reservoirs on the o utskirts of th e metropo litan area with additional supplies comi ng from the Murray River fo r all except the M ypo n ga rese r vo ir. The w at er in Melbourne comes fro m rese r vo irs lo ca ted mainly in the north and eastern areas o f M elbourne, th e climate o f which tends to be wetter than that associa ted with the Ade laide water supply regions. Because o f the vari ety of so il mineral and chemical environmen ts w ith in th e catchm ents of reser vo irs in both th e Adelaide and Melbo urn e regions, it is unlikely that the observed consistent differe n ces in th e quantity or qua lity of tran sported DOM between the two regi ons co uld be specificall y expla ined by the mi neral or chemica l processes. l n additi on, the two regions are sim ilar w ith respect to cloud cover so photoc h e mi ca I pro cesses wo uld not be expected to be abl e to explain the differen ces eith er. Th e tempe rature and rainfall variatio n for regions in corporating th e three reservoirs m en tione d above (Fig. 1), show that although there is very little diffe rence in the average high and low daily temp eratures in the two regio ns, there is a striking difference in the rainfall patterns. T he rainfa ll in the two Melbou rn e regions is m ainly aro und 50- 100 111111 per month over the w hole year, whereas the Adelaide regio n receives the majo rity of th e rainfall in th e 7 months around the w inter period. Th e differe n ce in ra infa ll p att e rn between the two regions may lea d to differences in soil sorption prop erties and tra nsport paths as a consequence of th e wet/ dry characteristics of the so il s. Microbial processes are also strongly dependent on moisture co ntent and availability of nutrients. Th erefore, the difference in rain fall pattern between the 26
QUALITY
AND
TREATMENT
80
Myponga
120
70
Yan Yean
100
e:
60
80
::,
50
60 iil (0
40
40 iil
30
20
~ 20
0
~
"@
(t)
Cl)
0..
E
5
2
Cl)
Ol
~
(t)
~
Cl)
-
10
~ Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
2,
-20
MinT
0 Jan
~
3
Nov
-40
Dec
Month Figure 1. Average monthly rainfall and temperature (daily maximum and minimum) for t he Myponga region in South Aust ralia, and for the Yan Yean and Sugarloaf regions north of Melbourne in Victori a.
two regions may also result in quite different mi cro bial processes in the litter and soil profiles in response to the differing moisture content. Learning more about ho w these different environmental conditions affect th e quality of the transported water will be th e fo cus of fu rther research in this proj ect. H opefull y, the results of these studies wi ll indi cate possible catchment m anagem ent strategies to improve th e quality of transported DOM. References Anstis, S., 1999 "The biodegradation of natural organic matter under di!Tcrent vegetation systems" J-/o11011rs pro;ect tlicsis - B. App. Sci,
U11iversity ~f SA Chow, C., van Leeuwen , J. , Drikas, M ., Fabris, R.. , Spark, K. & Page, D. , 1999. 'The impact of the character o f natural organic matter in co n ve n ti on a l treatment with alu m ',
Proceedi11gs ef 1/,e IAWQ/IWSA /11tematio11nl Co11fere11ce 011 tl,e Re1111)1Ja/ ,!f J-/11111ic S11bsta11ces .fro111 l1Vater, T rondheim , Norway.June "1999, pp 105-¡J12 Fabris, R. and Spark, K. 1998. 'The efTect of the soil: environmental factors on the nature o f t he dissolved organ ic matter', Proceedi11gs of
tlie 141/, W orld Co11gress
~f
Soil Scie11ce,
Symposium 7, M ontpdlier, France, August 1998 (CD-R.OM ) N elson, P.N. , Baldock, J.A. and Oades, J. M. 1993. Concentration and composition of disso lved organic carbo n in streams in re lation to catch ment soil properties. Biogeoche111istry, ¡19:27-50. Nelson, P.N. , Cotsaris, E., Oades, J. M. and I3ursill, D.13. 1990. The influence of soil clay content on dissolved organic matter in stream waters. A 11st. J. i\llnr Freshwater Res. 41 :76 174 .
WATER MARCH/APRIL 2000 \.RC: for \'(/,Jlcr Quality and Tn:,1tmenl
Page, D., van Leeuwen, ). , Spark, K. , Mulcahy, D. & Anstis, S., 1998. ' Biopolymer degradatio n products as specific tracers of plant classes and input to dissolved o rganic matter in reservoirs', 9tl, Meeri11,~ of tlie illtematio11a/ J-/11111ic S11bstn11ces Society, Adelaide, September 1998, (in print} Spark, K. 1998. Variation in the nature of dissolved organic matter in a catchm ent.,
Proceedi11gs ef the 211d A W W A WarerTecl, Co,!ferwcc, Brisbane, April 1998. C D- ROM. Spark , K., 1999a. ' Dissolved organic matter in reservoirs; a review', Water, 26 (3), pp 35-38. Spark , K. , "19996. ' Improving drinking water quality from surface water stud ies; catchment studies', R.esearch Report. A11stralia-Gemrn11y
collaboratio11 011 c11vir011111e11/ R &D. Spark, K.M., Stevens, D. P. Cox, J.W. and C hittlcborough, D .J. 1998 . The efTect of transport path on the nature of soluble organic matter in soil leachates. Pmceedi11gs ef the N11tio11al Soils Co1ifere11ce, Australian Soil Science Society Inc. Brisban e, April 1998. Stevens, D .P. , Cox, J .W . and C hitt!eborough, D.J. 1998. Measurement and treat ment of phosphorus and carbon subsoil m ovemen t . Project UA D 10, La11d a,rd l;f/afer R esources
Co1pomlio11
Author
Dr K.M. Spark is the Project Leader at the Australian Water Quality Centre. T he team includ es Dr J. v an Leeuwen, a Senior Res earch Offi ce r , a nd T ec hni ca l Offi ce rs Lidia Sledz and Rolando Fabris. T hey acknowledge the assistance of PhD students Graem e Jablonskas and Jon Varco e (University of Ad elaide) and D eclan Pa ge (University of South Australia), and Hono urs student Simon Anstis.
The world's best range of performance pipework systems is available now in Australia, off-the-shelf, from GPS.
GPS Australia ca rries a comprehensive ran ge of fitt ings, va lves, couplings, flowmeters and instrumentation suitable for the water, gas, mining, process and industrial sectors. The vast range of GPS products has been developed from unmatched expertise in the processing of all the important polymers - PP, PVCu, PVCc, PE, ABS and PVDF - as well as t he manufacture of forged and cast comp onents. The GPS product range includes such well-known brands as Durapipe S&LP, Viking Johnson, WASKRMF, Straub, FIP, Akatherm, SEO and Philmac. We're m ore than happy to work wit h you in designing a piping system that's righ t for your project, or simply to recommen d the best solution to your needs.
Glynwed Call 1800 246 167 freecall or email info@gpsaustralia.com.au for more information
PIPE SYSTEMS
A U S T R A L
'
.
e
WATER QUALITY AND TREATMENT
TASTES, ODOURS AND ALGAL TOXINS, WHICH PAC IS BEST? D Cook, G Newcombe, and J Morrison Abstract Prograrn 3 has, as 011e of its projects, the testi11g of activated carbo11s. Bloom s of cya nobacteria (bl ue-green algae) in reservoirs produ ce nume rous n1etabolites during their li fe cycle. Som e of these compounds impa rt an unpleasant taste and odo ur, while others are toxic. Powdered activated carbon (PAC) is oft en co nsidered as a treatment option for these compou nds and ca n be very effective and easy to apply. However, there is no "un iversal" PA C capable of r e moving al l of these prob lem compo unds to the desired leve ls. The aim of this work was to determ ine the bes t activated carbon fo r th e adsorption of M lB, geosmin, and th e sax itoxins, and to compare these resu lts with previo us work on microcystin - LR. Six co mmerciall y ava ilabl e carbo ns we re tested, over contact tim es that m ig ht be realisticall y ava ilabl e in a wa ter treatment fac ility (a few minutes to several ho u rs), and at equilibrium (three days co ntact). It was found that, fo r both MIB and geosmin , the best activated carbo n depended o n the contact tim e. In su mmary, at contact tim es of 2 hours or less, it was fo und that one particular ac tiva ted carbon cou ld be appli ed to remove both geosmin an d m icrocysti nLR, while tha t particular activate d carbon would not be considered the opti111¡u111 fo r the rem oval of M IB and saxitoxins.
Key Words Activated ca rbo n , adso rpti o n , MIB, geosmin , sax itoxins, mi crocystin, water treatment
Pictured from left to right are Janina Morrison, Dr Gayle Newcombe, and David Cook
rial metabolites such as 2-methylisoborneol (M!B) and geosmin at low ng/L levels. T wo types of algal toxins of concern are the hep atotoxic microcystins and the neu rotoxic saxitoxins. In Australia the main sou rce o f mi cro cystins in drinking wate r so urces is the cya nobacte rium Microcystis ae ruginosa . The m icrocystins have been fo u nd to cause liver damage and promote tumo urs in m ice. More than 60 different hepatotoxic m icrocystins have been charac te rised (Sivonen and Jones (1999)). Studies have co ncentrated on the 111.ost toxic hepatotoxin, mi crocysti n-LR. A provisional guideline valu e of 1 ~tg/L of microcystin-LR has
been adopted by the W o rld H ealth Organisati on. Saxito x in s (o r paralytic sh ellfis h poisons) are co mmonl y produced by A11abaena circi11alis in Austral ian waters (Humpage et al. , 1994; Negri et al. , 1997) . T his group of compounds is well known fo r causing illness an d sometimes death in people eating contaminated marin e shellfish. No illness similar to para lytic shellfish poisoning has been reported in h umans from th e consumption of drinking wate r containi ng saxitox in s (Fitzgera ld et al. 1999) . Saxitoxins can be divided in to three groups of varian ts according to their che mica l structu re, w hich in turn affects
Introduction Cyanobacterial bloom s in waters used as a source to produ ce potable water are of concern as they can p r oduce so m e co mpounds that are aesth etically unpleasant and o th ers that can be toxic if co nsum ed. M u s t y-ea rth y ta ste and odo urs ca n be caused by cyano bacte-
28
Table 1 . Physical charact eri stics of PACs Carbon
HP CA-10 SA-30 *F-400 P-1100 PC0
Pore Volume (cm 3 g-1 ) Surface Area Primary micropores Secondary micropores Mesopores m2g-1 w<O.Snm 0 .8<w<2nm 2<w<50nm
2183 1 491 1 077 1 047 1 241 987
* GAC ground to a PAC
0 .57 0 .39 0.35 0.34 0.38 0.31 w refers to pore width
WAT ER MARCH/APRI L 2000 C.RC: r, ,r \Xlatcr Quality ilnd Trc;Hmcnt
0.54 0.47 0.13 0.12 0 .18 0. 13
0.51 0.41 0.09 0.09 0.05 0.03
Starting Material
Activation Method
wood wood wood coal coconut coconut
chemical chemical steam steam steam steam
WATER
QUALITY
AND
TREATMENT
equilibrium o nto PA C could b e direc tl y re la te d to th e geosmln 70 vo l u m e of m cso por cs MIB (lnm < width > 50 11111) . Th e m eso p o r o u s wood- b ase d 50 PA Cs di sp la y ed superi o r ,o mi c rocys tin - LR. ad sorp ti o n 30 e ffi c ie ncy at equilibrium and at sho rt contact times expected 20 in a water treatment plane. 10 Littl e research has foc used o n saxitoxin adsorptio n w ith HP CA- 10 SA-30 F 400 P 1100 PCO PAC PA C . Bailey e t al. (1999) re port g reate r than 90 % removal of sax itoxin with fou r Figure 1 Equilibrium adsorption of MIB and wood and coa l based PACs. geosmin for each PAC H o wever, th is was for the r e m ova l o ne sa x ito x in the ir toxicity. Due to the differe nt toxicanalogue (STX). T o assess th e ca pability ities and con centrations of saxitoxi ns th e of PAC to re move saxito xin s, t h e e x tent of PAC adso rption sho uld be re mo va l o f th e vario us analogues at examin ed in terms of to tal tox ic ity con ce ntrati o ns e xpec ted in a natu ral re moval. Th is ca n be acco mplished by bloo m needs to be e xamined. c onvertin g toxin c on ce n tra ti on co So far, there is no "u n iversal" PA C sa xitoxin equival ent (STX-eq) con cencapabl e of rem oving all o f these problem tration. compo unds to the desired levels. Some Powdered activated carbon (PA C) is acti vated carbons may be more suitable o ft en consid ered as a treatme nt o ption for the re moval o f tastes and odours, fo r rem ov in g tastes a nd odo urs an d wh ereas o chers are superior for toxin cya notox ins as it can be very effecti ve re mo val. This is partl y due to th e differa n d easy to appl y. Newco m be and en ce in th e struc ture, functi o nality and D ri kas ( 1995) tested ten PA Cs for th e mo lecular weight of the compounds in adsorption of M I B and geosmin. Th ey question. M I B and gcosm in are re lati vel y raced th e PA Cs in order o f effecti veness sma ll molecular wei ght co mpo unds of fo r the rem oval of th e two compo un ds . 168 and 182 respectively, compared w ith They found that at equilibrium , the th e saxitox ins and the microcyscins th at ranking o f the carbons w as th e same fo r have mol ecular w e ights be tween 300th e two co m pou nds; ic, the best carbo n 500 and 909- 1 115 respectively. The aim for MIB re moval w as also th e best fo r of chis w ork w as to de te rmine the best geosmin re moval. Th e autho rs made the activated carbo n for th e adso rption of assumption that the kineti cs o f adsorpM 113 , geosmin , and the saxitoxins, and to tio n o f th e cw o co mpounds w ould also co mpare th ese results w ith previou s b e similar, and th ey tested the te n PACs work o n microcys tin- LR. at shorter contact times for the adsorptio n of MIB only. T hey found that the Materials and Methods ratings changed at shorter concact times, Powdered Activated Carbon and th e three PACs w ith a high volum e Six co mmercially- ava ilable activated of mesopo res (transport po res) raced ca rbons w e re used in this study. Be fore hi ghe r at a shorter contact tim e o f 1.25 use each PA C was dried at 105â&#x20AC;˘c for at h o urs than they did w ith a contact tim e lease 24 hours and aUowcd to cool in an o f three da ys. The authors assumed that airtight desicca tor. Table I shows the the same rating w o uld appl y fo r geosmin c haracte ristics o f th e PA Cs. adsorption. Howeve r, recent work by Bailey ct al. (1999) indi cated chat th e Carbon Characteristics kinetics of adsorption of geosmin may be Pore size distribution was dete rm ined quite diffe rent fro m that of MIB , and as outlined in Donati c t al. ( 1993) . that the most effec tive carbon fo r M IB Analysis re m oval unde r water treatment plan t G cosm in was an alysed usin g so lid conditions is not necessarily also th e best pha~e microextraction - gas ch ro macografo r geosmin removal. p h y / ma ss sp ec trometry (SPME PAC adsorpti o n studies of microC C / M S) . cystin-LR by D onati et al. ( 1994) and 14 C - labelled MIB was analysed using Sc humann et al. ( 1997) showed tha t the a liquid scin tillatio n counte r. amount of microcystin-LR adso rbe d at
\ RC' for \'(/Jft'."r Qu.-.luy .ind Trc:,umc:nl
TOUGH but economical
PUMPS for
Water Sewage Dosing Sludge
PCM Progressing Cavity Dosing
Svedala Robot Submersible
BBA Self-Priming We specialize in providing and supporting a wide range of pumps and pumping systems for demanding applications.
PUMP!iJwDf1DtJW
I
A S/ISO 9002 QEC 2977
20 Uralla Road Port M acquarie NSW 2444 Tel: 02 6581 0744 Fax: 02 6581 0790
Email: pumps@nor.com.au
WATER
QUALITY
AND
TREATMENT
expected, the equilibrium adsorp- reported by Negri et al. (1997) and tion o f the two compounds showed Velzeboer et al. (1998) . 60 minute con1act time Figu re 3 shows some interes ting similar trends. 120 minute contact time ,,G 10090 Figure 2 shows the percentage trends for saxitoxi n adsorption. Th ese > removal of th e two compounds at trends can be summarised:0 ,o E co ntact times that could reasonably (i) D oses higher than 30 mg/ L PAC 70 !! .E to be applied in most water treatment and/ or longer contact times are required E plants. The figure shows that the to co mpl etely rem ove aU saxitoxins. 0 •• G 40 kinetics of adsorption for the two (ii) Removal effic iency varies among the "'C .g 30 co mpounds is indeed different. For PACs tested . u 20 example at a contact time of 60 (iii) T he different saxitoxin analogues ,t 10 minutes HP is not effective for M lB show various removal efficiencies. T he S A•30 F ... 00 HP C A•10 while it is the bes t for geosmin general trend of sax itoxin remova l PAC removal. CA-10 was also effective appears to be GTX4 > STX > GTX2 = 60 minute con1act time for geosmi n re moval. Th e GTX 3 >C l = C2. Figure 4 shows that SA-30, P- 1100, 100 120 minute con tac t tim e m esopore volume of HP is 0.5 1 "g 90 crn 3/g and 0 .41 cm3/g for CA-10. and F-400 were the most effective PACs > ,. 0 Th e hi gh er volume of transport for reducing toxicity in terms ofSTX-eq. E f 70 pores allows quicker access to Trea tment with SA-30, P-1 100, and F00 to adsorption sites, h e n ce bette r 400 redu ced the toxin concentration iC 50 adsorption at shorter contact times. expressed as STX-eq to 0.17, 0.22 and ,2 40 ;; Th ese resu lts indi cate that 0.31 µg/L respectively. These values are ~ 30 LI. meso p ore volume may be an significa ntl y lower than the proposed 3µg 20 10 importa nt factor concerni ng th e STX-eq/ L health alert leve l (Fitzgerald successfu l rem.oval of geos min. et al. 1999). At a contact time of 60 Furth er experim entation with other minutes SA - 30, P- '1100, and F- 400 m esoporous PACs is required for PA Cs removed 66, 65, and 73 % of toxin va lidation of th is proposition . These as STX-eq respectively. Th ese results Figure 2 . Comparison of MIB and geosmin results along w ith th ose reported by show that a signifi can t amount of removal wit h 15 mg/L of each PAC Donati et al. ( 1994) for micro- sax it oxin , or, m o r e import antl y, cystin-Ll<.. adsorptio n ind icate that saxitox in equivalents, can be removed at Saxitoxins were quantified as ou tlined the same PAC could be appl ied to a contact time relevant to a water treatment plant wi th a reasonable dose of remove both geosmin and mi crocystinby R.ositano et al. (1998) LR . H owever, this PA C would not be PA C. Equilibrium adsorption Th e selection of a PAC for saxitoxin co nsidered the best one fo r the remova l l<..a w water was added to clean dry adsorption sh ou ld be based on the of Mm. Pyrex bottles to a level just below the top T able 2 shows how the choice of removal of the saxitox in analogues that to obtain minimum · h eadspace. The PAC ca n change when contact ti me and are most likely to be present and the required amount of MI13 or geos111in was the compound that is causing taste and desired reduction in terms ofSTX-eq. As added and the solution shaken and then odour problems are cons idered. For toxin profiles Australia w ide have not allowed to stand. After several minutes example , at a contact time of60 minutes, been found vary sign ificantly, the best PAC was added and the bottle scaled and HP wo uld be th e PAC of choice for th e method to assess a PAC vvould be to agitated for 3 days. removal of geosmin, while for conta ct spike test waters with extracted toxins Kinetics times above 60 minutes C A-10 would be from bloom material. Examining the Dry carbon was weighed o ut 24 hours chosen. Similarly fo r M 1B removal the adsorption of one sax ito xin analogue prior to use and prewetted with 3-5111 L best PAC cha nges fro m CA- 10 to SA-30 such as STX (the most toxic) m ay be of deioni sed water. M 1.8 , geos min or for conta ct times below and above 60 risky as it is a minor constitu ent of all the saxitoxins were added to a constantly min utes. When choosing the appropriate toxins present and successful removal of stirred raw water sa mpl e . After 15 PAC, se lection should be based on the th is compou nd may not mean a PAC mi nutes of mixing, a sample was taken to most likely operatin g conditions at the wi ll su ccessfully remove the dominant determine the initial conce ntration. PAC water treatment plant (contact time) and C-toxins and gonyautox ins. T h e was added an d samples were taken at the taste and odo ur co mpo und(s) of saxitoxin removal here cou ld not be related to the pore size distribution of the predetermined intervals. Samples were importance. PA Cs shown in Table 1. pressure filtered though 0.45 µm disposSaxitoxins able filters to remove the PAC. The results r eResults and Discussion ported here are for Table 2. Choice of PAC as a function of contact time saxitox ins that would M IB and Geosmin and taste and odour compound most likely be found Six PA Cs were tested for equilibrium in a bloom of toxic Contact Compound(s) causing taste and odour problem removal of M IB. Four of th ose carbons MIB MIB and geosmin time (min) geosmin Anaba ena ci rcinalis. were also tested for ad so rpti o n of T he saxi toxin profile CA-10 CA-10/SA-30 HP 60 geosmin (Figure I). The conditions for (F igure 3) was very SA-30 SA-30 CA-10 120 these experiments were PAC dose = 4 sim il ar to tho se mg / L , natural water, Myponga ~<..eservoir water, DOC = 10.7 mg/ L. As
..
.
30
WATER MARCH/APRI L 2000 \.RC: for \X'.itt:r Quality and Tr<:<llmcnt
-
STX
c. 2
O.2 !il 1,1 g fl
c. z
(3 .7 % )
-
GTX2
GTX4
c. • 0 ,0 1 1,1gfl
{0.1 % )
-
Ct
c.'"' 5,34 1,1 g /L (7 .0 "' )
GTX3
c. • 0.43 .,gll
{5.5'.4 )
-
C2
c.
11:
1.20
11 g fl
( 15 .4%)
0.55 11 g tl (61.4%)
10 0 Ill
>
80
0
E Q)
...
60
c Q)
40
Q)
20
...u
Q,,
0 HP
PAC
· No r11 ull f or G TX2 with HP PA C
Figure 3 . Saxit oxin removal from Hope Valley WTP inlet water (DOC= 5 .7 mg/ L) after treatment with 30 mg/ L of PAC and contact time of 180 minutes. Total saxitoxin concentration = 7 .8 µg/ L, concent ration and the rel ative abundance of individual t oxins is shown above.
Summary and Conclusions For M!B and geosmin rem o val the best activated carbo n d epended on th e co ntact time. For geosmin adsorption the w ood based PACs, HP and CA- 10, with a h igher vo lume o f transport po res allo wing quicker access to adsorpti on sites, perform ed very we ll at sho rt co ntact tim es. T hese types o f activated carbons were fo u nd by Donati et al. (1993) to be th e most effective fo r mi crocysti n LR removal. At lo nge r contac t times, several other carbo ns were superior. At very long contact times, the ranking of the activated carbo ns for geosmin and MIB remova l was sim ilar; ie, th e best carbon for M IB was also the best for geosm in . H oweve r, at shorter co ntact times o f u nd er two hou rs, the ranking was foun d to be different fo r the two compounds. N o ne of the PA Cs comp letely removed all o f the sax itoxins, altho ugh so me pe rfo rm ed better than o thers. T he most toxic, STX and G TX 2, 3 and 4 analogues were mo re easil y re mo ved than th e less toxic C - toxins (Cl and C2) . H igh er PAC doses or lo nger co ntact tim e would be required to completely 100
ai >
0
E
80
~
0"
60
'!I
X
I-
40
c ...u
20
(/)
Q)
Ql Q,,
0 HP
CA-10
SA-30
F-400
P- 1100
PAC •value for HP will be lower as GTX 2 could not be Included In th is calcula tion
Figure 4. Saxitoxin equivalent remova l with 30 mg/ L of each PAC after 180 minutes. Initial STX-eq = 1.1 µg/ L
PCO
WATER
remove all th e saxitoxins. Based o n reported levels of saxitoxins found in natural blooms PAC may be a viable option for the treatment of contaminated waters at moderate doses (eg 30 mg/L) provided there is sufficient contact time. The carbons displaying the best removal fo r these toxins were also the most effective for MIB removal. In summary none of the PACs was the best at removing all the compounds tested. At co ntact times of 2 ho urs or less, it was found that o ne activated carb on could be app lied to re1nove both geosmin and microcysti n- LR while that activated carbon would not be considered the optimum for the removal of MIB and saxitoxins.
References Bailey, D., Bowen , 13., Goodwin , L. and Procter, L. (1999). An organic removal strategy for Grahamstown WTP. Proceedings 18th A WW A Federa l Convention, Adelaide, 11 - 14 April 1999. Do nati, C. D., Drikas, M ., H ayes, R.. and N ewcombe G. (1994). Mi crocystin- LR adsorption by powdered activated carbon. Wat. Res., 28 (8), 1735- 1742. Donati, C. D ., Drikas, M. , Hayes, rt. and Newcombe G. (1993). Adsorption ofmicrocystin- LR by powdered activated carbon.
QUALITY
AND
TREATMENT
Wat. J. Aust. Wat. Wastewat. Ass., 20, 2528. Fitzgerald, D . J., C unliffe, D. A. and Burch, M. D. (1999). Development of health alerts for cyanobacteria and rdated toxins in drin king water in South Australia. Environ. Toxicol. Wat. Q ual. , 14(1) 203-209 H umpage, A. R. , Rositano, J. , l3retag, A. H ., Brown, rt., Baker, P. D., Nicholson, B. C. and Steffensen, D. A. (1994). Paralytic shellfi sh poisons from Australian cyanobactcrial bloo ms. Aust. J. Mar. freshwater Res., 45 , 761-771. Negri, A. P., Jones, G. J., Blackburn, S. I. , Oshima, Y. and Onodera, H ., (1997) . Effect of culture bloom development and of sample storage on paralytic shellfish poisons in the cyanobacterium Anabaena circinalis. J. Phycol., 33, 26-35 N ewcombe, G. and Drikas, M. (1995) . T he removal of2-methylisoborn eol and geosmin using powdered activated carbon. Proc. Australian Water and Waste Wate r Association 16th Federal Co nvent ion , Sydney, pp. 175-182. Rositano, J. , Nicholson, B. C., Heresztyn, T. and Velczeboer, R. M . A. ( 1998). Charact<c:risation and determination of PSP to xins in neurotox ic cyanob;icteria and m.:thods for their re moval from water. Urban Water l"l.. esea rch Assoc ia tion Aus trali a , R esea rc h l"l..eport N o 148, Melbourne. Sch umann, l"l..., Wong, S. H ., Pendleton, P. ,
Mulcahy, D. and Levay, G. (1997) . Activated carbon properties controlling removal of blue-green algae hepatotoxins from water. Proc . Australian Water and Waste Water Association 17th Federal Convention, Sydney, Vol 2 pp. 574-581 . Sivonen, K . and Jones, G. (1999) Cyanobactial Toxins. In Chorus, I. and Bartram J. (eds) Toxic cyanobactia in water: A guide to their public health consequences, monitoring and managem ent. E & FN Spon Publish ing, London, 41-11 1. Velezeboer, R. M. A. , Baker, P. D . and R o sitano, J. (1998) C haracterisation of saxitox ins produced by the cyanobacteria genus Anabaena in Aust ralia. Urban Water Research Association Australia, Research R eport No 135, Melbourne.
ACKNOWLEDGMENTS
T he authors thank Tom Woods, J en ny M orrall and J oanna Rositan o for their assistance with saxitoxin analyses. THE AUTHORS .
David Cook, Gayle Newcombe (a Project Leader) and Janina Morrison are with th e Australian Water Quality Cen tre, a party in the CRC WQT. David C oo k's e mail is <lave.coo k@ sawater. sa.gov .au
TOVEKO® Gravity Sand Filters TOVEKO® is a compact upward flow, automatic self cleaning sand filter suitable for tertiary, potable & industrial applications. Features include: • Compact, modular & low level design. + Can handle fluctuating flow and loads. + Simple installation on-site. • Simple & reliable operation. + Low maintenance & operating costs. + Sand is thoroughly scrubbed before being returned to the top of the bed. + Removal of TSS, BOD, COD, P, Fe, Mg oil & greases is achieved.
TRIANGLE FILTRATION 17 Hinkler Rd, Mordialloc, Vic. 3195 Ph: (03) 9580 2122, Fax: (03) 9580 3131, www.filtomat.com.au
WATER QUALITY AND TREATMENT
BIOFILMS - A STICKY SITUATION FOR DRINKING WATER? I Fisher, M Angles, J Chandy, P Cox, M Warnecke, G Kastl, V Jegatheesan R esearch Progra111 Fo11r deals 111ir/1 1nai11fai11i1t(.! water qua/ii)'
i11
disrri!J11tio11
s yste111s. Introduction A drinking wate r distribution syste m extends from a water treatme nt plant o u tl et to co nsum e rs' taps. H isto ri call y, distributio n systems ha ve been see n as a compl ex ne twork of pi pes, p u mps and se r vic e re se rvo irs d esig ne d to m ee t o perational flow re qu ire me nts su ch as mini mu m p ressu re a nd ma xim um variations in dai ly de mand. T his vie w may be c h a nging, howeve r, as it is apparenc that th e mi c robi a l bio ta in di stributi o n sys tems have a rea l impac t on drinkin g w a te r quality. Certainl y, the adverse effects which des ign dec isio ns may have o n the micro bi ology of dri nki ng wate r syste ms is poorly unde rstood; yet m icrobia l quality is a key targe t in gu idelines, re qu ires the use of disinfectants an d is n o w a basis for re vocatio n of wate r authoriti es' ope rating lice nces. Adva nces in water treatme nt processes make it po ssible to produ ce higher qu ality wate r but also put mo re strin gent requ ire m ents o n a distributio n system to sustain chat qu ality during de livery . T here is no w in c reased interest in improving tap w ater quality, through both managem ent of ex isting system s and design of m odifie d or new systems.
cost-e ffective manage me nt strategies can be deve loped to concrol wa ter quality and he nce m eet th e in creasin gly stringent quality guid elines be ing im posed o n wa ter supplies. A major fo cus of the program 4 agenda is bio films and de te rmining the ir effects on w ate r quality. Bi o fi lm form atio n arises from the in evitable colo nisatio n of sur faces by mi c ro o rga ni sm s
w he never a solid surface is in contac t w ith an aqueous phase . Utilisa tion o f di sso lved nutri e n ts lea ds to ba cte rial grow th , repl icatio n and th e formation of ex trace llul a r p o l y m e ri c su bsta n ces . Bio film cell den sity in creases through growth and ce ll deposition such ch at a mature biofilm contains a hi gh density o f bacte rial cells embedded in an exopo lymer m atri x. Biofilms o n th e surfa ce of
The Project R esea rc h Program 4 of th e C ooperative R esearch Ce ntre for W ate r Quality and Treatment (C R C WQT) is specifically concerned w ith de termining and predicting wate r quality characteristics, to assist in maintaining water quality wi thin distribution systems. It aims to understand th e m echanisms by which the maj o r components of a d ist ributio n sys te m such as bio films, path o ge ns, w ate r flow and influ e nt w ater quality inte ract within pipes and service reservoirs, thus affecting health-relate d and aesth eti c di mensio ns of th e water qua lity at consumers' taps. From this basis, so und
WATER MARCH/APRIL 2000 C RC for Waler Quality
and Trc:atmc:nt
33
WATER
QUALITY
AND
TREATMENT
"Cryptosporidium oocyst interaction s with 0.80
6.00
0.70
l
"'E 5.00
i 4.00
0.60
~ 3.00
0.40
:0
0.30
(") ,.,-
0.50 iil en
en en 0
§ 2.00
0.20
'5
ii5 1.00
a: C ~
3
(0
?
0.10 - -6-0- - 7--0_ _ ___ ______J 0.00 90 120 T (days)
O.OO L.a1._0_ _ 2__0_ ___ 30-- 4--0-
•
Biofilm control
Biofilm 0.1 mg/L acetate
Residual chlorine control
Residual chlorine 0.1 mg/L acetate
Figure 1. Biofilm development and resultant chlorami ne decay in drinking water biofi lms grown in unsupplemented drinking water (control) and drinking water supplemented with 0.1 mg/L acetate
dri nking water pipes contribu te grea tly to the quali ty of drinking water by redu cing the efficie ncy of disi nfection pro cesses, assisting in the protection and regrowth of microo rgani sms in retic ulatio n systems, a nd b y affectin g th e aesthetic quality of drinkin g water at th e customer's tap . T raditionally, ba cterial regrowth in distri bution systems has been inhibited by the use o f di sinfecta nts (mainl y c hlorin e and c hloramine) . At prese nt, th e re appears to be a divergence devel oping between compliance with more stringent microbiological water quality guidelines and the redu ctio n of disinfecta nt usage, to minimise both th e development of potentia lly hazardo us disinfect ion by-produ cts (DBPs) and capital and mainte nance costs asso ciated w ith disinfection. As a result, alternati ve treatm ent m ethods are required to minimise ba cte rial regrowth , and hence biofilm format ion. The minimisa tion of bac te rial regrowth in distribution syste ms w ill also assist in maintainin g effec tive disinfectant residuals throughout the system without th e n eed for inc reasing di sinfectant usage. On e way in w h ich bacterial growth may be controlled w ithou t excessive disi nfectant usage wou ld be to decrease nutrie nt levels in th e wa ter. Of th e th ree main nutrie nts - carbon , nitroge n and phosphorus - carbon is required th e m ost fo r bacterial growth and is considered the most likely nutrient to be limitin g in drinking water. O rga nic ca rbon also contributes th e precursors of DBPs, contributes to th e decay of drinking water disin fecta nts and contributes to colo ur. The re moval of organi c carbon from drinking water would , therefore ,
34
e nabl e the control of bacterial regrowth throu gh nutri ent limitation, minim ise th e production of disinfection byprodu cts, redu ce the loss of di sinfectant res iduals and contribute to th e redu ction of drinking water colour. Program 4 th erefore se t specific goals within three research areas: 4. l. Biofilm and flo w interactions under controlled conditions 4.2. Pathogen and biofilm interactions under controlled conditions 4.3. Management strategies for biofilms and pathogens in distribution systems. Curr e ntly , t wo proj ec t s are includ ed in area 4. 1, " Facto rs affectin g biofilm develop ment unde r controlled conditions" and " Mod ellin g bi ofilms and int e rv e nti ons"; on e in 4. 2
drinking water pipe biofilms", and two in area 4.3 - "Optimisation of chlorin e re sidual in a di stri bution sys te m M el bourn e" a nd " R eal-time water qu ality m odelli n g in a di stributi o n system - P erth " . A proposed proj ect within area 4.1 "Physical and c hemica l effects on distributio n syste m biofilms and incorporated pathogens" has recently been approved and is exp ected to start early in 2000. Factors affecting biofilm developm ent under controll ed conditions: Little is kn own about the leve ls of nutrients in Sydney drinking wa te r, gen erally, and the re have been very fe w studi es of biofilm formation in Australian distribution systems. C onsequently, the nature of biofilms and th e effects of nutri ents on th eir development are largely unknown. This study was conducted to investiga te th e effects of o rganic m ate rial o n biofilm grow th and the subsequ ent effects of developed bi ofilm on disinfectant residuals. Annular reactors were used to form biofi lms in drinking water in o rder to have so me control over environmental paramete rs w hi ch affect biofilm deve lopment , which cou ld not ha ve be e n achieved in situ. Outco mes from the study indicated that biofilm development was limited by organic carbon in Sydney drinkin g water. Bio film development also contributed considerably to disinfec tant loss within th e reacto r system s, partic ularly in the presen ce o f assimilable organic carbon levels above 100 µ g/L (Fi g. 1). lt is hypo thesised that biofilm s act as a biological catalyst , w h ich e nabl e the
Figure 2. Alice Gilyou using t he biofilm reactor developed to study biofilm development in waters containing very low organic carbon concentrations and disinfectants
WATER MARCH/APRIL 2000 CRC for \Xfatcr Qualit )' and Trc:;;nmenl
WATER
QUALITY
AND
TREATMENT
destruction
Biofilm Pipe Wall Figure 3. Elements and processes incorporated into a simple model to describes t he biofi lm growth and disinfectant decay in a distribution system
co n version of o rgan ic carbo n spec ies unable to react with di sinfec tants to those that arc able to. C hlo ramine is usuall y put forward as a m o re e ffi c ient d isinfecta nt aga inst biofilms than c hlorin e. A co mpari so n of c hlori ne and chlora m in e showed t h at c hl oramine wa s more
e fficient at re tarding bi o film developm e nt, using th e sa m e treated wate r so urce . N everth eless, bio film deve lopm ent was eas ily de tected, by a va ri e ty o f m e thods, in the prese nce of 1. 4 mg/ L ch loramine . R em oval of organic carbo n from th e water used , th rough th e use of
Nutrient Analyser
*Phosphate *Nitrate *Ammonia Can also include:
pH Conductivity Dissolved Oxygen Temperature Redox Turbidity
g ranu lated activated ca rbon filtration, produ ced a very stab le drinking water w ith enhanced disinfectant stabili ty and associated loss of biofilm development. To ca rry o ur the project successfull y, an a nnular reac tor w as develo ped w hi ch all owed the study of very lo w nutrien t water o ve r lo ng periods, w hile maintain in g stable disinfectant residual s (Fig. 2). T he re is no know n co mm erciall y avail able reac tor wh ich has these capabilities. In addi tion, a novel reprodu cible m ethod fo r th e dete rmin a tio n of assi mi lable organic carbon (carbon able to be easily used by bac te ria), based o n flu o resce nt stain s and th e nati ve bacte rial populatio n , was also develo ped. M o de lli ng biofilms and interventions: Alth o ug h mod els o f fl o w in distribution system s have been used for man y years, quality m o dels arc relati vely new and often in acc urate . This project is deve loping a d ynamic model of interactions be twee n bulk water and bio film devel o pme nt in a pipe elemen t, in cludin g th e m anagerial inte rventi o n of disin fec ti on.
See u Single & Multi-parameter 4-20mA RS232 SDl-12 Built-in Logger
Water Quality Instrumentation
G reenspan manufacture a range of rugged, reliable and hig hly accurate sing le and multiparameter sensors des igned spec ifica lly for long The Greenspan Nutrient Analyser is proven technology term unattended monitoring. Th ese sensors suit for the Wastewater industry. Unfiltered samples can be a wide ran ge of water qu ality monitoring drawn directly from Mixed Liquor. Monitoring points applications and are available with a broad can be either in-process or on final effluent. variety of outputs and communication s options.
1800 250 270 T E C H N O L O G Y
I
Quality Endorsed Company ISO 9001 LIC 3067 Standards Australla
Ph: +61 7 4660 1888 Fax: +61 7 4660 1800
www.greenspan.com.au
-
---
Monitoring Quality
WATER
QUALITY
AND TREATMENT
1n >,
pathogens ente rin g the syste m acc umu la t e or grow in pipe or se rvice reservoir bi o film s . Th e be haviou r o f th e parasite
3 8 0
C ryptosporidi,1111
7
Mean observed (n=6)
6
Theoretical
5
:5 a. :5 4 0
0)
g
in drinking wate r distributi on sys t e m s is n o t w ell defi ned. [n particular, interac tio ns betwee n
2
0 0
100
50
150
Time (h)
Cryptosporidium ooc y st s and biofi lms attac hed to p ipe surfaces h ave n o t bee n cl ose ly exami ned prev iou sly. T h e po ten tial ex ists fo r oo cysts to b ec om e attac hed to pipe biofi lm s, and to subseque ntl y detac h inte rmittently, or foll owing a change in syste m conditi ons. Disinfec tio n o f water w ith c hl o rin e is u n like ly to affec t C ryptosporidi11111 oocysts but may cause d e t ac hm e nt o f biofi lm c onta i nin g attached oocysts. This proj ect used a la b o rato ry sca le pip e rig made o f ex humed p ipe to study th e poten tial of Cryptosporidi11111 oocysts to accu mu late in re;il pipe b iofilms and subseq ue ntly be released to the water phase by bi ofi lm sloughing. The majority o f C rypt os poridi11111 oocysts intro duced in to the experim ental system were not su bsequentl y recove red in efflu e nt sampl es, which did not agree with theoretical efflu ent oocyst numb ers (Fig. 4) . It was hypoth esised that m ost of the oocysts not accoun ted for in effl uent samples we re probabl y attached to surfa ces with in the syste m , although these were not accounted fo r by direct exam inatio n of the su rface biofilm. T his m ay have bee n du e to he teroge no us di stri bu tio n , inaccu rate recove ry calc u lati on th ro ugh oocyst attach me n t and e mbedd ing, and oocyst destruc tio n and disintegratio n . O ocyst recove r y from wa te r s;i mp les was regard ed as ac c u ra te, with rec o ve ry typical o f this type of sa mple . Should surfa ce attachme nt and subsequ ent interm itte n t de ta c hm e n t be a maj o r fac tor in actual field condi tions, variability between samples ove r ti me wou ld be hi gh . C omposite samples taken ove r ti me or sp;ice wo uld be of mo re value than grab sa mpl es. Such sa mples would be o f m ost va lue ;it a few key
Figure 4. Log of actual Cryptosporidium oocyst numbers iso lated from t he effluent of a laboratory scale pipe rig compared to t heoretica l effluent oocyst numbers
Experime ntal resu lts from the proj ect described above are being compared ini t ia lly w i th pr o c ess d es c r ipt io n s co ntained in ex isting models. At th is sta ge, a simpl e b io fi lm m odel has been de ve lo ped base d o n exp erime ntal data derived from b iofi lm reactor e xpe rim e nts. T he m odel can be used to predict th e bacterial deve lop men t and disin fe cta nt decay in distribu tion syste ms. Th e m o de l i nco rp orates impo rta n t system p rocesses, su ch as attachm ent o f free bacteria to the bio fi lm o n a pipe wall surface, detac hm ent o f bacteria from the biofilm , growth of bio film bac te ria with chl o rami ne inh ibition , chloram ine deca y in th e bu lk water ph ase, and ch lo ramine decay due to bio fil m bacteria and wa ll surfaces (Fig. 3) . Further, the biological stabili ty of diffe re n t wate rs ca n b e predicted by the mo del through the es timatio n of th e co n ce n t ra tio n o f organ ic substances. These developme nts will resu lt in a robust mathe mati cal desc ription of exp erimental results o f know n acc uracy, prio r to in corporation into a netwo rk flo w / quality model. A m odel of th is ki nd is a useful too l in developin g system manage men t strategies su ch as limi tin g th e input of nutri ents fro m treated water, optimisin g the disin fec tant regim e fo r co ntrol o fb io fi lm activities, adoptin g regimes fo r rap id wate r m ovemen t through th e system, as well as comparing the effects of th e type a n d ti min g of m ain s cl ea nin g, to u ltima t ely improv e d r ink in g wa te r quality . Cryptosporidi11111 oocyst inte rac tio ns w ith dri n k in g wate r p ip e bio films: M icrobial h ealth risk may be substan tially modified w ith in distribution syste ms, if
~-:-, 36
WATER MARCH/APRIL 2 000
(~ ) CRC for ~ Inter Quality .:ind l i-e atm<.'ll l
co ntrol points o f a system . Should variation in real water samples occur as in the e xp eri m en ta l system , the se tting of arbitrary acceptable le vels o f oocysts in di scou r age d . w ate r sh o u ld be H e te roge nous distribu tio n and uncertain ty with oocyst recovery make ad- hoc bio fi lm or sedi me nt samples of lim ited value . T he in teracti on o f shea r stress and oocyst de ta c hm e nt and res usp ensio n remain s incomple tely understood. It is also uncle;i r as to how lon g attac hed oocysts ca n re main viable and infective ;ind, once no n- viable, how long they can be de tected . D esirable techni cal adva nces wou ld in clude im prove d isolatio n of Cryptosporidi11111 from surface m ate rials su ch as bio film or sedim e nts. Furth e r e xperim e ntati o n and development to ;ichieve these obj ectives is proposed in a new proj ect to improve o ur understandin g o f C ryptosporidi111n be havio ur in drinki ng water distribution syste ms, an d co n se qu e ntl y d eve lo pi n g in fo r m e d manage rial strategies to deal with distribution syste m com promise. Optim isation of c hlo rine residual in a distribution system - M elbou rn e: This proj ect wi ll validate th e me thods and res ul ts from the lab orato ry b iofi lm studies, by sa mpling and analysin g both bi ofilms and bu lk wa te r fro m a real di stribution sys te m , over a summ erwin te r period . T hree different methods of obtain ing biofi lm samples will be co mpa red directly. In additi o n , th e bio film / disin fec tion models developed in " M odelling bi o fil ms and inte rventions" will be valid;ited against the wa ter quality and biofi lm data obt;iin ed. This is necessa ry b e fo re u si n g th e m to d e ri ve ma nage n1e nt strategies, such as co ntrolling chlorin e resid uals to mee t coli fo rm and D B P gui delines th ro u gho u t the distrib u tion syste m . As flu c tuation s in flow can h ave substantial impact o n measured qu ality, fl ow time series fo r eac h p ipe w ill fi rst be deri ved from ex isti n g pip e netwo rk m o dels . Th ese flow s wi ll the n b e provided as inputs to th e quality models, to produ ce th e predicted qua lity for direct comparison w ith me;isured data du ring va lidatio n. A range of realistic flows w ill al so be gene rated from the n e two rk mod e ls fo r ev;i lua tion of managem ent strategies. Conclusion
Th e proj ec ts under C R C WQT Progra m 4 have all indicated t ha t b iofi lrns co n tribute de tri m e nta lly to
water quality through contributions to disinfectant decay and pathogen incorporation. The primary driver of biofilm develop ment is clearly the p resence of organic carbon, which is seen to be the k ey component influencing quality in water distribution systems. Further work in the program w ill involve studying the different physical and chemical param eters, whi ch may affect bi ofiJm recruitm.ent of pathogens and su bseque nt detachmen t. [t is also envisaged that th e quality model developed from experimental data obtained from o t her research areas w ithin the program will be incorporated in to a network m odelling package that w ill be read ily ap plied initiall y t o management of disinfec tion in distribution systems. T his is to address the cu rrent inadequac ies of network models to predict water quality. A complementary joint project with WA Water Co rporation in Perth has
commenced, w hich will explore the feasi bi lity of real-time modelling o f disinfection fo r operati onal control in a system supp li ed from groundwate r. Through p rojects such as those in Melbourne and Perth, the sound process understanding developed in the laboratory and process model studies is being transferred to assist managers of water quality in real distribution systems. Acknowlegements
We ackn ow le dg e th e ot h er Cooperative Research Centre for Water Q uali ty & Treatment project partners invo lved in the above projects: M ary Dri kas and Naomi W ith e rs from Au st ralian Wate r Qua li ty Centre, Adelaid e; Nataly Orr and Kevin H ellier from M elbourne Water Corporation; and Ca rl Rouhiainen fro m WA Water Corporation, Perth. We thank Alice Gi lyo u and Hopi Yip, the analysts with A WT ES&T involv e d in biofi lm research for Program 4 .
Authors
Dr. Ian Fisher is a Prin cip al Co n su ltan t with Australian Water Tech nologies Environ m ent, Science & Tech n ology (AW T E S&T) and Coordinator of CRC WQT Program 4. Dr. Mark Angles, Dr. Joseph Chandy, Dr. Peter Cox and Malcolm Warnecke are senior consul tants w ith A WT E S&T involve d in b iofi l m research and project developm ent for P rogram 4. George Kastl and Veeriah Jegatheesan are sen ior consultan ts w ith A WT ES&T involved in the development of models and projects in rea l systems within Program 4. Contact Address: Austral ian Water T echnologies - En vironment, Science and Techno logy, PO Box 73, West Ryde, NSW 2114. T el: 02-9334 0938 Fax: 02- 9334 0840 Emai l: ian.fisher@ awtpl.com .au
AquaMaster The next generation electronic water meter • Battery or Mains Operation • Bi-Directional as standard • Buriable Sensor [sizes range from 15mm-600mm]
• Submersible Sensor & Electronic Display Unit • Help to reduce leakage/ unaccounted for water
ABB Instrumentation
ABB Automation
Sydney: (02) 9540 0000 Melbourne: (03) 9837 1000 Brisbane: (07) 3848 6123 Tasmania: (03) 6344 911 0 Adelaide: (02) 8279 3000 Perth: (08) 9240 0600 Email: A88 .instrumentation@au.abb.com Website: www.abb.com
jl ••••
11111•••
WATER MARC H/APRIL 2000
37
WASTEWATER
~
QUEENSLAND COASTAL STPS RJ Thomas, EA Gardner, GA Barry, HN Chinivasagam, PE Green, AV Klieve, PJ Blackall, GW Blight, BJ Blaney Abstract The microbiological qu ality of the eilluent from 33 sewage treatment plants (STPs) along the Queensland coast was investigated as part o f a project to develop a quantitative model to assess the health risks assoc iated with irrigating sugar can e w ith treated emuent. Initially a sin gle sample from each of the 33 STPs was tested to provide an indi cation of the microbiological quality of effluent available for irrigation use. T e n of these STPs, selected to give a range of e illuent quality and treatm ent types, were then tested 4 to 6 times over a 12 m onth period to in vestigate in - plant variatio n. Microbia l counts var ied widely between and within STPs. Ran ges for all 92 samples (with geome tric m eans in brackets) we r e: t h er m oto lerant colifor m s, 0 to 24,000,000 (388) / 1 00mL; Clostridi11111 pe,jri11ge11s, 0 to 15,000 (90) /100mL; coliph ages , 0 to 6,370 (4) / 1 00m L; Cryptosporidi11111, 0 to 600 (<10) /L and Cinrdin, 0 to 4,840 (98) / L. All of the 10 STPs w h ich were sampled over 12 m onths had Cinrdin cou nts above 150/L in at least one sample , and one STP had co unts above 1 ,000/ L for all 5 sam ples. No seaso nal trends were evident in the coun ts of any of the microorgan isms assayed , and no relationship was evident between micro-
38
WATER MARCH/APRIL 2000
bial count and STP type, population size o r STP throu ghput. None of the statistica lly sign ifi ca nt relationships between assays reported in this paper were strong enough fo r the reliable prediction of the results of one assay from th e results of another. Key Words Sewage emuent, indi cator o rganisms, pathogens, wastewater reuse, irrigation , sugar can e, thermotoleran t colifo rms,
Cinrdin Introduction U se of secondary-treated e fflu ent for spray irrigation of sugar cane is an attractive proposition for Queensland cane farmers. Effiuent co mpares favourab ly in cost with other sou rces of irrigation water, and has some nutritive valu e for th e cane , whi le ca ne fields may prov ide an environm entall y safer site for disposal of the emuent compa red w ith direct o utfa ll to water bodies (Ga rdner et al, 1 998) . Disease-causing orga ni sms (pathogens) persisting in the secondarytreated emuent and subsequentl y transfe rred to t he cane are eli minated during processing of the suga r. H owever there may be an increased risk of infection to farmers and their fa mili es and ne ighbo urs who come into direct conta ct w ith the effluent and it s ae r oso ls, and to
cons um e rs of vegetable and sa lad crops g rown ne ar eff1 u en t - irrigate d farms (M ara and Cairncross, 1989) . This risk is difficult to assess, althou g h empiri cal evidence and epidem iologica l in vestigati ons suggest the risk is smaLI (Fattal et al, 1986, Shu val et al, 1989). Effiuent-borne pathogens wh ich cause human infections t hrou gh direct contact include bacteria su ch as Sn /111 011elln, Shigelln and Eschcricl,in coli, the protozoa n parasites Cinrdin and Cryptosporidi11111, parasitic worms (he lminths) and e nte ric viruses (l3itton, 1994) . Enteroviruses, coliphages, coli fo nns and Snlmo11elln ha ve been de mon stra ted in aerosols form ed d uring spra y-irrigation of cro ps and recreational areas w ith sewage eilluent (M cN e ill, 1985). Mi crobio l ogica l monitoring of Queensland STPs is largely restricted to assaying faecal co liform levels, w ith o nl y a few ST Ps testing fo r bacteriophages, enterococc i, Sal11lo11elln, Cinrdin and/ or Cryptosporidi11111. Frequency of testing, source of testing (STP , loca l government, state gove rnmen t , hospital or private laboratory) and test m ethods used vary markedly between STPs, and the small amount of data available is of little use in comparing effiue nts from di fferent STPs, o r in quantifying health risks associated w ith th e ir re use.
ASTE
Most regulations and guidelines controlling the reuse of wastewater specify acceptable levels of thcrmotolcrant coliforms for various practices, and usually do not specify levels of protozoans and viruses (Table I). This is an acknowledgment of the difficulties involved in assaying the latter, rather than confidence in the predictive value of the indicator assays. The inability of assays for traditional indicator organisms such as coliforms, thcrmotolerant (faecal) coliforms and coliphagcs to accurately predict levels of the majority of the faccally-transmittcd infectious agents is well documented (Bitton, 1994). However, in the absence of practical and specific methods for routinely assessing pathogen levels the traditional methods remain useful, primarily to monitor the disinfection process. The thermotolerant coliform count also provides an indication of the likely levels of pathogenic enterobacteriaceae present in an effiuent. The coliphage count is useful as a conservative indicator of viral disinfection (Havclaar, 199 !). The work reported here was done as an initial part of a project to develop a quantitative model to assess the health risks involved in spray-irrigating sugar cane with secondary-treated efl-luent. Methods
To assess the microbiological qu:1lity of efl-luents available for sugar cane irrigation, we determined the levels of selected microorganisms in the effiuents from 33 sewage treatment plants (STPs) near sugar-producing areas in Queensland in 1995/9<1, induding some samples from stabilisation ponds receiving secondary-treated efl-luent. Ten of these STPs, selected to provide a range of cffiuent quality and STP characteristics, were then sampled on several occasions during 1996/97 in order to investigate in-plant variation and to determine whether any seasonal efl-Ccts could be detected. Microorganisms assayed included thermotolerant coliforms, Sal111011c/la, Clostridi11111 pqji'i11grns, the protozoan parasites Ciardia and Cryptosporidi11111, hclminths (parasitic worms) and coliphagcs of E. coli strain K13. For the initial, single sample survey, 33 STPs were chosen on the lxisis of their proximity to sugar-producing areas, the feasibility of using their efl-luents for irrigation and to give the widest geographical spread over the length of the Queensland coastline involved in sugar production, from l6°S to 28°S. Sites are identified in Tables 2, 3 and 4
by number only, from l (the most northerly) to 33 (the most southerly). Trc,ltment processes used induded oxidation ditch, biofiltration, Imhoff tank, activated sludge and nitrogen reduction. Chlorine disinfection was used by 27 of the STPs, one used ultraviolet irradiation and 5 had no final disinfection process. Equivalent person (EP) capacities ranged from 1,500 to 240,000 and flow rates from 0.4 to 200ML/day. This initial survey was done to provide an indication of the quality of efl-luents available for irrigating sugar cane, rather than to provide a precise estimate of the average efl-luent quality from these sites. For the multiple analysis survey, IO of the STPs were chosen for regular sampling on the basis of STP and effluent characteristics in the initial survey, to provide low to high microbial counts, 5 diflCrent STP types, a range ofSTP sizes and use and non-use of chlorine to disinfect the cffiucnt before discharge from the STP.
Four of the STPs, including one of the 10 from the second survey, disdurged into holding ponds before final discharge. Samples of the influents/ effluents from these ponds were also tested. Sa111ple collection and preparation. Samples were collected from the outlet stream of STPs (before ponding) and from pond outlets into sterile polyethylene containers. One litre was collected for bacteriology (with sodium thiosulphate solution to 0.1 % final concentration to remove residual chlorine) and 2 L for parasitology and virology (without thiosulphate). Samples were held on ice during transport to the laboratory. 13acterial counts were done within 24 hours of collection. Samples for parasitology :md virology were held at 4°C until tested. Bacteriology. Appropriate volumes of sample, and appropriate ten-fold dilutions wen.' filtered through 0.45~tm filter discs (Millipore). Faecal coliforms were determined on m-FC agar (Difeo)
NHMRC May 1995 USE STANDARD Injection to aquifer < I 000 thermotolerant coliforms/100 mL Unrestricted public access, urban " .. < 10 14 Controlled public access, urban < 1000 " Spray irrigation of crops <10 Pasture irrigation < l000 " Dairy cattle pasture <200 "## <10 " Aquaculture < lO Primary contact recreational < 150 Non contact recreational < l000 " Ornamental water bodies < l0000" VICTORIAN EPA 1996 USE STANDARD Unrestricted public access < 1 thermotolerant coliforms/100 mL < 2 virus/50 L < 1 parasite*/50 L Unrestricted access but no food crops < JO thermotolerant/100 mL Restricted public access < !000 thermotolerant coliforms/100 mL NSW RECYCLED WATER COMMITTEE 1993 USE STANDARD Unrestricted access - urban use < I thermotolerant coliforms/100 mL < 2 virus/SOL < I arasite* /50 L INTERIM GUIDELINES FOR DECLARED WASTE WATER DNR 1996 USE STANDARD Unrestricted public access < 10 thennotolerant coliforms/100 mL Urban with controlled access < 150 Agriculture (fodder crops) < l000 " Agriculture (unprocessed food crop) < IO Agriculture (processed food crop) < 1000 " "
.'
.
#
5 days withholding
## no withholding period
* 11 Parasite1t includes C1Jplosporidium and Giardia oocysts/cysts Table 1. Extracts from National and State Guidelines
WATER MARCH/APRIL 2000
39
WASTEWATER
Crypto· Giardia Salmo11ella' C.perfringens Phage E.coli Thennosporidium tolerant Colifonns /L /L / 100ml / 100ml / lOOmL / 100ml 1100ml MUd < 10 20 15 0 0 Cl 0 0 2 1 OD < 10 300 240 0 0 11 0 Cl I 2 OD <5 190 0 180 0.7 2800 600 C l 14 BIO 3 <5' <5 0 2 0 I I Cl I 4 OD 100 5 1004 300 11 0 430000 1050000 No 14 BIO 5 3080 120 68 2000 0 129 30 Cl 15 BIO 6 1010 < 10 0 90 0 0 Cl 3 I IT 7 760 <20 0 120 0 6900 0 Cl 2 BIO 8 680 <20 10 2.3 500 1900 13600 UV 9 BIO 9 40 10 2 0 70 480 390 NA No OD 10 1530 49 390 0 0 3200 3200 Cl 3 BIO II 100 < 10 0 0 0 34 34 Cl 1 12 BIO 20 I 10 0 0 1400 0 2 Cl BIO 13 30 < 10 5 0 0 4 0 I Cl BIO 14 170 11 10 0 0 0 3200 20 Cl BIO 15 110 30 < 10 0 0 0 600 2 Cl OD 55 16 NT NT' 6 1430 0.9 80000 54 0.6 No 20 OD 17 NT 13 NT 14000 0 15 9 3 Cl 134 BIO 18 NT NT 5 440 0 4 I 6 Cl Ox/BIO 380 19 NT NT NT 2900 14 8 0 7 Cl AS 300 20 50 < 10 0 0 0 0 0 3 Cl BIO 150 21 <5 <5 480 0 15 0 59 Cl 3 OD 100 22 85 70 0 40 0 0 20 2 Cl NR 100 23 <5 10 0 160 0 40 0 15 Cl 24 AS 850 I I 10 3500 6000 0.3 1570 50 Cl I 25 IT 30 105 120 5 130 0 2120000 0 I No 22 OD 26 I 2 10 320 6 0 0 Cl 0 5 AS 150 27 30 30 I 1200 7 0 7 11 Cl AS 350 28 NT NT 367 NT 3.9 40000 200 No 100000 AS 6000 29 < 10 50 10 0 0 0 45 Cl 0 2400 AS 30 < 10 10 6 0 .0 0 9 Cl 0 OD 530 31 NT NT 50 8 15 0 22 Cl 5 BIO 220 32 10 < 10 < 0 0 0 0 Cl 1 0.4 15 AS 33 OD: oxidation, BIO: biofiltration, IT: Imhoff tank, OX/BIO: combination OD and BIO, AS: activated sludge, NR: nitrogen reduction Cl: chlorination, UV: ultravio let irradiation Most Probable Number, 0 = <0.3/ IOOmL Limit of detection varies with turbidity of sample NA: not available NT: not tested
Plant Number
• • ' • ' r
Plant Type•
Equiv. Person Capacity xlOO 80 25 500 75 550 60 30 60 NA' NA 100 15 50 38 550
Flow
Effluent Disinf .•
Table 2. Survey of 3 3 sewage treatment plants on the Queensland coast
incubated at 35°C for Sh as a resuscitation step, followed by 44.5°C fo r 18h (AP H A 1992). A representative number of coloni es on each filter disc was identifi ed using th e Microbact- 12 system (Medvet) whi ch identifies E. coli to the species level and Klebsie/la, Enterobacter and Citrobacter to the genus level, with some isolates identifiable to species level. C. peifringens numbers were determined by the method o f Bisson and C abelli (1979) as mo dified by Armon and Payment (1988) . Most probable numb ers of salmonellae were estimated in the initial, single sample survey by resuscitating th ree lO0mL (co nc entrate d by membrane filtration), three 10ml and three lmL aliquots of the sample in 200ml, 90mL and 9mL volumes of buffered peptone w ater respec tively, follo wed by selective enrichment in Rappapo rt-Vassiliadis broth at 42°C and Muller-Kaufmann tetrathionate broth at 37°C . T hese w ere su bcultured to BGSA and XLD plates after 24h and 48h of incubation. Presumptive colonies were confi rmed using Kahn's 2- tube medium 40
WATER MARCH/APRIL 2000
(Oxoid) and agglutination w ith polyvalent typ e O antiserum (Murex). In the multiple sample analysis of 10 STPs, the presen ce or absence of Salmonella in 100ml of sample was determin ed by res uscitation in 900 ml of bu ffe red peptone water fo ll owed by selecti ve enrichment and culture for Salmonella as described above. Parasito logy. For the detectio n of helm.inth and protozoan parasites, l L of eilluent sample to wh.i ch 0.01 % Tween 80 was added was centrifuged at 3000g for 10 minutes and the resultant pellet was resuspended to a volume of 1 to 2 mL, depending on the amo unt of residue. H elminths were detected under light microscopy by their characteristic size and shape in two 100µL aliquots. Giardia cysts and Cryptosporidium oocysts were simultaneously detected by staining two S0µL aliquots with a conunercially available di rec t im mu n o flu o resce nt tes t kit (Cellabs) prior to examination with an O lympus BH2 epifluorescent m.icroscope. Virology. B ac t e r io ph ages we re analysed acco rdin g to G rabo w and
Co ubrou gh (1986) a nd Bowd e n (personal commu nicatio n). O ne mL of CaCl2 .H 20 (13%) was added to 100ml sewage efflu ent. T o this sample SmL of an overnight culture of E. coli K1 3 ATCC 15766 (host bacterium) cultured in Tryptic Soy broth (Amyl) was added and mixed well. T his n1ixture was then combined with 100mL of molten Phage Agar (Amyl) containin g 1% triphenyltet razolium c hloride, gently mixed , po ured in 50ml aliquots into petri dishes (1 45mm) and incubated at 37°C for 24h . T he prese nce of phage particles was indi cated by cleared zones (plaqu es) in an otherwise continuo us bac terial lawn. Plaques w ere enumerated and expressed as plaque fo rming units (pfu) per 100ml of effluent sample. As a positive control, E. coli K13 was plate d with a sampl e known to contain virulent phages. Statistical methods . D ata were con1bined from the initial and follow-up surveys including the ponded efilue nt discharge samples, giving a total of n = 92 samples for data analysis (Tables 2, 3 and 4) . All test results shown as below
ASTEWATER
Plant Number
Date
5
Thennotolerant Colifonns
ÂŁ,coli
Salmo11el/a
C.pe1:fri11gens
Phage
/IOOmL 480000 1080000 24000000 119000 153000
/JOOmL 220000 150000 6000000 119000 99000
+/- in IOOmL
/IOOmL
14/05/96 31/07/96 21/08/96 9/10/96 29/01/97
+
700
+
1000 6500 380
/JOOmL 189 115 6370 1515 25
26/06/96 7/08/96 25/09/96 27/11/96 4/03/97
26000 I 65 200 7
5000
10000 490 180000 2070000
5000 90
8 8
9/05/96 24/07/96 3/09/96 26/11/96
II II II II II
14/05/96 23/07/96 28/08/96 26/11/96 5103/97
13 13 13 13 13
2/07/96 27/08/96 8/10/96 20/11/96 12/03/97
15 15 15 15 15
5 5
5 5 7 7 7 7
7 8
8
a
b
2
+ +
2
900 4200 2400 940 710
0
54 200 0
NT 2070000
+
300 970 880 15000
0 2
2 0 2
150 61000
150 27000
1300 550
600 5 60
0
0
9900
2700
2100 1000 570 210 630
8/05/96 16/07/96 21/08/96 1/10/96 4/02/97
95 4800 1200000 540000 148000
65 2400 350000 60000 92000
80 2000 2900 220 1140
19 19 19 19 19
14/05/96 23/07/96 27/08/96 8/10/96 29/01/97
43
43
2
0
640
530
6 0
3 0
20 20 20 20 20 20
16/04/96 2/07/96 3/09/96 1/10/96 27/11/96 I 1/02/97
260 16000 380 870 27000 150
50 8300 60 80 27000
24 24 24 24 24
13/05/96 30/07/96 3/09/96 1/10/96 5/02/97
65
26 26/06/96 26 6/08/96 26 9/09/96 26 15/10/96 26 12/02/97 Limit of detection varies with NT: not tested
9
0
I 21 4
1600 46 960 1000 8
+
+
+
200 1500 420 I 0
0 0 0 0
0
421 0
218 940 0
0 0 0 0
5 0 0 0
0 0 0
591 I 1067 0 2
0 0 0
94 140 5
0
110 2800 1000 1040 200 300
0 0 0
120 150 240
0
21 0
0
530000 40000 2700 0 10600 10600 3300 0 73000 45000 turbidity of sample
+
162
96
130 0
0
5 0 0
700 70 36 I
29
0
3
0
I 0
Cl),ptosporidium /L 10 20 <301 <20 <20
Giardia /L 40 110 390 120 460
<10 <20 <20 40 <10
1390 2580 4840 2260 1120
<20 <30 <20
2850 600
NT"
720
NT
<20 240 <10 <20 <20
480 220 10 380 320
<10 <10 <20 <20 <20
JO 70 560 100 360
<20 10 <JO <20 80
200 20 220 340
10 <20 <JO <10 <10
160 520 80 <10 10
<JO 20 <20 <JO 20 <10
140 320 520 270 1060 <JO
<JO 60 <10 <20 <20
30 80 140 280 60
60 <10 <20 <10 <10
180 80 100 <10 <10
260
Table 3. Multiple sample analysis of 10 sewage treatment plants
detectable limits were taken to be zero, while samples not tested were regarded as missing values. Simple linear correlation coefficients were calculated among assays after transformation to the logarithm 10 (concentration + 1) scale. For the binary valued Sal111011ella assay, mean thermotolerant coliforms m samples positive for Sal111011ella were compared with negatives by t-test after transforma-
tion to logarithm 1u(concentration + 1). Results Microbiological assay results for the single sample survey of 33 STPs are presented in Table 2. The results for the multiple analysis survey in which 10 of the STPs were sampled at regular intervals for 8 to 10 months are presented in Table 3. Results for ponded effiuents
from 3 STPs sampled once, and one STP sampled 6 times are presented in Table 4. A high variability was seen between and within the STPs with respect to the numbers of the assayed microorganisms m the secondary-treated effluents. Ranges for all 92 samples (with geometric means in brackets) were: thermotolerant coliforms, 0 to 24,000,000 (388)/!00mL; C. perfri//ge//s, 0 to 15,000 WATER MARCH/APRIL 2000
41
WASTEWATER
Plant Number
Date
&
Thennotolerant Colifonns / IOOmL
Giardia
0 29 0 0 I I 0 0 239 3 0
120 600 60 <20· < JO NTb <20 <20 < 10 < 10 < 10 NT
105 850 180 80 80 NT JOO <20" 10 < 10 < 10 NT
0 0
0 0
< 10 NT
50 NT
1430 0
6 I
NT NT
NT NT
6000 4
1 60
110 NT
3500 NT
Salmo11ella
C.perfringens
Phage
/ IOOmL
+/- in IOOmL
/IOOmL
/ IOOmL
130 II 700 0 70 17 36 2 I 4 0 0
5
Effluent Stream 26 in out in out in out in out in out in out
11.10.95 11.1 0.95 26.6.96 26.6.96 6.8.96 6.8.96 9.9.96 9.9.96 15.10.96 15.10.96 12.2.97 12.2.97
2 120000 35000 530000 3 10000 2700 130 10600 3300 9 73000 0
0 0 40000 0 0 0 10600 0 0 9 45000 0
21 in out
16.4.96 16.4.96
0 87
0 34
17 in out
3 1.8.95 3 1.8.95
80000 0
54 0
55
Cryptosporidium /L
E.coli
+
/L
25 10.10.95 1570 m 1800 out 10. 10.95 • Limit of detection varies with turbidity of sample b NT: not tested
50 1800
Table 4. Analysis of ponded effluents from 4 sewage treatment plants
(90)/1 00mL; co liphages, 0 to 6,370 (4) / 100mL; Cryptosporirli11111, 0 to 600 (< 10)/ L and Ciarrlia O to 4,840 (98)/ L. H clmin ths were not detected in any sa mples. The surve y was not d esign ed to q u antitatively test for the effects of loca tion of STPs, STP types, equi valent pe rson (E P) capacities, daily flow rates o r e fflu e nt d isin fection w ith c hl orin e. Other than a significant (P <0.05) effect of eillue nt disinfection with chlo rin e on th e rmot ole ran t co li fo rm counts (geometri c means 130 with chlorin e, 13,370 w ithout chlo rin e), no ne of the other STP attri butes appear to be impo rtant. In the initial survey, 11 of the 33 sa mples co ntain ed more t han 1,000 therrnoto lerant co li forms/ I 00ml and wo uld therefore have been considered unsuitabl e fo r spray irrigatio n of sugar
C.pe1:fringens Coli phage C1yptosporidi11m Giardia
Thermotolerant Coliforms 0.266* {n=9 I1) 0.594** {n=9 l) 0.122 0.230*
cane under most g uidelines. On ly 4 of th e I I contained m o re than I ,000 E. coli/ I 00111 L, w ith th e o th er 7 sa mpl es containing large numbe rs of Klcbsiella. Sal111011clla was detected in 6 of the 33 samples, all w ith th errnotolerant colifo rm counts above 1,000/ I 00111L and 3 with E. coli counts above 1,000/ I 00m l. Of th e 6 STPs not using chlorin e to disinfect their effl ue nt, 5 (83%) had therm oto le r a nt co liform co unt s above I ,000/ l00111L, co111pared w ith 6 (22%) out of27 STPs using c hlo rin e. Sal111011ella was isolated fro111 4 of the 6 non-chlorinated effluen ts and 2 o f th e 27 c hlorinated e fflu e nts. In th e second , 111ultiplc sa mpli ng survey of IO STPs, 2 were consistently over, and 2 were consisten tly un der, 1,000 th er111otole rant co li forms/ I 00111L. T he o th ers varied w idely fro111 sampling to sa mplin g. Counts were consistently
C.12.er[J:i11ge11s 0.268* (n=90) 0. 11 8 0.446**
ColiQhage
0.124 0.144
CrJ!J2.IOS/2.0ridium
0.298*
•• Significant at the I% probability level (P < 0.01) • Significant at the 5% probability level (P < 0.05) n = 80 except where indicated Table 5. Matrix of simple correlation coefficients between organism concentrations , after transformation to log10 (concentration + 1 )
42
WATER MARCH/APRIL 2000
high in the 2 non- chlorinated STPs. H owever 6 of t he 8 STPs using chlorine also h ad ve ry hi gh the r111 oto lera nt coliform counts inter111ittently. A nonc h l o r i n a ted ST P was positive for Sal111011ella in 4 o ut of 5 sa111p lings. H owever Sal111011clla was isolated from 4 of the 8 ch lori nated STPs on one (3 ST Ps) o r 2 ( I STP) occasions. M ean thermotolcrant co li form coun ts were sign ificantly h igher (P<0.05) in sa111ples positive fo r Sal111011ella com pared w ith negative sa mpl es (geom etric means 76,560 and 140, respectively). Significa nt correl atio ns were also fou nd between t her111otolerant co lifo rm co un ts a nd counts fo r co liphage (P<0.01), C. pe1fri11 J!<'IIS (P<0.05) and Ciarrlia (P<0.05), and betwee n C . pe,fri11gens co u nts and cou nts for col iphage (P<0 .05) and Ciarrlia (P<0.0 I) (T able 5). I 11 more than hal f of the samples in which therm otoleran t coliforms were detected (48 of 80, I sample not tested fo r E. coli) E. coli accounted for less than half of the t he rmoto lera nt co liform count. T h is was m ost often due to high numbers of Klcbsiella spp ., pa rticularly K. p11e11111011iac . Ot her non -E. coli thermotolerant co li forms regularly contrib utin g to the thermoco le rant coliform co unt included E11tcrobacter
aJ!,(.!lo111em11s, E11tcrobactcr cloacae, Citrobacter _fre1111rlii and Scrmtia liq11~facic11s .
WASTEWATER
Discussion
On the basis of these results and c o n sidering curr ent g uid e lin es fo r thc rmotolcra nt coliforms in wastewater u sed for irrigation (Q ld D ept Natural flcsources, 1996), I I (33'¼,) of the 33 e muc nts in the single samp le su rvey would have been unacceptable for spray irr igation of sugar ca ne . Of the IO ST Ps in the mu ltiple sa mple survey, 2 fail e d th e thcrmotolerant col ifor m g uidelines o n all occasio ns they were sampled, and an othe r 2 fo il ed on most sam pli ngs. Th e va riability o f results bet ween and w ith in ST Ps re flects the variability in the m icrobial populati o n in th e incomi ng sewage, va r iati o n in the performance of a plant typ e com pared w ith the other plant types , day-to- day variabi lity in eac h STP 's perform ance and sa mpling errors associa t e d w i th sa mplin g from a hc tcrogeno us product. Variability w ithin plants is addressed in g u idelines recom mending "safe" levels of indicator o rganis m s, thro u g h speci fi cati o n of t h e am o un t of comp liance testing to be done in a given period . The m etho d use d in this study to co un t t h e rm otoleran t co li fo rm s e m p loyed a res usc itatin g in cubati o n at
35 C prior to selective incubation at 44.5 C: . Thu~, ~rre~~cd rherm orolcrant coliforms w hic h wou ld no t otherw ise be detected arc incl uded in the the rmorolerant coliform count. Th e resuscitat io n step also favours th e growth o f n o n-£. w li therm otole rant coliforms and this may ha ve contri buted to the relatively low ratio o f E. wli to no n-£. coli see n in this study. The variable contri butio n of Klcbsiclln to th e the rmoto lcrant coliform count red uces its value as an ind icator of e ntc ric pathogens , particularl y in the t ro pi cs w h ere Kl cbs ielln i s more abu ndant, and for the exami nation of samp les fro m ho ldin g ponds w h ere Klebsielln ca n multipl y to hig h co ncent ration s {McN c ill, 1985). H ad th e D e par tm e nt of Natura l R.cso urc es' g uidel ines specified an E. coli co unt rather tha n a thc rmo to lera nt coliform coun t of< 1,000/ I 00111 L, 29 of the e mu c nts in the single sample survey wou ld have passed for spray irrigation o f sugar can e, rather than 22. C. pe~(ri11,'.(c11s w as assayed sin ce it is resistant to che mi cal disinfection (T y rre ll ct al 1995), and m ay be a better ind icator of di sinfection of viruses tha n is the t h c rmot o l c r a n t co li fo r m co unt.
H owever, our results showed o nl y weak co rrelation with phage conce ntrations . Som e sam ples had very hig h counts of C. peifri11,(?e11s and th e po tential fo r build- up of spores in the soil as a resu lt of longterm re- use of wastewa te r may be a cause fo r conce rn. In the present study there was a significant co rrelation bet ween C. pe,fri11,'.(CIIS co unts a nd th ermoto lc rant colifo rm counts (r =0 .27, P<0.01). In a com parison be twee n cou nts o f C. pe,fri11ge11s and E. coli in marine water, Bonde ( 1975) found no distinct and constan t relationship between the 2 indi cators. Ferguso n ct al ( 1996) identi fi ed C. pe~/i·iJ1,',!e11s as the m ost useful indicato r of fae ca l pollutio n in an urban estuary in Sydney, and fou nd it was the o nly indicator significa ntl y corre lated to th e presence of pathogeni c Cinrdin (r =0.4 I, P<0.05). A similar correlat io n was o btaine d in th e presen t study (r = 0.45 , P<0.01 ). Cinrdin leve ls were mu ch highe r than those reported by Sykora ct al (l 99 1) for seco ndary- trea ted c ffiu c nt in th e USA . T hree of the ST Ps in o ur multiple anal ysis survey were always above 150 cysts/ L, and none was always below that level. W h ile coun ts of C. pe,f,·i11,'.(e11s were
ProMinent have the right metering pump and instrumentation tor any ot your chemical dosing appplications. Our Beta range features reproducible metering accuracy set by Quartz locked frequency in 10% increments with maximum capacity ranges from 0.6 to 32 1/h.
Vario & Sigma are motor driven pumps which offer a level of control previously only available with solenoid pumps, control from mA _ ___ or pulse signal.
Newly released hi-tech Gamma /L range. Accessories include flow failure monitors.
®
Ir
= = ·a)
1' The motor driven Meta , Makro and Orlita range are renowned for their reliability. Capacities up to 100,000 1/h and pressures to 380 bar.
Litres/hr displayed.
::i:
e A.
ProMinent S pectra progressive · either block
Instrumentation for feed forward PIO control. Measu pH, free & total chlorine, ORP, 0 3 , CI02 and conductivity. ProMinent Fluid Controls Pty Ltd Head Ofllce: Ph (02)94600995
Dis tributors
OLD Offtee: Ph: (07)32438330 VIC Offtee: Ph (03) 9562 9366
TAS:
NT:
WA: Ph: (08)94589555 SA: Ph: (03)6244 7575 NZ:
Ph (08) 89471008 Ph. (08) 82TT 7799 Ph· (64) 9579 2633
WATER MARCH/ APRI L 2000
43
WASTEWATER
sig nifi ca ntl y co rre lated w ith Giardia counts, this assay could no t be used to predict levels of Giardia in indi vidual sa mples. Similarly effiu encs with very high th erm otolerant coli fo rm coun ts gen erall y had high Giardia counts, but one plant with a low thermotol erant colifo rm count had the highest recorded Giardia co unt. He nce t he n eed fo r specific testing fo r Giardia, o r development o f a better indica tor assay. N o helminths (includ ing Tae11ia sp.) and r e la tiv e ly low le v el s o f Cryptosporidi11111 were detected in the survey. This is reassuring given their resistan ce to most di sin fec ti on processes, but probabl y refl ec ts th e low levels o f these parasites in the pop u lation serviced by these STPs. Coliphage counts we re correlated with thermotolerant colifo rm coun ts in the present study (r =0.59, P<0.01) . T he cnteri c viruses were no t assayed. Th e relevan ce o f bacteri ophages as surroga tes of other vi ruses has been reviewed (Havelaar '1991) an d they arc regarded as useful indi ca tors o f th e effi ciency o f th e d isin fect ion pro cess in e limin atin g viruses. Coliphage levels in this study were generally lo w in chlo rinated emu ents, so it is li kely th at enterovirus
numbers in these effiu e nts were also low. at least one sa mpling, for all o rgan isms M etho do logies fo r p h age assays are except Sa/111011el/a. Multiplication of E. beco min g m o re stan dardised than th ey coli can be expected in nutrient rich have been, and there may be a case fo r wa ter in trop ical and sub-tropica l spec ifyi ng acceptable phage levels in temperatures (H azen 1988, McNe ill water fo r agricultural re use, using a l 985, Rivera et al 1988) and excreta phage with similar disin fec tio n resistance from ducks and geese have increased E. to enteroviruses. M S2 phage appears a coli leve ls in wetland systems in reaso nabl e ca ndidate as it has rece ived Q ueensla nd (Gree n way and Simpson considerabl e study as a possible indi cator 1996) so the va lu e o f assays for thermoof enteroviruses, and its characteristics tolera nt coli forms and colip hages in these comparative to enteroviruses are well ponds is li mi ted. Also little was known of doc umented (H avelaa r 1991 ). th e conditions u nder which the holdin g This study was no t abl e to quantita- po nds were operati ng, an d it is possible tively compare trea tme nt types and STP that detention times were too short to sizes w ith respect to effiu ent quali ty, effect consiste nt reductions in the ind icabeyond the effect of chlorinati on on the tor bacteria and protozoans. therm o tolerant coli form co unt and on Conclusions Sa/111011ella isolatio n rate. The apparent T his work has provided data o n the link between non- use o f chlorin e disinleve ls of in d ica to r o rga ni sms and fection an d high thermotoleran t coli form pathogens in effiuents from Queensland counts and high Sa/111011ella isolation rates STPs, and th is data will be used in develis noteworthy considerin g th e pressure to o pi ng a quantitative model to assess the ph ase-o ut chlo rinatio n as the fi nal d isin- heal th risks associated with the re use of fectio n p rocess . emuent to irrigate sugar cane. Further Data in Table 4 co mparin g effiuent data are requ ired o n the numbers and q uality at the in let and o utlet o f hold ing , 'survival o f poten tially pa thogenic viruses ponds at 4 STPs indicate a general fall in in the effluents, and on the numbers, indi cator orga nism leve ls a(ter pond ing. survival and spread of viruses and o ther Ho wever higher levels w ere see n in pathogen s in aerosols generated by spraypo nd o utflows compared with inflows on irrigation . T he variability of the levels of organisms in effi uents, and the lack of reliable in dicators for the presence of many pathogens indicate a need fo r cau ti on in m anagin g effluen t-i rriga tio n, beyo n d tive health risk based on the degree o f sa tisfying gui de li nes listing acceptable e fflu e nt tre atm e nt , dis in fec ti o n levels of indicator organ isms. Regular m e thod , irri ga ti o n m e th o d an d monitoring o f treatment processes, and withholding period after irrigation. the appl ica tion of a well-constru cted In a companion study of disinfected and relevant risk assessment model are effiu ent in south east Quee nsland, also important facets in ensuring public Rynne and Dart identified enteric hea lth risks assoc iated wi th effi uent revirus levels in excess of 104 pfu / L, use for irriga tio n are acceptab ly low. w ith the effiu ent often being used for Acknowledgments golf course irrigation. The abse nce of W e are very grateful to the responsifrequent outbreaks of gastro intestinal ble agencies an d op erationa l staff o f STPs illnesses by golfers suggest to us either in these surveys. The work was assisted very high levels o f personal hygiene, by a grant from the Suga r Industry or the considerable difficulty in associR esou rce Panel. Literature review by atin g efflu e nt quality with health Na relle Fegan and the technical work of resp onse. Ralph Stutchbury, The rese Del Dot, R ynne, F. and D art, P. (1998) . Microbial Aaron Fielder, Paul Duffy an d Bo nni e health risks associated with effiu ent reuse. R esearch R eport N o. 144. Urban W ater Thomas is gratefull y acknowledged.
HEALTH RISKS This study co mpl em ents two o ther studies on hea lth risks fro m effiu ent reuse, which ha ve been published by Fegan et al. (l 998) and Rynne and D art (1998) . T he Fegan study review ed much of the histo ric data on effiu ent irrigation and co ncluded that "the publish ed literature provides ve1y littl e epidemiological evidence that indi cates a significant risk to health exists fro m the reuse of wastewa ter" . H oweve r finan c iall y r e ali s ti c epidemiological studies are a relatively blunt instrum ent to use for health studies o n reuse, es p ec ially wh en acceptable risk can be as low as on e extra infection per 10,000 persons per year (U SEPA 1989) . Rynne and Dart (1998) argued fo r the use of Quantitative Microbial Risk Assessment techniques, but acknowledged the difficulty of enumerating the concentration of the organisms (enteric virus or protozoan) causing infection , and the computational difficulty in doing the risk assessment on a routine basis. As an alternative , Rynne and D art presented Decision Trees of quantita-
44
WATER MARCH/APRIL 2000
R.esearch Association of Australia . U S-EPA (1989). N atio nal drinking water regulations: filtration, disinfection: turbidity Giardia lamblia viruses Lc,~io11e/le, h eterotro phic bacreria: Final rule 40 C FR, Parts 14 l and 1~2. Federal R e,~ister, 54: 27486. Fegan, N. , Gardner, T. and Blackall, P. (1998). Health Risks associated w ith the R e use o f Efflu e nt fo r Irriga tion. Queensla nd D ep artme nt o f Natural R.esources, D NRQ 980130 62pp.
References APHA (1992) Standard Methods fo r the examination of water and wastewater. 18th edition. Armon R. and Payment P (1988) A modifi ed m-CP medium for enumerating Clostl'idi11111 pe,fri11ge11s from water samples. C1111 . j. Microbial. 34: 78-79. Bisson JW and CabeIii VJ (1979) Membrane filtration enu mt:ratio n method for Clostridi11111 peifri11ge11s. Appl. E11viro11 . Micmbiol. 37 (1): 55-66.
WASTEWATER
Bitton G (1994) Waste Water Microbiology. Wiley-Liss, N ew York. Bonde GJ (1975) cited in Cabelli V J (1977) C lostridi111n peifringens as a water quality indicator. In H oadley AW and Dutka BJ (Eds .), Bacterial Indicators/ Health Haz ards A ssociated wit/, Water (pp. 65-79) Philadelphia: American Society for Testing and Materials. Bowden J (1995) pers.comm. Fattal B, Yekutiel P, Wax Y and Shuval HI (1986) Prospective epidemiological study of health risks associated w ith wastewater utilization in agricu lture. Wat er Science Technology 18:199-209. Ferguson C M , Coote BG, Ashbolt NJ and Stevenson IM (1996) R.elationsh ip between indicators, pathogens and water quality in an estuarine system . Water R esearch 30 (9) : 20452054. Gar dner EA , Ch inivasagam N , [tao A , Vieritz A , Blackall P, R.ynnc F, Thomas R., Klieve A, Blaney B , Green P and Barry G (1998) Q uantifying the health risk of spray-irrigating treated sewage e ffiu ent. Watertech '98, Austra lian Waste Water Assoc iat ion, Brisbane. Grabow W O K and Coubrough P (1986) Practical direct plaque assay for coliphages in 100 mL samples of drinking water. Appl. E1111 .Microbial. 52: 430-433. G reenway M and Simpson JS (1996) Artifi cial wetlands for wastewater treatment, water reuse and wildlife in Queensland, Australia. Wat. Sci. Tech. 33 (10-1 1): 221-229. Havelaar AH (1991) Bacteriophages as model viruses in water quality control. IA W P R.C
Study Group on Health Related Water Microbiology. Water Research 25 (5): 529545. H azen TC (1988) Faecal coliforms as indicators in tropical wate rs: a review. Toxicity Assess111e11t: An illtemational j o11mal 3: 461477. McNeill AR (1985) Mi crobiological water quality criteria: a review for Australia. A W R.C Tech. Paper No. 85. Mara D and Cairncross D (eds) (1989) Guidelines for the safe use of wastewater and excreta in agriculture and aquaculwre. World H ealth Organisation , London. Q ld . Department of Natural R.esources (l 996) Interim guidelines for t he reuse of reclaimed wastewater in Queensland. Queensland Department of Natural R.esources, Brisbane. R.ivera SC, Hazen TC and T o ranzos GA (1988) Isolation of faecal coliforms from pristine sites in a t ropical rain forest. Appl. Env. Microbial. 54 (2): 513-517. Shuval HI , W ax Y, Yekutiel P and Fattal B (1989) Transmission of enteric disease associated with wastewater irrigation : a prospective epidemiological study. Am. j. ,!f Public 1-1/tl, 79: 850-852. Sykora JL, Sorber CA, Jakubowski W, Casson LW, Gavaghan P D, Shaprio MA and Scho tt MJ (1991) Distribution of Ciardia cysts in wastewater. Water Science & Tec/1110/ogy 24: 187-192. T yrrell SA, R.ippey SR. and Watki ns W D (1995) Inactivation of bacterial and viral indicators in secondary sewage diluents, using chlorine and ozone. War. Res. 29 (1 1): 2483-2490.
Acknowledgments
We are very grateful to the responsible agencies and operational staff of STPs in these surveys. T he work was assisted by a grant from the Sugar Industry R eso urce Panel. Literatu re review by Narelle Fegan and the technical work of Ralph Stu tchbu ry, T herese Del Dot, Aaron Fielder, Paul Duffy and Bonnie T ho mas is gratefully acknowledged. Authors
Rod Thomas, Nalini Chinivasagam, Athol Klieve and Pat Blackall (microbi ologists), Peter Green (parasitologist), Gary Blight (biometricia n) and Barry Blaney (chemist and admini strati on co ntact) ar e w ith the Queens la n d D epartment of Primary Industries, based at the An imal Resea rch Institu te at Yeerongpilly (Locked M ail Bag No 4, M oorooka 4105) . E ma il: tho ma sr@ dpi .qld.gov.au Ted Gardner (Principal Scientist and project leader) and Glen Barry (chem ist) are w ith the Q uee nsland Department o f N atura l Resources , Catch m ent Processes, Natural Sciences Precinct, 80 Me iers R oad, lndooroop illy 4068. They are parties in th e CRC fo r Sustai nable Suga r Produ ction. Email: T ed.Gardner@dnr.gld. gov.a u
I L I T I E
s
Vinidex continues to
develop new &
ativc water mains e in the world is the ult of endless research, evelopment and testing.
inn9vative
so1ut1ons for utility providers & installers.
A continuous high-dens,~ polyethylene conduit ma1<1· factured by Vinidex using an innovative and advan5qd extrusion process. /
"always specify Vinidex products."
Ill
.•
Ultra rib These new generation material efficient prdduf tS offer the most ad/anc.'ed pipe systems availabl1· providing innovative/ systems & solutions. 1
II
• •
.
SUPERMAIN VINYL IRON CABLECON ULTRA RIB
• • • •
CORFLO CONDUITS DUCTING DRAINCOIL
POLYETHYLENE
III I Sales H otline
13 11 69
w w w .vinid ex. co m . a u e-mai l: info@vinidex.co m.au II
,I I
I
a
BUSINESS
Abstract In Australia, there is a growing awa reness a mo ngst water authorities for the need to identify the ir overall business risks, demonstrate due diligence and set priorities for capita l in vestm ent. This can best be ac hieved through a complete tec hni cal and lega l risk manage m ent progra m . For regio nal and m etropolitan water authorities a key co mpo nent of such a program wi ll be risk managem ent for drinkin g water quality. In th is paper w e d e ta il ou r approac h for suc h a progra m , w hich has been applied by m aj or water authorities in Australia . It invol ves a screening level risk assessm ent cove ring all components of a wate r supply system: catch men t, reservoir, bulk water transfer, treatm en t, distribution and delivery. T hrough this first step we can identify priority areas that can the n be subj ected to a more de tailed risk assessment, m aki ng the overall process cost-effective. In practice, m uch of the risk assessm en t can ca rried out by water authority staff using me thods developed by Parame trix Australia and AW T , and drawing o n the staffs intimate u nderstanding of the systems that they manage. For eac h ide ntified ri sk we develop a nu mber of risk managem e nt options. These options ca n th e n be ranked 46
WATER MARCH/APRIL 2000
according to th eir risk redu ction per unit cost to obtain an optimal risk managemen t program . Furth er details of this process, w ith exa mples drawn fr o m recent consu ltancies conducted by AWT and P arametrix Australia, are give n in the paper.
Introduction Th e c urre n t Au strali a n D rinki n g W ater G u idelines (NH &MR C, 1996) pro vide many elem ents of a preventati ve man agement system , th ey inco rporate recomm endations to manage catchm ents to avo id co ntam ination fro m human and an imal wastes, and to maintain multiple barri e rs from catc hm ent to tap. H owever the re is little guidance o n how to implement these elem.e nts and the emphasis remains on traditional compli an ce w ith num e ri cal guidelin e values and reac tive in ves tigation a ft e r m on it o rin g h as revealed excee dances . T he on- going review of these Guidelin es is to provide th e elem ents of a preventive, quality ass ura n ce approach a nd will bring compliance with num e rica l limits i n to th e context of water qu ality manageme nt. T he re is a need to shift from quality co ntrol to quality assurance. Th e water industry has unde rgon e ama lgamatio ns and corporatisation unde r
the COAG reform process crea ting ne w c hallenges for water managers, particularly in Vi cto ria. Non - m et ropolitan water authorities now have respo nsibilities for many more town supplies than in the past, in so me cases more than fifty to wn suppli es . T he Bo ards overseei ng th ese authoriti es are selected for the ir commercial skills and th e me mbers are insisting that due diligence requ ire men ts are m et. Vi ctorian water auth orities have sig ned m e moranda of unde rstandin g (MOU ) wi t h th e M i n iste r for Agric ulture and R esources to upgrade drinkin g water quality to meet WHO 1984 Guideli nes by the year 2000. As a result of the M OU, water authorities have made substantial in vestme nts in treatm ent facilities w hich com e on- line in the n ext 12 months. Th ese new and improved fa cilities need to be managed in conc ert with the distri bution system ope ra tio n to achi eve t he mandated improvem ents in water quality. There is a also n eed to manage priorities for water quality imp rovem ent, bo th in equipme nt and ope rations, wh ich takes account of the cost effectiveness of indivi dual action s.
Risk Management R isk manage me nt (AS/ NZS 4360 :
BUSINESS Table 1. Typical Hazards CATCHMENT
STORAGE
Facility Failure
General Pollution
Facility Failure
Faclllty Failure
Wastewater TP failure
Algal blooms Stratification
Overall power failure Dosing equipment
Management
Chemical Contaminants
Pipe break Treatment dosing failure Mains condition
Communication Public access Experience/ expertise Reduced monitoring Reduced maintenance
Variation of chemical strength Backwash supernatant return Disinfection by-product formation
General Pollution
Land use - mining - farming and grazing - horticulture - forestry - stormwater runoff Industrial Sewage/ TP effluent
TREATMENT
Sabotage/ Accidents
Accidental spi lls - transport Management
Communication Experience/ expertise Reduced maintenance
1995) is de fin ed as th e syste111atic application of //lanage111e11I policies, proced11res a11d practices lo the tasks ef idrntifyi11g, analysi11g, assessi11,i;, treati11,~ a11d 111011ito ri11g risk. A risk-based approach in c lu des th e ide ntifica tion of haza rds and th e assessn1e nt and m a nagem e nt of risk; wh ere hazard is d e fin ed as th e po te ntial fo r an u nd esirab le conseque nce or impac t and risk as th e overal l m easure o f th e co nseq uen ce and freque n cy o f a hazard . T he m e th odology described belo w is an o pe ra tional , se mi - quantitati ve app roac h based o n w o rkshops to capture th e expe ri ence and judge m ent of water autho rity o pe rating pe rsonnel and managem ent. A consultant tea m w ith expe rtise in w ater supply and wa te r qu ality provides worksho p fac ili tatio n and prese ntati o n o f the results. Th e p rocess o utlin ed be lo w usually invo lves at least two w orkshops, on e to agree o n the suppli es to be exam ined , ide ntify th e hazards and associated ri sks and th e second to develo p and eva luate approp riate respo nses.
Design - Appropriate treatment processes - Operational flexibility - Process control - Inadequate of mixing (e.g. chlorine) Operation - Sudden or large raw water variation - Over/ under dosage of chemicals - High tu rbidity during ripening periods Equipment failure
Establishing objectives Th e first stage is to establish and agree upo n g round ru les and o bj ecti ves fo r the process. As the m etho do logy requ i res substantial comrnitme nt fro m the w ate r auth o rity, nam ely the cl ose involvem ent of key o perational staff, it is impo rtant that expectati o ns o f t he process are accu rate. It is also importan t that a!J sys te m co mp o n ents (i.e . ca tc h m e n t, storages, o ffta kes, treatm ent, and distrib utio n) are includ ed in the process as th e re is a hi g h degree of interdepe nde nce betwee n these compo nen ts fro m a risk managem ent pe rspec tive.
Hazard Events
Duration
Frequency
+
X
Magnitude
Consequence Factor
Diagram 1. Ri sk Assessment Process
General Pollution
System Operation
The de tail of th e m e th odo logy fo r eac h proj ect, specifi call y th e scales used for scoring and th e weig hting fact o rs, is develop ed in conj unc ti o n w ith t he cl ient. Th e sco ring sca les are c hose n to re present the range of haza rd eve nts re levan t to t he proj ect, and th e w eightings are develo ped to reflect th e c li e nt's judgem ent o f w hat is important.
1/ !
DELIVERY/ DISTRIBUTION
+
Customers Affected
Risk Score
Uncovered storages Corrosion Human/ animal/ vermin access Biofilms Disinfection by product formation Cross connections Bacteria growth Commissioning new mains System Operation/ Design Inadequate operating procedures Stagnation in dead ends Decay in chlorine due to detention
Screening Process Since the o perational risk managem e nt m etho dology ca n o nly be ca rri ed o u t co ncurre ntly o n a sm all n umbe r of suppl ies, an ini tial screenin g process is employed to identify chose w ith hig h risk pro fi les. As part o f th is screening pro cess a questio nnaire has bee n deve lo ped to score risks in th e ca tchm ent, sto rages, treatm ent and distribu tio n co mpo ne nts of each individu al syste m . T he scores are w e ig hted acco rdin g to th e re lati ve imp o rta n ce of c ontributi o n s . Fo r exa mpl e, for mi c robio logical contam inatio n in a ca tch m ent the co ntribu ti o ns are we ig hted in the orde r of human acti vity, inte nsive anim al indu stry and native animals - based upon t he human infectivity of mi cro bial pathogens associate d w ith each ac tivity and the intensity o f the acti vity. Th e scores fo r th e catchme nt are th en m odified by valu es attributed to aspects o f t reatm ent and distribu tio n su ch fac tors as th e effecti ve ness o f disinfec tio n and the p o tenti al fo r co ntaminati o n due to un covered sto rages in the distributio n system . Th e po pulatio n of eac h suppl y is also conside red, w ith a score based o n th e logl O o f the to wn popu lation to e nsure th at th is fac tor is no t o ve r- re presented. Th e screenin g process w ill sep arate o ut the hig h risk profil e supplies to be subjecte d to th e m o re detail ed risk assessm e nt/ m anagem ent process.
First Workshop Ha z ard Identification: At a first workshop, a list of generic hazards (Table 1) is pro vided to "seed" a brainsto rmj ng process: ch is is aimed at revealing sig nifica nt hazards associated w ith each supply system . Partic ipants are asked to rev iew the list, de te rmin e w hich o nes appl y and describe them in te rms spec ifi c to the
WATER MARCH/ APR IL 2000
47
BUSINESS Table 2. Summary list of ranked hazard events and their contributions to total risk score. Duration
Magnitude
Customers
Frequency events pa.
Percentage Risk
1-7 d.
High
35,000
1 20
46
< 12 h.
Medium
35,000
50
10
Dairy sheds , piggeries,
1-7 d.
Medium
35,000
30
6.5
poultry farms Run-off from road s into
12-24 h.
Low
35,000
30
5.9
1-7 d.
High
35,000
10
3 .8
12-24 h.
High
35,000
10
3.8
< 12 h.
High
12,000
10
2.8
< 12 h.
Medium
35,000
12
2 .5
Hazard Event Name
Description
Agricultural drainage water Poor particle removal during
Entry of drainage water into the River As shown by particle counts
fi lter ripening periods
and turbidity
Agricu ltural/ farming/ grazi ng activities in catchment Transport corridors Sudden or large water quality changes Town stormwater run-off
channels and then to River Floods , increase in rainfall in catchment Stormwater e nters approx 20m upstream of water supply offtake
Design of chlorination system suspect
Control process relies on inaccurate monitoring
Insufficient t raining of treatment plant operators
Incorrect sample collection and chlorine residual measure ment
equipment
system being considered . T hey are also asked to brainsto rm other events that ma y represe nt hazards to water qu ality in their syste m. For each hazard participants are asked to estimate th e frequency, duratio n, m agnitude and numb er o f custom ers affected, rem embering that th e frequency is not that of the hazard bu t of the number o f times it impacts o n water qual ity . For example, filter ripening may occur at daily frequ ency bu t th e number o f times that the system is vulnerable to pathoge ns breakthrough may only be twenty or thirty times over a year. T he hazards are then assessed as o utlined in th e flo w chart dia gram below. Th e m eans fo r determining scores for the various factors are included in the m ethodology, w hich is specific to each project, as are the weighting factors used
to co mbine the scores into a fina l R isk Score fo r each hazard. Risk Ranking: The hazards are then ranked in descending risk sco re and their individual percentage contribution to the total risk score for the suppl y system can then be calculated. T his process qu ickly identifies the maj or risk con tributio ns, typically the first ten to fifteen hazards constitute 80% to 95% of the total risk, and it is this group that sho uld be selected for further attentio n. At this stage the total risk profile is reviewed , ideally at the second workshop so that participants have a chance to reflect o n their w ork so fa r. The output of this sessio n is in the fo rm described in T able 2.
Second Workshop Risk Responses: For each hazard risk m anage ment responses are develop ed.
T hese are aimed at frequency reducti on reducing the probabili ty o f the hazard; or consequ ence reduction - reducing th e duratio n , m agnitude , or numbe r o f customers affected; o r ways of avoiding the hazard completely. In some cases furt her info rmatio n may be required before id entify in g management options. These m ay in clud e wate r qu ality mo nitoring programs to determine actual sources of contamin ation to be the targets of future actio ns. Other respo nses will include m odificati ons to existing facilities, or intro duction of new equipment, operating procedures, Q A or Q C measures or other actions. Operational staff will be a rich source of ideas; their collec tive experience will throw up the results of relevant past investigations and discussions on manageme nt options. D ecision Scores: Using the process
Identify Risk Management Responses
t
t
t
j
Recalculate Risk Scores
t Effectiveness Score
+
Diagram 2 . Risk Management Process 48
WATER MARCH/ APR IL 2000
Cost Score
Estimate Feasibility of Response
Estimate Cost of Response
Estimate Reduction in Consequence produced by Response
Estimate Reduction in Frequency produced by Response
+
Feasibility Score
=
Decision Score
BUSINESS
outlined in Diagram 2 the probable reductions in fr equ ency and consequence are determined fo r each response (to recalculate the reduced risk scores) , together with an es timate of the feasibility and cost of implementation. This provides a decision score ranking the respo nses in order of best outcome for a particu lar hazard. These scores cannot be used to compare the valu e of responses for o ne hazard to those in another hazard, wh ich is essential to developi ng an overall risk management strategy. For comparison between hazards, the cost effectiveness of eac h response is calculated in terms of annual cost for one percent reduction in risk . An exampl e of thi s o utcome is shown in Table 3. Th ese calcu lated cost effectiveness values can be used, with care, for compariso ns be tween water supply system s. Factors, su ch as the popu lation affected, need to b e cons id ere d w h en integ ratin g responses into a managem ent strategy.
Action Plans Coord ination of the acti on phase is esse ntial as there may co nflicting or mu tua l b e n e fits in impl e m ent ing
indi vidua l res p onses . A listing of optimum responses, sorted in order of cost, wi!J reveal a series of actions with minimal implemen tation cost. These most often relate to improvements in communi catio ns an d, cl early, these shoul d go ahea d regardless of further cost-benefit analysis. Th ere w ill be other responses that can be logically grouped together in action plans eg. training actions - workforce skills, operating procedures, in cident management. Th e fi nal actions w i!J be assigned to individuals to manage co mpl etion in a reasonabl e, agreed timeframe.
Monitoring and Review Ongoing commitment is criti cal to effective risk managem ent, this includes fo llow up on actions - monitorin g of implementati o n and determini ng that risk reduction has been achieved . Th e whole process must be repeated periodica!Jy to reprioritise significa nt hazards taking accou nt of actions achi eved, and to in clude other supply systems not su bjected to detailed analysis in the first round. Th is methodology, o nce adopted by an orga nisa tion , can be efficiently
carried ou t using internal resources.
Conclusion T he methodology outlined above has provid ed a means for maj or wa te r authorities in Australia to manage water quality in a proactive man ner foc ussing on predictio n and prevention rather than reactio n to non-compliant monitoring results. It m eets du e diligence req uirements and can be readily integrated w ith existing asset and quality management programs. A logical extension is to introduce H azard Analysis C riti ca l Control Point (HACCP) as a quality assurance m easure to maintain th e effectiveness of ac ti ons id ent ifi e d by the pro cess (H aavelaar, 1994), (D eere and Davison , 1998). The methodology can be readily extend ed to the operation of other water-related systems eg. sewerage. W hi lst the description above fo cuses o n issues relating to microbial pathogens, the process methodology captu res a wide ran ge of haza rds, which may represen t chemical, biologica l or other risks. T he pro cess, th ro u g h conce ntratin g o n events rather than spec ific water quali ty indicators, avoids preocc upation with
WATER MARCH /APRI L 2000
49
BUSINESS
Table 3. Hazard responses/risk reductions for use in risk management planning Rank No. 2
T16 a. b. C.
e. f. 3
S5 a.
b. C.
d. e. 4
S10 a.
b. c. d. 5
Hazard Event/Responses High turbidity during ripening period of filter Redirect first 5-10 min of filtered water run to waste Slow closure of backwash valve {water hammer dislodges particle from bed) Variable speed drive pumps to avoid on-off operation of filters Monitor particle counts and turbidity, analyse results and optimise performance Improved f ilter maintenance {reduce fissuring of media) Agrlculturaljfarmlng/ grazlng activities Mon itor quality {locate , assess and verify problem) Target farming practices - reci rculation systems, education, incentives, fines etc Construction of wetlands to improve drai nage waters Move offtake upstream of area Tertiary treatment of drainage waters Transport corridors Red irection of table drains (Water Authority and Councils) Identify specific hazards and location of high risk entry points Staff plant more frequently during high rainfall (out of hou rs) Ripari an zones
T13A Sudden large water quality changes Transfer offtake upstream of river j unction a. b. Improved record keeping/planning and interpretation of records SOP' s to cover poor water quality events Staff plant more frequently during high C. rainfal l events, continuously monitor raw water and j ar test frequently d. Global catchment management improvements, such as stream bank stabilisation Construction of large sto rage upstream of offtake e.
cu rrent, copical iss ues and has the potential co mitigate othe r c he mi ca l and mi crobiologica l hazards of wh ic h we are not yet aware. T h e scori ng and we ighting factors used in the m ethodo logy are tailored to suit each proj ect, based o n the exte nsive experie nce o f Parametrix and AWT in risk assessm ent and manageme nt. Wate r authori t ies can u se t hi s approach to manage risks from ca tc hm e nt to tap , or for indi vidual com ponents of a system. For in stance, where new trea tm e nt pla n ts have b een co nstru cted to meet water qua lity targets this risk management process can pro vide needed improvem ents co management of the distribution system to deli ver high quality water to c ustomers. The process ca n target sou rce water p rote c tion 50
WATER MARC H/ APRI L 2000
Decision Score
Risk Reduction
Response Cost Per Year
Cost/Year/% Risk Reduction
0.50
10.6
$125,000
$11 ,745
0.65
4.3
$5,000
$1,163
0.78
8.6
$12 ,500
$1,453
0 .71
5.4
$10,000
$1,860
0 .67
3 .8
$10,000
$2,584
STUDY
0
$50,000
no risk reduction
0.77
4.9
$100,000
$20,450
0.64
4.9
$500,000
$102,249
0.55 0.46
3.4 6.2
$500,000 $1,500,000
$146,071 $242,170
0.42
0
STUDY
0
$50,000
no risk reduction
0.54
4.9
$6 ,000
$1,205
0.78
4.5
0.58 0.68
2.8 1.4
$500,000 $31,000
$174,064 $22,483
0.91
3.2
$6,000
$1 ,865
0.48
2 .3
$500,000
$217 ,581
0.48
3.4
$800 ,000
$232,086
ac tions tha t can be coordin ated with other catchme nt management agen cies, o r passive actions suc h as water bird mitigation programs to prevent contamin at ion of water in sto rages. Such programs have proved very successful in achieving water quality improvements through low cost measures.
References N H MRC ( 1996) A11mali1111 Dri11ki11g Wam C11iddi11es Ag ri c ultt1re and R.esou rce M anageme nt C ouncil of Australia and New Zealand , N ati ona l H ea lt h and M edica l R esearch C ouncil N H MR.C (l 999) Re11ised A11srraliau Dri11ki11g Wafer C 11ideli11es (Cryplosporidi11111 and Ciardia) Drq{r Agric ulture and R esource Management Council of Australia and New Zea land, N ational H ealth and Medica l Research Council.
no risk reduction
$0
D eere D and D avison A (I 998) Safe Drinking Water , J,1/arer 25 (6) p21 H aavelaar A H (l 994) Applicatio n Of H ACC P T o Drinking Water Supply r:ood Co11rrol 5 (3) p l 45
Authors Ross Bannister a nd Dr Nick O'Connor are both Prin c ipal Co nsu lta n ts w it h Australia n Water T ec hn o logies (A WT), 68 Ri cke t ts Road , Mt Wave rley, Victoria 3 149 . Carl Stivers is a co n su ltant with Parametrix Australia. T ogether they have worked on toxicological and operational risk assessmen t proj ects for a nu mber of Victorian water authori ties. Carl has worked throughout Australia on risk m an age m ent proje cts, a nu mb e r of which have been on water-related issues. R oss ' e-ma il is rbanniste r@wes.com .au
ENVIRONMENT
IIJ
ENERGY EFFICIENCY - A MANAGEMENT APPROACH A McCleery, G Weiss Summary Organi sa tions that address e ne rgy m a nage m en t in a syste m atic way as they w o uld an y other m ajor busin ess issu e tend to be the leaders in energy performa nce. Th ey e nsure there is commitm ent to en e rg y m a nag e m e nt , ad e qu a te reso urces to bac k th e ir commitmen t, that key e nergy performance indi ca tors arc m easured and repo rte d, and that staff are fu ll y aw are o f the ir ro le in e n e rgy m a nage m ent. These organisation s also sec th e chall enges to th e wo rld e nviro nm ent as a n oppo rtunity fo r introdu c ing ene rgy effi cie ncy rather th an a risk to ope rations.
Introduction Th e w ate r and wastewater industry is a large consum er of e ne rgy, for example Syd ney Water alone co nsum es around 400 m iLl ion kW hours of electri city per year. En e rgy e ffi ciency has mu ch to offe r the w ater and waste wa te r industry in botto m line performance. Already th e industry has em b raced a number of e nergy saving techn ologies. These in clude: • V ariab le speed d ri ves • Co-gene ratio n and hea t recovery • L oad shiftin g • Timed and loa d co ntro ls • Auto mated Dissolved O xygen controls • Co111pressor opti misatio n • F ine bubble diffu sers So m e of these technol ogies have lead to signifi ca nt e ne rgy sav ings. H oweve r, th e y arc o nl y o ne part of the ene rgy mana ge m e nt e q u a tion. A r o bu s t m a n agem e nt syste m is neede d to lock in savi ngs and to dri ve a process of co n tinuo u s impro ve 111e11t. Furthe r, tec hnical en e rgy e ffi c ie ncy proj ects do no th ing to a d dress sho r tco min gs in op e ra tin g, ma intenance and pu rchasing prac tices. T h is m ay be robbing th e fa c ility o f m ajor e ffi ciency gains and cost savings. As an exa111 plc, savings in excess of $500,000 were identifi ed in a recent study co mplete d at a m ajor wastewate r treat111 ent fa c ility. Th e site has an annual en e rgy exp enditure of $ 1 .2 million. M a n y of these savin gs cam e fro m ope rational o r syste m changes rather th an n ew e nergy effi c iency tech nology. T his is t y p i ca l o f th e majority o f studi es
*
Costs +5%
Management says costs are too high again: Where's that last audit?
~
savings herij)
0
.. *
.....
This feels familiar
-5% -10% Oh well, that's enough. Manager says get on with other programs
-15% -20% -25% 0
5 Years
10
Figure 1
Ill!
SPECIALISING IN ENVIRONMENTAL SERVICES
iiiiiiiiiiiiiii
ENVIRONMENTAL GROUP NATIONALLY Neville Ward, Ron Edwards & Dr Peter Nadebaum (03) 9272 6666 Peter Zemek (07) 3233 161 1
Water Supply & Treatment Water Quality Management Mana~ement Systems & Comp iance Auditing Water Resources Development, Hydrology, Irrigation & Drainage Wastewater Collection, reatment & Reuse Environmental AudiV Site Investigations
NEWCASTLE Ian Gregson (02) 4929 3255 NSW David Russell, Dr Ian Garrard (02) 9412 9999
Contamination Assessment & Remediation
QLD Stephen Trainor (07) 3233 1611
Hazardous/Industrial Waste Management Solid Waste Management Environmental Impact Statements Environmental Management Planning & System Development Emissions Testing & Air Quality Monitoring
t£eg i5 consulting Australia Offices throughout Australia and South East Asia
VIC Don Vincent, Dr Jeff Bazelmans, Mike Juleff, Dr Wayne Drew (03) 9272 6666 WA Scott Bird (08) 9220 9300 SA Paul Lindon (08) 8271 2322
WATER MARCH /APRIL 2000
51
ENVIRONMENT
Senior management want to help us reduce costs Costs
*~-
+5% 0
The operations and maintenance guys are finding more savings. They have put in systems to lock them in
-5% That new process modeling system is saving energy and boosting production!
-10%
That new plant is
-15%
really energy efficient
*
-20% -25% 5 Years
0
10
Figure 2 completed by o ur company across a wide range o f industries. CO 2 equivalent emiss ions attributable to the sam e facility were estimated to be aro und 20,000 tonnes per year. T his is a significant greenh ouse contribution fo r a single site and th e site risks significant
financia l impost if the green ho use debate lead s to carbon taxes o r emissions permits.
Why energy audits alone do not work The old approach to the man agement
FILTRATION MEDIA SAND, GRAVEL, COAL, GARNET MANGANESEGREENSAND Invest in accurately graded, durable media from your complete filter media professionals.
U'eaufduuedU,~~: • Decades of mineral processing experience. • • • • • •
Media produced to the Al/'-/'NA B100-89 Standard. An extensive product range at competitive prices. Manufacturing In accordance with Quality System AS 3902/ISO 9002 Packaging alternatives to suit any requirement. Proficient technical assistance and support. Prompt delivery Australia-wide and Overseas.
So, tap into our extensive experience and helpful service when next you require filter media.
---~-RIVER SANDSrt
683 Beenlelgh-Redland Bay Road, Carbrook Qld 4130 INTERNATIONAL: +6173287 6444
~ E E CALL: 008 077 744
52
WATER MARCH /APRIL 2000
FAX: (07) 3287 6445
of energy efficiency was the Energy Audit. Unfortu nately the audit is at the centre of an ad hoc approach to energy cost control, one that results in a rolJ er coaster ride for energy performance . The orga ni sation gets co ncerned about energy costs and undertakes an energy audit. This identifies 5% to 10% easy savings. As there is no systematic process to implem ent these findings the easy (low capital cost) work gets done and the rest is shelved. After some time the 'Champion' pushing these actions moves o nt o anot h er task , energy management is seen as complete, and is quickly forgo tten. Some energy savings from capital works w ill be sustained but m ost of the initial savings wi ll be lost. Then in another 3 to 5 years the whole process is repeated . Fig ure 1 tells the story . The problems with audits include: • Th e audit j ust provides a snapshot of opportu nities at one time • There is limited ti me to complete & audi tors are not always experts in your operation. • Audits do not focus on 'management ' aspects w hi ch are usually critical in cost control. Nor are they necessary aligned w ith business objectives. • T hey are in evitably highly technical, with a strong engineering focus and are not directed towards the real decision makers • Actions stemming from audits are usually dealt with on a project by project basis, which guarantees that th ey disappeared from agendas. An audit is not an energy m anagem ent program . Th e outcomes of the audit (usually a report) exist in isolation wh ich in a typical busin ess environment, m eans that they soon are disca rded.
Energy Management Systems Achi eving true energy e ffi c ien cy requires much more than occasional audit ing and tec hnical ap plications. Before any capital program for energy efficiency is considered , man y other critical managem ent elements must be in place. This enables costs to be driven down as demonstrated in the Figure 2. This is nothing more than the trend set with quality managem ent systems and picked up by safety and environment management systems. Energy management has historically been neglected because energy costs are often not seen as controllable costs, energy performa nce was not seen as being critical to business performance and there was neve r an overriding complian ce issue. Now, however the situation is chan ging as businesses cannot ignore any cost input and the greenhouse
ENVIRONMENT
d e bate is leading to greate r external interest in th e use of ene rgy . • S u ccess ful e n e rg y m a n age m e nt re q uires organ isa tions to address at least t h e fo ll owi ng ele ments: Leadersh ip • Understanding & Plannin g Finan c ia l & S u ppl y • P e opl e Managem e nt • Operating, M ainte nance & Plan t R e p o rting a nd • M e as u r e m e nt , Feed bac k So m e of the issu es around th e above elem ents of the energy manageme nt syste m are discussed below. Leadership
• Does the orga11isalio11 recog11ise rne~fZ)' as a co11trollable cost? • Is there a specific policy lo address e11e1;gy eff,cie11ry a11d gree11ho11se? Most leadi ng water and wastewate r utiliti es ha ve mo ved on fr o m thi s positio n . T here is usuall y a se nior executive sponsoring ene rgy managem en t and oversee in g a cost reco very program. M.any o rgan isations have publish ed an En e rgy Management Poli cy and suppo rt th is w ith reg ul ar publi c reporti ng of e n e rgy pe rformance. It is criti cal that seni o r managem ent has contro l o ve r e n e rgy m anagem e nt so that in te rnal ba rri ers can be e liminated . Understanding & Planning
• Do yo11 k1101.1• when a11d where energy is 11sed, how 11111ch and what type? • I-low does your peifomia11re co111pare to i11 r/11stry a,1m1cQe and bes/ practice? • H a,1e base KPls bee11 established and targets set.for.facilities a,1d regions? • Is there 1111 ene,;Ry 11iariaJ/Cl'lle11t pla11 with defi11ite lh11elines and respo11sibilitics? Often e nergy and g reen ho use redu ctio n targets arc set without a co m p lete u n d e rstandin g of e ne rgy co nsumption an d po te ntial for redu ction. Some o f o u r rec e nt experie nces w ithi n and outside t h e water indu stry show po te ntial en e rgy sav ings of aro u nd 10 per cent (by operating at no w o rse than average) and u p to 20% (by operating near a level of pe rforman ce th at has already been achieved). These savings are t yp icall y availa ble w ithout signifi ca nt capital expe nditu re . Se ttin g K e y Pe rfo rma nc e Ind ic ato rs (K P I) is th e fi rst step in identifying suitab le targets, priority sites and savi ngs potential. F o llowing a diagnost ic o f ene rgy manageme nt pe rforman ce, an en e rgy managem ent plan should be prepare d to imp le m ent desired im provem ents. It is o ur experie nce that most plans need not in clu de te c h ni cal a pplic ation s until signi fica nt system and people d evelopm e n t has occurre d . We have also found
that plans should be implemented in disc re te phases cove ring fo r example a thr ee- m o nth p e riod. Th is a llo ws m anageable prog ress and clearly defi n es responsibilities, necessary resources and acti viti es.
yet th e e n ergy bi ll , som etim es in the $100,000s, is not o verseen by a senior manager. Finance & Supply
• f,Vho is res ponsible .fo r E11 e1ly Maiiage111e111' • ls t/1is recog11ised in positio11 descriptions' • Has mi1are11ess an d trai11i11g been provided?
• Is e11e1;gy ~ff,rie11cy a11d l[fe cycle costing i11co1porated into capital allocatio11? • H ow are e11er;_ey operatillJ! budgets set a11d are they regularl y re11icwed? • Is e11er;_~y bcillif p11rrhased at the best rate' • How do quality <if supply issues a_lfect your b11si11ess?
An enginee r is usually g iven respo nsibility for e ne rg y im pro vem en t in iti at ives thoug h w itho u t authori ty over energy consumption. Look at the energy expend iture and the n the resources dedi cate d to that ex penditu re. Afte r com pleting an ev aluatio n of e n e rgy manage ment perform ance one facility m anager stated " we have 20 peopl e managing a mai ntenance ex penditure o f SS mi llion and on ly ha lf a pe rso n m anaging an ene rgy expe nditu re of S I O m illion " . W hile energy ca n be inco rporate d into o ther du t ies, it is important tha t it is rec ognised as a co n tro llab le cost that needs to be effective ly ma naged . Ofte n w e find o rgan isat ion wit h rigorou s con trols o ver operational ex pe nd iture
Often ca pital prog ram s ignore life cycl e co sts assoc iated with e n e rg y e ffi cie ncy . Co ntrac tua l arrange m e n ts n eed to be structured such that su pplie rs o r co ntrac tors are not d riven to using c heaper yet less e ffi c ie nt i te ms t hat in the lon ger run cost sign ifica ntly more, on ce o pe rational and maintenance exp ense is included. M aintenan ce co ntracts that pay a m anagement fee for new capital w o rks ca n be cou nter product ive to e ne rgy e ffi cien cy as there is a pre fere nce to replace rather than decom m issio n o r d o wn size . If th ose w ho are respon sibl e fo r ene rgy use are not aware of co ntract provisio ns th e re is th e po te ntial for major financial expos u re. For exa mple if
People
C&SBRAND AUSTRALIAN FILTER COAL & C&SBRAND GRANULAR& POWDERED ACTIVATED CARBONS INCLUDING OUR INNOVATIVE COAL/WOOD BLENDED P.A.C. Phone or fax for sound technical advice from experienced personnel.
JAMES CUMMING & SONS PTY LTD 319 Parramatta Road AUBURN NSW 2144 Phone: (02) 9748 2309 Fax: (02) 9648 4887
QUALITY ENDORSED COMPANY AS / ISO 9002 STANDARDS AUSTRALIA
WATER MARCH/APRIL 2000
53
ENVIRONMENT
contractors are not aware of ratchet clauses in demand contracts, they could inadv erten tly cause an in cr ease in demand charges for the remainder of the contract. Operations, M aintenance & Plant
• Has energy been incorporated into operations and maintenance procedures? • Have economic projects been implemented? • Are capital programs supported with training and awareness? It is often possible to identify 5% to 10% or more of energy savings that ca n be achieved in buildings and industry simply by takin g basic maintenance steps and having properly co mmissioned control system s. Successful organ isatio ns develop and u tilise formal operating procedures and work instructio ns for energy efficiency. These are integrated with overalJ operating procedures. T hey also incorpo rate energy efficiency policies for purchasing of plant and equipment . T o ac hieve best practice e nergy management, it is not sufficient to have good systems; the technology employed m ust also be energy efficient. In order to achieve this, existin g plant may need to be supported by additio nal equ ipment.
R etrofits can be undertaken to upgrade energy efficiency. It is essential that guidelines are ava ilable for design, new instalJations, retrofits and replacements The business sho uld also actively pursue technology inno vation and R &D with energy conservation and efficiency in mind. Measurement, Reporting & Feedback
• Is critical data collected and reported? • Are variables reported and acted 011? • Are reg ular performance app raisals completed? (eg Post Implementation R eviews). • ls there formal recognition of achievements? This is frequently the m ost critical elem ent in any management system. Without accurate and accessible data, improvement programs wilJ falter. It is also our experience that once sites establish ap p rop ria te monitoring syst em s immediate savings arise. Recently a streamlining of a data management system for a multi-facility organisation identified savin gs of over $50,000 simply by stopping payment for sites no lo nger owned. The sam e system was also used to hi ghligh t variances of m o re than 100% in energy use at different sites, for the same type of facility.
Environmental and Water Industry Specialists • • • • •
asset management water and w astewat er t echnology risk management infrastructure planning strategic partnership
Contact Andrew Osborne or Peter Everist
Tel 03 9694 1200 Fax 03 9694 1211 Web www.fisherstewart.com.au Engineers • Surveyors • Project Managers • Planners • En vironmental Consultants
MELBOURNE • CANBERRA • BRISBANE • JAKARTA • SINGAPORE • HANOI
54
WATER MARCH /APRIL 20 0 0
Auditing and checking are powerful tools when integrated into a full management system. E nergy efficiency initiatives need regular appraisal to ensure they are delivering value. Often the tracking of n1ajor programs alJows diversion of fun ds from one project to another if figures on actual financial returns are m ade available to managem ent.
Improving Energy Performance Improved energy performa nce comes w hen a continuous improvement plan is developed within the fra m ework offered by an energy managem en t system. The plan should address both improvements in energy performance technologies and shortcomings in the energy management system.. The process begins with an assessment of existing energy management practices. To this end, our company developed the One-2-Five® En ergy diagnostic. The diagnostic allows orga nisa tions to assess their performance in each of the key elements of an effective energy managemen t system on a scale from one to five stars, and then develop an improvement plan. The verification step confirms the diagnostic but also iden tifies specific energy saving opportun ities, and coupled with the diagnostic resu lts, leads to the im.provem ent plan. The 3 month and 12 month r ev i ew cycles ensu re tha t m omentum is main tained and are used to keep the energy improve men t plan aligned with business objectives. In the diagnostics completed in the water and wastewater industry we are finding that th e average overall result is around 2 stars. Some of the specific issues raised in these diagnostics have been mentioned above. The depth of the energy management system should depend on the size of the potential benefi ts and potential risks. For example an organisati on with an energy expenditure of less than $200,000 may find that simple procedural controls are t h e most cost effecti ve ap proac h. Organisations with significantly greater energy expenditure w ill find that a more comprehensive managem ent approach is necessary to deliver the desired cost savings and greenhouse emission reductions.
Authors
I
Quality Endorsed Company
A.S90Cl1 LIC1~
SlandM:KAuolfW
Andrew McClee ry is a Senior Consultant with Energetics. H e may be co nta cted at e mail: M ccleeryA@ energetics.com .au. Gordon Weiss as a Principal Co nsultant with Energetics, and may be contacted at weissg@ energetics.com.au