Page 1

Volume 37 No 3

MAY 2010

AWA JOURNAL OF THE AUSTRALIAN WATER ASSOCIATION


The Industry's Most Widely Used Water Modelling Software

Major water network operators like Brisbane City Council, ActewAGL Water Division, SA Water, and Barwon Water all rely on Bentley products every day


water

Journal of the Australian Water Association ISSN 0310-0367

Volume 37 No 3 May 2010

contents REGULAR FEATURES From the AWA Chief Executive

Election Fever

T Mollenkopf 4

CPovey 5

My Point of View Crosscurrent

8

Aquaphemera

Dutch Mission to Australia - see page 16

R Knee

8

Industry News

14

AWA News

20

Events Calendar

28

FEATURE REPORTS Sharing Australia's Water Expertise with the World P Oliver

30

New Water Recycling Centre Launched

34 Sharing Australia's Water Expertise with the World - see page 30

AWA CONTACT DETAILS Australian Water Association ABN 78 096 035 773 Level 6, 655 Pacific Hwy, PO Box 222, St Leonards NSW 1590 Tel: +61 2 9436 0055 Fax: +61 2 9436 0155 Email: info@awa.asn.au Web: www.awa.asn.au DISCLAIMER Australian Water Association assumes no responsibility for opinion or statements of facts expressed by contributors or advertisers.

I

COPYRIGHT AWA Water Journal is subject to copyright and may not be reproduced in any format without written permission of the AWA. To seek permission to reproduce Water Journal materials, send your request to media@awa.asn.au WATER JOURNAL MISSION STATEMENT 'To provide a journal that interests and informs on water matters, Australian and international, covering technological, environmental, economic and social aspects, and to provide a repository of useful refereed papers. ' PUBLISH DATES Water Journal is published eight times per year: February, April, May, June, August, September, November and December. EDITORIAL BOARD Chair: Frank RBishop; Dr Bruce Anderson, AECOM; Dr Terry Anderson, Consultant SEWL; Michael Chapman, GHD; Robert Ford, Central Highlands Water (rtd); Anthony Gibson, Ecowise; Dr Brian Labza, Vic Health; Dr Robbert van Dorschot, GHD; John Poon, CH2M Hill; David Power, BEGA Consultants; Professor Felicity Roddick, RMIT University; Dr Ashok Sharma, CSIRO; and EA (Bob) Swinton, Technical Editor.

AWA

EDITORIAL SUBMISSIONS Water Journal welcomes editorial submissions for technical and topical articles, news, opinion pieces, business

information and letters to the editor. Acceptance of editorial submissions is at the discretion of the editor and editorial board. • Technical Papers and Features Bob Swinton, Technical Editor, Water Journal- bswinton@bigpond.net.au AND journal@awa.asn.au Papers 3,000-4,000 words and graphics; or topical articles of up to 2,000 words relating to all areas of the water cycle and water business. Submissions are tabled at monthly editorial board meetings and where appropriate are assigned referees. Referee comments will be forwarded to the principal author for further action. Authors should be mindful that Water Journal is published in a 3 column 'magazine' format rather than the full-page ' format of Word documents. Graphics should be set up so that they will still be clearly legible when reduced to two-column size (about 12cm wide). Tables and figures need to be numbered with the appropriate reference in the text e.g. see Fig ure 1, not just placed in the text with a (see below) reference as they may end up anywhere on the page when typeset. • Industry News, Opinion pieces and Media Releases Helen Kelton, Editor, Water Journal- journal@awa.asn.au • Water Business and Product News Brian Rault, National Sales and Advertising Manager, Hallmark Editions - brian.rault@halledit.com .au

ADVERTISING Advertisements are included as an information service to readers and are reviewed before publication to ensure relevance to the water sector and objectives of the AWA. Brian Raul!, National Sales and Advertising Manager, Hallmark Editions - brian.rault@halledit.com.au Tel: +61 3 8534 5014 AWA BOOKSHOP Copies of Water Journal, including back issues, are available from the AWA Bookshop for $12.50 plus postage and handling. Email: bookshop@awa.asn.au PUBLISHER Hallmark Editions, PO Box 84, Hampton, Vic 3188 Tel: 61 3 8534 5000 Fax: 61 3 9530 8911 Email: hallmark.editions@halledit.com.au

Eynesbury is the first completely new town to be developed in Victoria for many years, in the arid plains west of Melbourne, and the Eynesbury Development JV is installing third-pipe recycling and other water efficiency systems. The company is preserving many buildings of the original 1870s Eynesbury Station. This water tank was the header tank for the homestead, presumably supplied by a windmill. How it was made watertight is a mystery. Water would be pleased to receive photos and stories of other heritage water assets for publication in subsequent issues. Photo by Brian Labza.

water

MAY 201o 1


water

Journal of the Australian Water Association ISSN 0310-0367

contents

,.

Strategy Selection for Project Delivery - see page 55

TECHNICAL FEATURES(~

Volume 37 No 3 May 2010

_..,'

Comparison of Tertiary Filters Operational Performance - see page 71

INDICATES THE PAPER HAS BEEN REFEREED)

WATER BUSINESS POLICY

Urban Water Reform - Achievements and Challenges

Reported by Andrew Speers

36

P Liggins

40

D Francis, A Dijanosic

48

GFox

55

S Johnston

59

L Ho. K Craig, H Bustamante. G Newcombe

65

MP Thomas

71

K J Hartley, PA Lant

77

S Beckwith, N Ning, J Adeane

84

~ Economic Regulation of the Australian Water Sector - Past, Present and Future Achieving consistency in approach to urban water charging

[ยงI Preparing for a New Regulatory Pricing Framework Getting the most out of the regulatory pricing review process PROJECT DELIVERY & FINANCING

rnll

Strategy Selection for Project Delivery

There is no single 'best' delivery model

~ PPPs in Water There is a strong precedent for water infrastructure projects being delivered via PPPs WATER TREATMENT

~ Removal of Cyanobacterial Metabolites: Laboratory Evaluation of Granular Activated Carbon Successfully mimics full-scale GAG filters WASTEWATER TREATMENT

[ii

Comparison of Tertiary Filters Operational Performance

The cloth filters ' operational performance depended on regular chemical cleaning

[i]

Sludge Settleability in BNR Processes

SSVI increased with primary anoxic mass fraction

Fault Prediction for Wastewater Pump Stations Using advanced pattern recognition techniques WATER BUSINESS

New Products and Business Information. Special Feature: Sludge Drying & Biosolids

87

Advertisers' Index

96

2 MAY 2010 water


8th-10th JUNE 2010 SYDNEY CONVENTION & EXHIBITION CENTRE, DARLING HARBOUR

Register on-line by visiting www.irrigationaustralia.com.au or to obtain more information Ph: (02) 9556 7988

T

CONFERENCE

• Theme: One Water Many Futures • International & National Keynote Speakers • Challenging Topics • Networking Opportunities

T

EXHIBITION

T

HALF DAY WOR

• International & National Exhibitors • Hundreds of Brands • Latest Technology

Co-hosts of the Irrigation Australia 2010 Conference:

~/2 ~

#

...

.····•.

.

IRRIGATION

~_/2 Cooperative Reseorch Centre for

AUSTRALIA

' / ( /~ ' IRRIGATION FUTURES

• •:

Organised by:

. :• •••••• EXHIBITIOOS •.. •.. &TRADE FAIRS


feature article

Sharing Australia's Water Expertise with the World

By Dr Peter Oliver, International Wat erCentre It's a hot February morning on a North Stradbroke Island beach. There are a couple of board riders perfecting their style; you ng holiday mothers sunbathing and a few retirees enjoying the water between the flags. Further up the beach and there is a different energy. Fifty young adults are eagerly talking, beach towels and clip boards in hand . They are working in small groups. Some begin to draw and dig in the sand, some grab pieces of driftwood and other jetsam to use in this work. There's constant discussion about groundwater dependent ecosystems, bore fields and effluent. The discussions contin ue as eighteen different three-dimensional models of North Stradbroke Island take shape in the sand. There are over twenty different accents. The discussion is enriched with perspectives from Australia, Africa, the United Ki ngdom, USA, The Pacific, The Middle East, South Asia, South-East Asia and South America. Concepts and language from engineering, ground water hydrology, aquatic ecology, planning, anthropology, law and economics are seamlessly woven through discussions explaining island water management, and how we need to take a t rans-disciplinary, wholeof-island, whole-of-water-cycle view if we are to manage water on North Stradbroke Island as sustainably as possible. These discussions are part of an orientation field trip for a group of students from t he International WaterCentre's Master of Integrated Water Management (MIWM) program at The University of Queensland.

3 0 MAY 2010 water

IWC Masters students make three-dimensional conceptual sand models.

These students are taking the first steps in an 18-month Masters program which immerses them in a unique, transdisciplinary learning environment focusing on developing an integrated, holistic understanding of the management of water, using Australian expertise and Australian and international case studies. By the ti me these students finish the MIWM program t hey will have been on a variety of field trips. These visits range from water infrastructure and other management issues throughout South East Queensland , to a ten-day field trip in Western Australia looking at aquatic ecology and salinity and catchment management, led by the University of Western Australia's Centre for Excellence in Natural Resource Management in Albany. Current students are doing their third semester research projects throughout Australia and in places as far afield as Nauru, Israel, South Africa and Bangladesh. They have gained water knowledge and honed their water management skills in Australia and are testing them throughout the world, particularly in developing countries. Programs such as this are sharing Australia's water expertise with the world. The first students graduated in 2009 and are already making a global

mark, with Alumni leading the monitoring and evaluation of wetlands management programs for the World Wildlife Fund in Pakistan and developing trans-boundary river basin plans in Ecuador and Peru. The high quality of overseas students enrolling in this Masters Program , the diversity of countries from which they come, and the enthusiasm with wh ich they return home to work in water management , are solid indicators that these graduates wil l make a positive difference to water management internationally. High-quality Australian graduate students are also entering this program, studying both fu ll and part-time. With student numbers on the increase, we can expect their impact on Australian water management to be felt in the near future. "As Australians, when we look at ourselves, we t end to look at the wat er crisis we're in, and feel we haven't been managing the resou rce in an integrated way," comments Mark Pascoe, CEO of the International WaterCentre. "We're not perfect. However, we are managing water with our eye on the whole-of-water cycle, in not only the biophysical dimension, but also in the human, social , cultural, economic and environmental dimensions.

feature articles


+GF+ DISCOVER THE BENEFIT OF DOUBLE FLOW The new ge nera tion diaphra gm valve of fers 100 percent more flow. Wi th consider ably reduce d energy co nsumption. How w as th is achieved? We have opti mised the valve bod y, minimised th e dead space an d maximi sed hygiene. Enjoy absol utely stable pr oc es ses thank s to l inea r flow proper ties. The new diaphragm valve maximises you r benefits. A worthwhi le proposition. The new gener ation of diaphrag m valves from Geor g Fischer Pi ping Systems. Inn ovatio n is profitab le. For types 514,5 15,51 7, 519 and Diastar six, ten and sixteen.

Contact Us for further Information: t : 02 9502 8000 or 1300 130 149 f: 02 9502 8090

e: austra lia .ps@ georgfischer.com www.georgefischer.com.au


feature article "Those who know about us tend to see Austral ia as innovative. They would probably compare us to the Israelis, Swedes and Californians in the way we've embraced water management challenges of today and the future. " Student feedback reinforces this view point. "This IWC Masters Program brings expertise from Australia, but at the same time we learn a lot from lecturers with experience from outside Australia," says current student, Janina Murta from Portugal. "We are always making the bridge between here and the rest of the world . We have learned from Australia's experience and seen the challenges of implementing best practices. We have looked at them critically from technical, economic, institutional and social perspectives." Canadian engineer and program graduate, Diane Cousineau observes that Australia is the only country offering such a well-desig ned and truly integrated curricul um, noting that Australia has some significant water challenges and has come far in changing people's water use habits.

Water practitioners from all over the world come to Australia for IWC's Master of Integrated Water Management.

Learning from students from other cultural contexts is also significant. Diane's experience is that the IWC Masters program can be a good place t o test water policy ideas, explaining that "Cultural, political and gender differences can thwart seemingly coherent water policies when it comes to actual implementation." The beach is a special place for many Australians. It is also part of our special,

national 'outdoor' classroom , from whic h the International WaterCentre and partner university teachin g team are successfully taking integrated water management education to the world. /WC Water Leader Scholarships will open on 1 May for the Master of Integrated Water Management program. For more information go to www.watercentre.org.

r FOR SALE ROTATING MANDRILL for filter production Designed for both low and high pressure vessels at 1.600mm diameter. The design allows for no join in the parralel section of the tank. It allows for the production of both vertical and horizontal tanks from 2 meters to 4.5 meters. The SALE also includes the moulds required for filter tank construction, including the lid and cradle to support the tank.

,.

For further information Email to Gary Hurne at

gdhume@uq.net.au Mobile 0414 522 719 Sunshine Coast Q!_;D

,. r


PIPING Bi. AUTOMATION SYSTEMS Ph: 03 9563 7666 Fax: 03 9563 7722 email:sales@paas.com.au


feature article

New Water Recycling Centre Launched

Australian Water Recycling Centre of Excellence

More than 130 water experts gathered in Brisbane in March for the launch of t he Australian Water Recycling Cent re of Excellence (AWRCE). Launched by Senator Penny Wong, the new centre will operate under the Chairmanship of Mr David Gray, who has significant commercial expertise. He is also the Chair of WaterSecure, the owner of Australia's largest wat er recycl ing project. The Australian Government has contributed funding of $20 million to t he Centre, along with its sister centre, the National Centre of Excellence in Desalination, which was established in Perth last year. In her speech, Penny Wong affirmed the Government's commitment to water recycling and said she hoped the Centre would conduct research that leads to energy efficient, future-focused water recycling technologies. She said, "The Rudd Government is committed to investing in urban water

Senator Penny Wong opens the centre.

Specify and use Link-SealÂŽ seals to seal pipe penetration

measures t hat help secure water supplies through alternative wat er sources that are less dependent on rainfall. "That's why a central part of our $12.9 billion Water for the Future initiative is a $1.7 billion investment towards new infrastructure project s, including desalination, recycl ing and stormwater harvesting, to help take t he pressure off mains water supplies." The launch provided an important opportunity for the Centre to engage with indust ry on the development of its Strategic Research Plan. The engagement was led by Research Advisory Committee Chair, Mr Ian Law, who released four discussion papers at t he launch , designed to stimulate dialogue with industry and research practitioners.

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

PR •jEX

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

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

34 MAY 2010 wat er

L to R: Ian Law (Research Advisory Committee), Senator Penny Wong, David Gray (Chairman).

feature articles


feature article The discussion papers outline potential priorities across the Centre's proposed research themes. Mr Law explains: • Theme 1: Technology, efficiency and integration

This theme has been developed in the knowledge that we already have the technology to produce water of any quality to meet the requirements of a wide range of recycling opportunities. The challenge often faced, however, is how to optimise these technologies to achieve the required quality of water that is fit for purpose. This theme suggests topics that relate to improving efficiency and integration with the aim of developing sustainable water recycling solutions. CEO Mark Donohue.

Mr Law stated that, "As a partnership between research and industry, the potential to create opportunities to increase the uptake of water recycling is real and encouraging and we are very excited about the outcomes the Centre will produce." The launch event and the following day's business breakfast provoked great interest from industry and drew invitations for AWRCE Management to continue the discussions with speaking engagements around Australia. As a result Mr Ian Law and AWRCE CEO Mark O'Donohue are organising a series of forums around Australia in comi ng months. "This is an opportunity for us to continue the discussion with industry and researchers about the purpose of the Centre and to get more feedback about the discussion papers," Dr O'Donohue said. "We hope to involve utilities, health and environment regulators , industry practitioners, and researchers in these conversations," he added. The input received during this tour will assist the members of the Research Advisory Committee in finalising the Strategic Research Plan to present to the AWRCE Board for endorsement. It is a crucial piece of work which will guide the Centre's investment strategy during its formative years. For further information regarding the tour or for a copy of the research discussion papers, please visit the AWRCE website www.australianwaterrecycling.com.au.

Suggested research topics include: • Management of salt and saline effluents in water recycling • Improved online monitoring processes • Optimisation of existing process technologies and trains

• Theme 3: Social, institutional and economic challenges

Water recycling has assumed a recognised and important role in the portfolio of water management strategies in metropolitan and regional areas. The social, economic and institutional components of water recycling are pivotal to achieving sustainability in Australia. The theme notes that recent national appraisals of the state of urban water reform, inclusive of water recycl ing, found that research was deficient in social, institutional and economic factors. The reviews concluded that improved understanding in these research areas is central to increasing the level of recycling application and acceptance. Suggested research topics include: • Decision Support (Institutions and Governance) • Prioritised Investment (Economics of Water Recycling)

• Optimal integration of source waters, technologies and end uses

• Implementation and Evaluation (Social Psychology Research into Water Recycling)

• Innovative and novel technologies for water recycling

• Theme 4: Sustainability in water recycling

• Theme 2: Risk management and validation

This theme proposes that by understanding the complexities of water recycling , and using the principles of sustainability as a platform, more effective outcomes for the water sector will be achieved.

This theme has been developed to further research in the assessment and management of risks when reclaimed water is returned to the water cycle. Much of the theme discussion is related to validation of the performance of the technology applied in water recycling schemes. The theme also identifies that the communication of risk issues relevant to the provision of safe, reliable and affordable reclaimed water are best managed with the appropriate involvement of four key groups of stakeholders - the water professionals, the policy and regulatory decision makers, the independent science researchers, and the consumers. Suggested research topics include:

Central to this theme is the suggestion that water recycling should align with national and international Life Cycle Assessment (LCA) initiatives to help ensure sustainable water recycling systems. The theme also suggests research topics that are focused on the adoption of technologies to achieve more effective and yet sustainable water recycling initiatives. Suggested research topics include: • Alignment of water recycling with national and international Life Cycle Assessment initiatives

• Toxicology and chemistry

• Management of salts and recovery of nutrients

• Measurement and assessment of pathogens

• Expanded use of low energy and energy recovery systems

• Risk assessment and communication

• Reduction and reuse of chemicals and consumables in water recycling.

• Policy and practice

water

MAY 2010 35


water business policy

URBAN WATER REFORM ACHIEVEMENTS AND CHALLENGES Reported by Andrew Speers Abstra ct This article discusses the presentations made at the session on Water Reform at Ozwater' 10. This session complement ed others in the Water Policy Reform stream that ran throughout the second day of the conference. Other papers of particular interest to readers might include those on Water Markets and Pricing, by Colin Reid (183), Shane Lee (184) and Pat Mccafferty (185) and Regulation by Paul Liggins (113), David Francis (1 14) and Arran Canning (115). The papers by Liggins and Francis are published in t his issue.

Introduction If t here is one lesson to be taken from the Ozwat er session on urban water

reform it is that urban water security lies in the diversification of water supplies. And a corollary of this is that each location will require a different mix of wat er sources to meet its needs in the particular circumstance in which it finds itself; one size no longer fits all. The mosaic of opinions and information presented by the four speakers in the Urban Water Reform session of Ozwater combined to provide the audience with a real sense of the achievements gained in urban water and the challenges ahead.

The Speakers Despite being the second speaker in the session, Will Fargher, Acting General Manager Water Markets and Efficiency Group at the National Water Commission, set the scene by reminding the audience

• Potable Water • Remote Well Sites • Booster Stations

Jlccurate & Consistent Hypoch/orite Solution

~ An.~_,,..s -~

-----~~

, ~.

~~-~~

~=-~

Constant Chlor and HTH are. registered trademarks of Arch Chemicals, Inc.

36 MAY 2010 water

of the intent of urban water reform : to secure water services; to protect public health and the environment; and to deliver water services cost effectively. While Mr Fargher observed that these objectives had not been fully achieved, he also noted the progress made in diversifying water sources while maintaining human health protection, and in price reform that has more closely aligned the cost of services with the price charged. Dan Spiller, Acting Executive Director of the Queensland Water Commission, in discussing the developing South East Queensland Water Grid, gave a practical example of supply diversity, explaining that t he linking of trad itional dam waters, with desalinated supplies and recycled water reduces system risk - as a number of these sources are climate independent - and wi ll reduce dependence on dam waters from the current 95% to 78% by 2012.

Providing an 'independent view of water reform' Andrew Speers, National Manager Policy at the Australian Water Association commented that we needed to strive to design 'systems for circumstances'. He also noted, however, that in doing so the widest possible range of options should be considered , including traditional ground and surface water sources, recycled water, stormwater, desalination, rainwater and inter-basin transfers and that the primary criteria for selection of a source should be the extent to which it is sustainable. In this regard he also commented that no sources should be subsidised by governments lest such subsidies promote economically inefficient proposals over more otherwise more viable alternatives. Presenters noted that the diversification of supplies also created the need to look closely at governance arrangements. If the most efficient new sources are to be brought on line there will be a need, as Mr Fargher remarked , to remove regulatory barriers, encourage th ird party access and to improve governance systems overall. Andrew Speers commented that in looking at future water supplies attention should be paid to fully separating operational and

technical features


ITT

(.lOWARA

The SV Series, Great Innovation, Great Performance. The Lowara SV Series of multistage pumps has already set a high standard within the pump industry and now with the inclusion of the new SV33, SV46, SV66 and SV92 this standard has been significantly raised. These new SV's bu ild upon the advanced features of the current SV's with innovative engineering, providi ng marked increases in efficiency, lower noise, easy seal replacement, compatible dimensions, construction options and su itable for use with standard motors, along w ith superior laser cutting and welding technology in the manufacturing process. This wider range extends flows up to 120m 3/ hr and heads up to 300m and combined with the existing SV2, SV4, SV8 and SV16, this ve rsatil e and flexible range of SV pumps can satisfy a wi der range of challenging pumping applications. Ask your Lowara Distributor or Dealer about the SV providing solutions for your pumping requirements. www.itt.com www.hydrovar.com www. lowara.com www.gou lds .com Lowara and Goulds Master Distributors:

,

'

l

Victoria & Tasmania

Queensland

Sth Aust & Nthn Territory

Brown Brother Engineers (03) 9793 9999

Brown Brothers Engineers (07) 3200 6488

WBS Pumps (08) 8344 9927

New South Wales

Western Australia

New Zealand

Papua New Guinea

Brown Brothers Engineers (02) 9671 3666

Advanced Pump Technologies (08) 6143 1000

Brown Brothers Engineers +64 (3) 365 0279

Fimali Limited +67 (5) 852 3658

.. · -~,.

~ \

.•

Engineered for life The "ITT Engineered Blocks" sym bol; and "Engineered for hfe~ are registered trademarks of ITT Industries. Q 2006


water business policy regulatory functions and to ensuring that prices and subsidies remained transparent. He also noted that openness and transparency would help to ensure that governments are not tempted to demand extraordinary dividends from their corporatised water service providers. Mr Fargher also noted that despite progress made in securing urban water supplies, potentially inefficient restrictions are still experienced in urban areas for lengthy periods. These comments were notable in that meeting customers' expectations by delivering a secure, reliable supply of water was one of the goals of the SEQ water grid. Mr Spiller reported that one of the Grid's Explicit Levels of Service objectives is that restrictions should occur one in 25 years on average. This contrasts markedly with the experience in urban areas over the past decade. In his presentation, Tony Slatyer of the Department of Environment, Water, Heritage and the Arts, referred to the water reform process underway over the past 16 years and, in particular, to the recent COAG Work Program on Water which includes a focus on improved urban water planning. He reported that national planning principles have recently been promulgated incorporating the establishment of agreed levels of service (a feature of SEQ's operational framework) consideration of all options for supply, inclusion of the stakeholders in decision-making and the use of correct price signals to balance supply and demand, among other things. There was no dissent by the speakers either during their presentations or during questions from the idea that these principles were desirable and would address some of the concerns about how supplies might be efficiently diversified or how regulation might be improved , but Andrew Speers did note that we need to remember that 'urban' is a term that does not only describe our capital cities. Rapidly growing districts such as Coffs Harbour, Bendigo, Rockhampton, and the like are also urban areas he said, as, even, are small townships. It is important that supply security objectives and level of service goals apply to these areas. Mr Speers commented too that it is important that we not consider urban areas as existing in isolation from rural districts. In fact, he noted, there is a

38 MAY 2010 water

close relationship between the two and many of the issues confronting rural areas - such as the need to improve the operation of water markets - are relevant to urban wat er reform as urban areas may wish to purchase water for their own use. Market transparency will be vital if such trades are to occur in an open manner. Of course no market or water supply system can operate without good information. One of the national planning principles is that good data are requi red . To this end, Mr Slatyer referred to the responsibilities that have now been given to the Bureau of Meteorology to provide regular reports on the status of Australia's water resources and patterns of usage, an annual National Water Account and real-time water availability forecasts.

Conclusion It was clear from all speakers that there has been a paradigm shift in urban water management. Much has been achieved to enhance security and to strengthen governance arrangements but we fall short in a number of areas. Particular targets for reform will include, but not be limited to, the development of Water Plans for each catchment; improvement in transparent cost-recovery including the implementation of national pricing principles; improved planning that is responsive to drought, risk and climate change; and further reform of regulatory frameworks. It was equally clear though that some caution is needed , particularly with regard to reform of instit utional arrangements. As And rew Speers noted, disaggregation of water utilities into retailers, operators, suppliers and planners creates the opportunity for poor interfaces to develop between. the new entities. By way of example, he referred to the need to ensure that information gained by, say, utilities on demand and demand trends is fed back to the various supply entities and to planners, lest future plans be poorly targeted. Urban water reform is a dynamic and challenging field. Responses to changing climate, economic, social and environmental imperatives will require new solutions and new frameworks. In moving forward , however, we must continue to ensure that we deliver safe, secure water supplies efficiently and sustainably.

technical features


The Siemens answer: Siemens TowBro速 sludge removal headers Siemens TowBro速 products can double sludge removal efficiencies. The computer designed Unitibe header provides the ultimate in rapid, uniform removal of final clarifier sludges. Siemens RimFlow速/TowBro速 clarifier technology can lower overall instal lation costs and enhance sludge removal. The unique design, proven in hundreds of installations, ensures effective results, simple operation and reduced maintenance, while delivering energy savings. Answers for Industry. To learn more about Siemens products in our eight solution areas of water, energy, environment, healthcare, productivity, mobility, safety and security ca ll 137 222 or visit www.siemens.com.au/water

Innovation for generations.

SIEMENS


water business policy

~ refere ed pap er

ECONOMIC REGULATION OF THE AUSTRALIAN WATER SECTOR PAST, PRESENT AND FUTURE P Liggins Abstract The term 'economic regulat ion' in the wat er sector covers a range of activities in three separate areas: • Determination/ recommendation of prices • Monitoring/establishment of service standards or target s • Oversight of competition and third party access (where applicable) This paper considers the matters of price sett ing and service standards set out above and compares and contrasts the regimes established in the various Australian states and territories. It goes on to suggest that in jurisdictions where independent regulation occurs there is little difference in wat er authority performance compared to those w here it does not exist. It then outlines the case for and against national regulation.

Introduction In Australia while much is still uncertain about the future of utility services, including water supply, one thing can be st ated with certainty: we have moved in t o a new era where resource scarcity and the demands on service suppliers require subst antially higher prices for end-users. This inevitably leads to a greater political foc us on the activity, with flow-on effect s for regulators as well as consumers. A fundamental factor in regulating this sector is that there has been very little direct competition for supply of water and wastewater services to customers. Consumers have had little choice over their supplier, the source of their water or the level of security of supply. While there are some moves under way t o open up the possibility of alternat ive suppliers, essentially customers remain dependent on monopoly services, making the quality of regulation all the more important t o the economy as a whole and t o the community's satisfaction that they

are getting a fair deal on water prices and service quality. The elephant in this room, of course, is the impact of population growth in an era where most cli mate scientists are continuously sound ing warnings that the southern regions of Aust ralia are going t o be subject to increasingly drier cond itions. As an example of the problem, the population of south-east Queensland is reported by the State Government t o have risen by 33 percent in the 12 years to 2009, placing enormous pressure on power and water services, a situation exacerbated in this area by periods of severe drought. Ensuring the quality and reliability of water supply is an issue that has quickly risen up the public agenda in Queensland and across the rest of Australia in t he past decade. An Essential Research poll (2009), list ed this issue as fourth on a list of 13 that would influence respondents in voting at a federal election. Water was only superseded by the quality of the health system, management of the economy and the quality of education with concern about climate change lying 12th. The water industry nationally has recognised the urgency of meeting this challenge by embarking on the largest infrastructure capital spending program in its history, calculated by the Water Services Association of Australia (WSAA) to amount to $30 billion in the next 10 years. How efficiently this large outlay is managed and how fairly consumers feel they are being charged for what they receive from these developments goes to the core of judgements about t he quality of regulation.

Achieving consistency in approach to urban

This paper was presented at Ozwater'10.

40 MAY 2010 water

water charging.

On the other hand, how effectively suppliers and regulators manage negotiations on the cost of capit al, a major adversarial issue in today's regulat ion of electricity in Australia, for example, is very important in ensuri ng that taxpayers do not bear unreasonable burdens in an industry where most of the sector is government-owned. The lower the assumed cost of capital, the lower t he price that suppliers are permitted to charge their customers and regulators may be tem pted to use this device to help lower t he unpopular cost burdens on the end-users. Government enterprise reform is now approaching its third decade in Australia. A case could be made for evaluation of the result s as they apply to utility services to be pursued as a priority. The tendency to date has been to assess outcomes piecemeal and in silos whereas, there is now a need for a holistic review of how much impact Government Trading Enterprise (GTE) reform has had on infrastructure performance in the water sector. This approach would need to include the evaluation of regulat ion of infrastructure industries. Such a review would include consideration, of three major issues: quality of service - where is it and is it improving or deteriorat ing, do the prices being charged enable a reasonable return to the suppliers, and is the process making adequat e provision for innovation in product offerings and ensuring that long-term planning processes are in place? These are common factors across the spectrum of utility services. Inherent in any such review wou ld be the understand ing that the long-term is more important than the short-term, even though it is the latter that receives the bulk of media and political focus when discussions are taking place on water and electricity supply.

technical features


l tl KDCH

MEMBRANE SYSTEMS

Our eng ineers are a breed apart. Experts in their field . Highly experienced. Passionate about their work.

Go to www.kochmembrane.com/bundamba. They immerse themselves in your situation, your organization and your needs. And they know how to work efficiently on a tight sched ule w ithout sacrificing quality. We bring together more tha n 40 years of:

Download our free white paper on the Bundamba Water Treatment Plant. Discover how the biggest membranes in the industry helped Bundamba become one of the most energy efficient treatment plants in Australia.

• Innovative membrane filtration technol ogy • Standard ized and custom-engineered systems • Superior quality products • Field service supporting nearly 20 ,000 locations w orldwi de All for one purpose .. .your next call.

Experience the Power of E3 ENGI N EERI NG • EXC ELLE NC E • EV ER YD AY

2008 • GWI WATER PROJECT OF YEAR

Also learn how: • Aggressive construction schedule was met (not only met but exceeded by a few weeks) • RO system demonstrated lower fouling rates • Using large diameter elements reduced footprint by 40%.

www.kochmembrane.com Corporate Headquarters: Koch Membrane Systems, Inc., 85 0 Main Street, W ilmington, MA 0 1887 USA T: 1.88 8.677.5 624, F: 1.978.657.5 208 Koch Membrone Systems, Inc . is a Koch C hemical Technology Group, LLC company.


water business policy One of the myths of national debate about utility servic es is that the regulators are somehow above the melee whereas, in fact , they are in the middle of the fray, under pressure from large end-user customers, advocacy groups on behalf of the community, the suppliers, governments and their political opponents. The question of whether customers are getting a fair deal hangs conti nuously over the National Water Commission (NWC) in its role as industry watchdog. Regulatory decisions made today, drawn from assessment of the wide range of factors influencing supply, can be seen very differently in two, three or five years time when one, or more, important facet is perceived in t he light of new circumstances. Ken Matthews, Chairman and CEO of the NWC, observes that, despite a lot of effort on water reform, our supply system is still in trouble. The Commission argues that State governments should be setting long-term, transparent wat er rel iability targets and investing targets and investing in their supply systems accordingly. The NWC also asks governments not to subsidise conventional urban water infrastructure developments, but to require consumers to pay for it, highlighting the political challenge facing regulated and policy makers.

Economic Regulation The term 'economic regulation' in the water sector covers a range of activit ies in three separate areas: • Determination/recommendation of prices • Monitoring/establishment of service standards or targets • Oversight of competition and third party access (where applicable) WSAA (2009) has acknowledged that economic regulation of the industry needs to be improved, not least because future supply wi ll involve a greater role for the private sector, which , as the association says, "will demand a transparent and clearly-defi ned regulatory framework before committing capital."

~ refereed paper

One can not argue with WSAA's call for the adoption of formal, independent price regulation by all governments, using nationally consistent approaches and common principles. However, while there is an increasing movement to bring regulation to a national level - or at least an eastern seaboard level, as with the Australian Energy Regulator - for a number of service sectors, water is one area where economic regulation of urban supply remains the responsibility of individual States and Territories. Th is is reflective of a century or more of government ownership and oversight. It is an environment where up until the 1990s many utilities set prices not to cover all the costs of water services (Government subsidies often filled the gap between revenue and costs). In fact, prices were often in the form of a fixed charge based on the rated value of the property connect ed to the system. The past decade has seen government enterprise reform , driven largely from Canberra and influenc ed in many industries by commercial interests. This has resulted in many utilities becoming more commercial in their operations and structure, however in some areas the trad itional mindset still persists. Trying to achieve uniformity of approach in such an environment and in a federal system is a difficult, complex and drawn out process, but nonetheless one with which the NWC is charged. Independent reg ulation is an important element of the National Water Initiative (NWI) principles, which requires that independent bodies set or review prices, or price setting processes, for water storage and delivery by government water service providers. However, ful l and independent regulation, as applied in the energy industry in real ity only occurs in two States, Victoria and New South Wales, and in the Australian Capital Territory. These arrangements have been in place since 2004 in Victoria, since 1997 in the ACT and since the early 1990s in NSW. The Queensland Competition Authority has new price monitoring roles but it has not undertaken an urban water price review. In Tasmania, the government still makes pricing decisions and the political pressures applying to all administrations was well-illustrated last November when the Tasmanian Government rescinded its water and wastewater pricing decisions after a public outcry.

Ashbma~

s1ma11-Hartley·

Ph 07 3279 3276 email: sludgetreatment@epco.com.au

42 MAY 201 o water

In South Australia at present the Minister for the Murray River and Water Security oversees the operations of SA Water and the Treasurer monitors its financial standing . The bottom line is that the State government determines all prices and directs the Essential Services Commission to review the process after the decision is taken. However, the State government has committed to shifting to independent economic regulation, with legislation due to be enact ed later in 2010. In Western Australia, there has been a move recently to assign

technical features


www.pipeandcivil.com.au

Pipe and Civil is a leader in the civil construction industry with a strong track record in pipeline construction. Located in Brisbane, we also have offices throughout Australia.

At the heart of Pipe and Civil you will find highly experienced, well qualified and motivated staff, enabling us to consistently exceed the expectations of our clients.

For further information: Brisbane - Head Office Tel: 07 3262 4600

Melbourne Office Tel: 08 9288 1805

Adelaide Office Tel: 08 8233 5941

Perth Office Tel: 08 9288 1805


G

water business policy 200 ~ - - - - - - - - - - - - -

i

180

.§. _2 160

!

120

••••••••• _.. +-_____,c..,_ ______,,_ =--

, ---==:::;;;;:;;;;o;;::,1i::-=----

i 100 + +---~ ~·~r·~~::~-~~- --"-r.-.~.-~.-~-~-"-_-_-_-_-_-, - -------------§

80

,.

60 - - ~-~ · - -- ·-- - ·- --

j

•o

t

20

-

+--------------

ACTEW

-

SydneyWater

-

CltyWestWater

- - - South Ea.st Water • • • • • • Yarr a Valley W ater

· ~- - · _- _ ·_ -_ · _· _ _

. - - -- · ---·

}

200 180 .§. 160

i

+ - - - - - - ~~ - - ~~ - -

l uo t----=--==--,,,,!!:.~ :..__~:---- :: --..:;:::-

refereed paper

Brisbane Water -

-

Power & Water

w .a te rCorp · Perth

! ~

.. 'll

f

'o

j

140 120 100 80 60

,o 20 0

~<

J

t

i

::

i

3:

t

.N

.fl

£

2006-07

2005·06

j

2007-08

Figure 1. Average duration of an unplanned interruption- water (minutes).

Figure 2. Average duration of an unplanned interruption- water (minutes) for 2007-8 only.

1000

900 . - - - - - - - - - - - - - - - - - - - 800 + - - - - - - - - - - -- - - - - - ---t-

900 -

800

C

700

"~

600

.2

~

i

~

z

;ii

-

.... .

500 400

~

....

300

---

-

-~.

0 2005·06

2006-07

• CltyWestWattr

• • •• •, Yarra Valley Water

~

100

Sydney Water

- - - South East Water

- .-.=· - ·......... ........................ - ...

200

• • ACTtW

• MelbourneWater

• • Brisbane Water -

• Power& Water

-

• SAWater-total

-

Water Corp · Pe rth

2007-08

700

+--- - - - - - -- - - - - - - - ---t-

"600 1500 i 400 : 300

+ - - - - - - - - - -- - - - -- - ---1+--- - - - - - - - - - - - - - - ---t+---------------+ - - - - - - - - - - - - -- - -

j

z 200 +-1- -- -- - - - - - - - - - - 100 0

~

<

j f

l

.

i3:

~

~

!

5

1

j 1

!E I. . j

1i s

~

! i :I

!

,ll

~

3:

~

]

~

~

~

ij

3:

Figure 3. Net profit after tax ($ per property).

Figure 4. Net profit after tax ($ per property) 2007-08 only.

reviewing prices and services for the south-west of the Stat e to the Economic Regu lation Authority (ERA). While this ent ails a public review, the ERA's findings are then subject to a political decision by the WA Cabinet.

businesses do not earn monopoly profits or provide substandard services while enabling them to cover the efficient costs of operation and maintenance.

In the Northern Territory, the Utilities Commission's role with respect to water and sewerage is mainly confined to licensing prices are decided by t he relevant minister.

Regulatory balance It is important to acknowledge that economic regulation of water supply cannot, and does not, operate in isolat ion from other issues affecting the service, most notably now the environmental challenges, including climate change policy. The crossover between environmental and economic regulation today is not so much a point as a four-lane highway. The key economic performance areas on which regulators judge suppliers in the water sector alone are enough to highlight the uniformity challenges for all concerned: • Affordability • Customer responsiveness and service • Network reliabi lity and efficiency • Water quality • Conservation and the environment • Comparisons of past performance Under the NWI a key goal of urban water supply is to provide safe, reliable and efficient services to cities in a sustainable manner - and this requires balancing economic, social, cu ltural and environmental goals. And, as ACIL Tasman (2008) has observed, economic reg ulation is pursued in order to reproduce the disciplines otherwise provided by com petition to ensure that the regulated

44 MAY 2010 wat er

Each of the performance factors set out above is open to subjective judgements by consumers as well as measurement by regulators. If there is a consensus at all in consideration of these measures, it is probably that there have been mixed results in delivery against them over the past 10 years. To broadly cite Frontier Economics (2008), water service delivery and current pricing arrangements are based on an industry paradigm which does not contemplate competition or the free entry of innovative supply options. There is a centrally determined price imposed on consumers based on a centrally determined plan to meet demand over a certain period. There is no scope in these arrangements for consumers to decide the value they place on a more reliable service - or, conversely, whether they would be comfortable with paying less for a less reliable service.

This leads, of cou rse, to substantial, long-term restrictions on water use when drought overtakes the supply arrangements, and results in the imposition of considerable costs for government agencies administering restrictions, households, businesses and the commu nity at large. As the Productivity Commission (2009) has observed, "the cost of water restrictions are many and varied - they range from structural damage to buildings and time spent on labourintensive methods of wateri ng to the emergence of 'water rage' with neighbours checking the water use of others in ways they would not contemplate for other services." Anecdotal evidence from one country town in Victoria suggests that the banning of hosing of lawns and gardens resulted in an almost immediate increase in injuries from elderly residents being forced to water with buckets and cans.

technical features


~ refereed paper

water business policy

Until State governments can all be persuaded that the best way to get price signals right in water supply is to have them fixed at arm's length by independent economic regulators, the goal of greater national efficiency in this area is likely to remain out of reach.

t

Outcomes of Economic Regulation

~

1400

,

(

1000

i

800

• Customer service

! ., "s

The Commission added: "Setting charges that do not cover costs, including the cost of capital, has implications for performance , particularly when c onsidering future investment decisions within the sector and across the economy."

Pricing In defence of the industry it is worth noting that the low returns were clearly influenced by lower than expected water sales due to the drought. It is likely that the independently regulated businesses are suffering from pre-set prices over a period w hen sales volumes are depressed due to restrictions, while the unregulated businesses have more leeway to adjust prices on a year t o year basis, subject to Ministerial control. One of the possibilities that could be explored by regulators to redress this situation is attempting to fine-tune water prices so they reflect storage levels, although this has a number of practical difficulties, not the least in terms of communication to

• City WestWater

• Melbourne Wate r

•• • •• • Brisbane Water

400

- - - Gokt Coast Water

-

Power & Water

-

• SA W ater· Adelaide

-

Water Corp · Perth

1400 , - - - - - - - - - - - - - - - - - - - - -

1200

li. 1000 ~

800

!"

400

+ - - - - - - - - - - - - -- - - - - - + - - - - - - - - - - - - ------ - - - -

}! 60 +--------------------+-1------------------1-----200

Customer performance

Further analysis is required to establish whether this is because the unregulated businesses are reapi ng monopoly profits from their customers, or whether they are simply earning a normal return on their assets. It may in fact be the latter: the most recent Productivity Commission report (2009) on the 24 water GTEs whose performance it studied suggested that nineteen of them in 2006-07 did not achieve a return on assets that exceeded the risk-free benchmark rate. "This," as the Commission observed, "suggests that these GTEs are not operating on a commercial ly viable basis".

-

Figure 5. Capital expenditure ($ per property).

i

In fact in only one area was there a clear difference between the independently regulated and unregulated businesses where the unregulated businesses have being earning higher net profits (Figures 3, 4).

ACTEW

Sydney Water

2002-03 2003-04 2004·05 2005-06 2006-07 2007-08

• Investment

Financial

-

-

600

200

• Financial

While the data set is incomplete in some areas, there appears no clear difference between customer performance by independently regulated water businesses in NSW, Victoria and the ACT, and those 'unregulated ' businesses in SA, WA, Queensland and the NT. This lack of differentiation exists both at trend and absolute levels. Figures 1 and 2 of the duration of unplanned water interruptions are typical of other customer service indicators:

-

•• • •• • Yarra Valley Water

!

Using data provided in the NWC and WSAA's National Performance Report a brief review of major utility performance in these areas demonstrates that the results are inconclusive ..

..

- - - South East Water

~

Given that in governments as well as in ordinary life, the most persuasive tool is often example, it makes sense to look at the outcomes in water service reg ulation between those jurisdictions where there is independent economic regulation and those where politicians fix the prices. Has performance been better in the former than in the latter? Three areas in particular lend themselves to this comparison:

...,..

.,

1200

~ i >

~

~

;J 3

t;

il:-

u

! !. !. !. ~ >

j i J

~

£

~

~

ij 3

;ii

I

;!

j

Figure 6. Capital expenditure($ per property) 2007-08 only.

Ii;.

1000 900

E

800

8 t

600

~

8

1 ,.'.ll

g

~

700

,-

-

I--

-

>>-

500 400

100

,,,-

0

>-

300 200

-

1--

-

1--

-

I--

~

~ "

ij 3

f

~

I I ul:-

i3 5

~

,S!

;J 3

> .!!

~

t

;!

i 3 ~

~

.;

._

-

>>-

~

"--,

.

:,,

! . !.. 1 ~

3

s

j

fl

i

~

Figure 7. Average residential customer bill 2008.

customers. It is likely to be a more attractive option in the advent of smart metering in the water industry, which may become a viable option if it can piggyback on the communications technology established by energy companies in their smart meter roll out.

Investment Care must be t aken w hen assessing "investment" as higher levels of investment are not necessarily a good thing and may just reflect gold plating or poor project management. They may also merely reflect historical under-investment. However, if anything investment by unregulated businesses may be higher than that of regulated businesses (Figures 5, 6). Finally, given that for unregulated businesses investment appears to be higher, customer performance no different, and financial performance possibly better, it is instructive to examine customer prices. And again there is no clear evidence that bills are lower for reg ulated entities (Figure 7). The analysis above is by no means conclusive and an array of historical factors (including the level of government subsidies

water MAY 201o 45


water business policy for water and sewerage services) will influence outcomes. The sample size is also small.

in theory an open market, being fully cost-reflective, not least when carbon charges start to impact on supply costs.

Nevertheless, given the apparently limited difference in customer performance between reg ulated and unregulated businesses, and possibly poorer financial performance from regulated businesses (at least in times of drought), it is also important to reflect on the fact that regulation is not without cost. For example, we estimate that the Victorian Essential Services Commission (ESC) spent around $3 million on water regulation in 2008/09. The regulated businesses are likely to have spent at least this amount again. Of course costs are not zero in jurisd ictions where Ministers and government retai n control over the price setting and performance oversight. Nevertheless, further analysis is required to conclude that customers are receivi ng value for money from the formal regu latory arrangements.

Even so, the water sector is rapidly heading down the same track. Grids are developing within, if not between, States, and the ACCC and NWC are pursuing less restrictive arrangements for trading across State boundaries. The Murray river can be considered as a giant pipeline interconnecting Queensland, NSW, Victoria, the ACT and South Australia although of course the majority of water used is for agricultural purposes.

A Single National Regulator? Given the inconclusive evidence regarding the benefits of Stat e-based regulation, it is useful to consider whether the desirable end-point of regulatory reform for the water sect or is either a consistent urban regulatory framework applied by State-based regulators, or even a single national urban water regulator. The competition reforms in the electricity and gas sector have led to this outcome and resulted in the establishment of the Australian Energy Market Commission to frame the rules and the Australian Energy Regulator to apply t hem. Of course there are a number of critical differences between the water and energy sectors. Gas and elect ricity grids are substantially interconnected and operate in an almost unfettered manner across State boundaries Further, no water entities operate across Stat e boundaries. And finally there is now a considerable degree of private sector involvement in the energy sector, with t he electricity supply sector having evolved over 20 years in t o an amalgam of govern ment-owned enterprises and private operators. The process of reform has been slow and is still a patchwork of privat e sector involvement, free markets and various forms of regulation. For example, energy ret ailers still have huge concerns about the 'safety net' prices paid by residential and sma!I business customers in what is

46 MAY 2010 water

To take a hypothetical example, is it reasonable for Lower Murray Water Authority's customers to pay a 5 per cent rate of return when downstream SA Water's customers are paying 10 per cent? And this leaves aside the multitude of current approaches to pricing water for agriculture. In the past it has been argued that environmental impacts and the importance of social factors in the water industry have required State-by-State factors to be taken int o account in determining prices and service st andards. However, it is no longer clear that the environmental issues are greater in the water industry than the energy sector, particularly for the urban water sector. The Chairman of the ESC has argued passionately that the social opt imality of a national regulatory arrangement needs to be tested and scrutinised publicly. He has observed that too often the benefit of national uniformity is merely asserted and that national bodies such as the Murray Darling Basin Authority and t he Australian Energy Regulator must "prove t hat they can satisfy the requirements of people and place; and not just the needs of puddles and pylons" (Ben-David, 2009). It makes more sense for the burden of proof to sit with State-based regulation and that national regulation should otherwise be the default position For the short and medium term at least, it appears that there is more to be gained by ensuring that urban water use is covered by the same regulatory and pricing framework, even if by different regulators. And this is certai nly not the case at the moment. While in Victoria the Water Industry Regulatory Order specifies the principles that must be applied to a water pricing decision, there is no equivalent document in other States.

~ refereed paper

Price reviews in the ACT must have regard to specific terms of reference issued by the relevant Minister, which can differ from review to review. Similarly, in its most recent review of Sydney Water's prices the Independent Pricing and Regulatory Tribunal was specifically directed by the Minister to have regard to certain matters. The National Wat er Commission's Steering Group on Water Charges is focusing on achieving consistency in approaches to urban water charging across sectors and jurisdictions. Whatever the regulatory future holds for the urban water industry, it seems reasonably sure that the tentative steps towards more efficient economic regulation taken in recent years in Australia need to be lengthened to giant strides if the service is to come to terms with what the Council of Aust ralian Governments has described as "a major challenge t o ensure sustainable water supply in the face of a drying climate and rising demand."

The Author

Paul Liggins is a director in the economics practice of Deloitte Corporate Finance. He has over 20 years financial and economic experience in natural resources and infrastructure management including the water, gas and electricity industries. Paul has also held a variety of policy roles with the Victorian public service. This included roles with the Essential Services Commission (formerly Office of the Regulator General) and the Department of Sustainability and Environment's water sector group. Email. paul.liggins@deloitte.com.au

References ACIL Tasman, 2008. Institutional Arrangements in the Australian Water Sector Ben-David R (2009, Can economic regulators save federalism in Australia? Submitted for publication by NETWORK: A publication by the Australian Competition and Consumer Commission for the Utility Regulators Forum Melbourne, Australia Can Essential Report (2009) National Issues of Importance, 26 October Frontier Economics (2008) Urban Water Markets, December Productivity Commission (2009) GTE Financial Performance Monitoring Water Service Association of Australia (2009) Position Paper No 03 - Vision for a Sustainable Urban Water Future

technical features


water business policy

~ refereed paper

PREPARING FOR A NEW REGULATORY PRICING FRAMEWORK D Francis, A Dijanosic Abstract Regulatory pricing review frameworks across Australia are currently being extended, reviewed , or developed as a result of structural developments in the water industry, changes to t he relevant legislative process, a desire to improve the effectiveness of the review process, and/or the natural evolution of the reg ulatory pricing review process. This paper examines the proposed changes, with a focus on Queensland, and offers some insights into the process. Water businesses must be fully aware of the changes to ensure they are prepared for t he regulatory review process. Appropriat e preparat ions for a review can facilitate an effective and efficient process and help achieve appropriate outcomes; for the water business, the regulator, and the community. It is intended that this paper might offer some guidance.

Introduction A number of regulatory pricing review frameworks across Australia are currently being extended, reviewed, or developed. These changes have arisen for a number of different reasons including: structural changes in the water industry (South East Queensland and Tasmania); changes to the relevant legislative process (South Australia); a desire to improve the effectiveness of the review process (Western Australia); or the natural evolution of the regulat ory pricing review process. It is imperative for water businesses to be fully aware of the scope of the changes to regu latory frameworks currently being implemented and proposed, to ensure they are prepared for the requirements of the regulatory review process. This paper examines the current state, proposed changes and potential new directions for regu latory pricing review frameworks and provides a brief update

Getting the most out of the regulatory pricing review process. 48 MAY 2010 water

on progress to date in implementing the changes. This paper offers some insights into the process from the perspective of an independent regulatory reviewer, based on our experience and involvement in a number of regulatory pricing reviews around Australia. The typical role of the regulatory reviewer is to act as an independent advisor to the regulator providi ng advice on the water business' strategic, capital and operational processes and the prudence and efficiency of the capital and operating expenditure proposed by the water business. It is intended that the insights in t his paper might offer a guide for water businesses as to what a regulatory reviewer, acting on behalf of the relevant regulator, expects t o assess in order to complete the review process. This paper aims to demonstrate that the appropriate preparations for a regulatory pricing review can facilitate an effective and efficient review process and go some way to achieving the most appropriate outcomes for the water business, the regulator, and t he community.

Review of Regulatory Frameworks This section provides a brief overview of the current regulatory pricing review frameworks in place around Australia, with a particular focus on Queensland. It also provides a brief summary of any proposed changes to these frameworks.

Queensland The Queensland water ind ustry is dominated by the South East Queensland region w ith Brisbane as the central focus. The region is home to the South East Queensland Water Grid, a multi-billion dollar network of interconnected bulk, distribution and retail infrastructure that supplies water and wastewat er services to a population of almost 2.8 million (QWC, 2008 & QCA, 2009). The majority of the water businesses in Queensland are owned and operated by local councils, either singularly or in groups of councils. The source and bulk infrastructure businesses withi n the

South East Queensland Wat er Grid are generally the except ion, with these businesses being State owned statutory authorities. Recent reforms to the industry in South East Queensland have created three new counci l-owned distribution-retail businesses based in three regions. The central entity includes Brisbane, Somerset, Lockyer Valley, Ipswich and Scenic Rim Councils. The northern entity includes the Sunshine Coast and Moreton Bay Councils, wh ile the southern ent ity is made up of the Redland, Logan and Gold Coast Councils. The new businesses take responsibility for the construction, operation and maintenance of the dist ribut ion and ret ail assets from the individual councils. Brisbane City Council's water business, 'Brisbane Water', has already moved across to the new central entity, 'Queensland Urban Utilities' , with the other Councils' water businesses scheduled to transfer across prior to the required date of 1 July 201 0. The Queensland Competition Authority (QCA) is the primary statutory authority for the st at e with its functions defined under the Queensland Competition Authority Act 1997 (the QCA Act). The QCA has responsibility for overseeing prices for water and wastewater related services provided by either government monopoly (or near monopoly) water businesses, where the QCA has the power to recommend prices; or water suppliers, as defined in the Act, w here the QCA has the power to actually set prices for services. Through the current regulatory pricing review process, the QCA either investigates the prici ng practices of the government monopolies or simply monitors the maximum prices charged by them. The QCA can only perform these functions on request from the relevant St ate Government Ministers, however, the QCA has a mandate to undertake an investigation of the SEQWater business without a referral from the Ministers. For the majority of cou ncil-owned water businesses, the current regulatory review framework for water and

technical features


The June 2010 issue of the AWA Water Journal will run the Essential Water Listings Guide in the centre of the publication. The Guide will be an alphabetical directory of companies within the Water Sector and is the most cost effective way of profiling your company and products to the extensive AWA readership. The format of the Essential Water Listings Guide is: • Company Name • Colour Logo • Contact Details • Key Contact • Business Summary (up to 70 words)

• Profile your company to the Water Industry in the most cost effective way • First time advertisers gain maximum exposure at minimal cost • Update your company details if they have changed in the last 6 months • Reinforce your marketing message to the key decision-makers in the Water Industry

Pricing Standard Listing - $250 inc GST Banner Strip Ads across listing page - $1,000 inc GST Rates for larger listings available on request Material Deadline - June 1st

Material Colour logo - print quality, 300 dpi Text - please provide company contact details and business summary (up to 70 words) as a Word doc

For more information contact: Brian Rault National Sales Manager Hallmark Editions +61 3 8534 5014

brian.rault@halledit.com.au


water business policy wastewater is incorporated within the review of the council's Infrastructure Charges Schedules (ICS), as set out in a Priority Infrastructure Plan. The ICS cover all trunk infrastructure including water, wastewater, drainage, roads , open spaces, pathways. With the recent water reforms occurring in the South East Queensland region , the role of the QCA in regulating the three new water businesses created has been investigated and defined . The Premier and Treasurer of Queensland have directed the QCA to recommend an interim price monitoring framework that wi ll apply during the transition period for the three businesses, that is, the period between 01 July 2010 and 30 June 2013 (QCA, 2009,pg iv). The interim framework seeks only to monitor the prices of services provided by the three water businesses to ensure t hat: 1. the businesses do not over-exercise their market power; 2. that information is provided to customers explaining the reasons behind annual increases in prices; and 3. to implement an approach that can be transitioned int o a more comprehensive regulatory review process in the future (QCA, 2009, pg iv). The interim framework has the followi ng functions: to focus on reven ues and pricing policies, to promote disaggregation of costs to align prices, cost s and service levels; and to incorporate trigger provisions for more a comprehensive pricesetting framework should revenues significantly exceed a guideline Maximum Allowable Reven ue for a sustained period (QCA, 2009,pg v).

Odours a problem(~ With 20 years experience, OCS has access to a wide range of technologies for odour control In wastewater, commercial and Industrial applications. Services include:

Investigation and monitoring Design, manufacture, Installation, operation and maintenance of odour control facilities

~ refereed paper

The interim framework also introduces the prudency and efficiency t est to capital and operating expenditure that is common in other jurisdictions. These two tests can have a major impact on the operation of the business. All proposed and historical expenditure needs to be assessed to identify a demonstrated need for the expenditure in line with planning strategies or legislative obligations (the prudency test) and further that the proposed expenditure is cost-effective in comparison to other options considered or market benchmarks (the efficiency test). While the interim framework involves only the monitoring of the prices set for services, it places a specific requi rement on each of the th ree water businesses to submit a fairly comprehensive annual information return covering information on revenues; service standards; demand volumes; the initial regulatory asset base; expenditure to be rolled into the reg ulat ory asset base; actual capital expenditure; contributed, donated and/or gifted asset values; depreciation; indexation; return on capital; and operating costs. Tasman ia

The water industry in Tasmania has recently undergone a significant consolidation with the previous twenty-eight local government owned retail entities and three bulk water authorities being transferred into three, regionally based, local council owned water corporations. Previously the water businesses were dominated by three bulk water entities, Cradle Coast and Esk Water in the north and Hobart Water in the south. The remaini ng local council owned entities were diverse in size and location but required stand alone systems for the provision of wat er and wastewat er services. The Government Prices Oversight Commission (GPOC), the independent regulator which had responsibility for the water industry, had a price monitoring mandate with urban retail prices actually set by the businesses themselves. The Stat e Government can however request that the GPOC undertake an investigation of pricing policies (NWC, 2007). The GPOC has powers under it's enabling legislation to conduct invest igations into the pricing policies and was undertaking such reviews of the th ree bulk water authorities on a three yearly cycle up until 2007 (NWC, 2007). The creation of the three new water businesses as part of the recent reform process has necessitated a change in the regulatory pricing review framework. The GPOC took on a more formal role as the Economic Reg ulator of Water and Sewerage. In this new role the GPOC took responsibility for providing advice on interim pricing and licensing arrangements that would apply until the transition period for the reform process expired . In the future the GPOC will provide advice to the St ate Government on regulatory price det erminations. South Australia

To solve your odour problems today

Call the experts! ((13

so

MAY 2010 water

The water ind ustry in South Australia is dominated by a single utility, the state owned South Australia Water Corporation, wh ich provides bulk, distribution and retail water and wastewater services to the entire state. The regulator in South Australia, t he Essential Services Commission of South Austral ia (ESCoSA) has been undertaking regu latory reviews into the water industry since 2004, however unlike the other states, water and wastewater services are not regu lated services. As such the ESCoSA's role in the regulatory review process has been limited to post-reviews of the State Government's processes for setting the prices of water and

technical features


wastewater charges; that is, the reviews have occurred after t he prices have been set. In June 2009 the State Government released a new strategy document for South Australia entitled Water for Good (Government of South Australia, 2009) which provides a plan to meet the water supply needs for South Austral ia up to 2050. A key component of t his strategy is the expansion of the role of the ESCoSA. This expansion includes the appointment of the ESCoSA as the independent economic regulator for monopo ly water and wastewater service providers in South Australia. Th e strategy also allows for an expansion of the regulatory review framework to include the implementation of operating licences for service providers, prices to be reviewed and set by ESCoSA, and third party access to service provision to be facilitated.

Western Australia The wat er industry in Western Australia is also dominated by a single utility providing bulk, distribution and some retai l wat er supply services and wastewater collection, treatment and disposal services across the state. The Water Corporation is a stat e owned corporation with an appointed board of directors. There are two other service providers that are currently reg ulated, the Aqwest/Bu nbury Water Board and the Busselton Water Board, both of which are counci l owned entities. The remainder of the state is serviced by local c ouncil ow ned businesses however at present these entities are not independently regulated. The regulatory pricing review framework for the state is managed by the Economic Regulation Authority (ERA) however the ERA does not have a formal regu latory role for water and wast ewater services. Rather, the ERA provides an advisory ro le to the State Government, responding to matters referred to it by the State Govern ment. In t hi s role, the ERA does not make decisions b ut provides recommendations and guidance for the State to consider (ERA, 201 0). The ERA does have a regulatory role in licensing the various water and wastewat er service providers across the state and has a regular program of operational performance audits against t he requirements of the operating licences and a review of asset manag ement planning processes. At present, there are no significant changes proposed for the regu latory framework in West ern Australia.

New South Wales The regu lated water industry in New South Wales is centred around the Sydney and Hunter reg ions with the key regulated businesses comprising Sydney Water Corporation, Sydney Catchment Authority, State Water Corporation and Hunter Water Corporation. Water management related services provided by the State Government, t hrough the Department of Envi ronment, Climate Change and Water, are also regu lated at present. More recently, reviews of water and wastewater services provided by Country Energy in Broken Hill have also commenced.

Leading the world in the modernisation of Irrigation Infrastructure Rubicon is leading the world in the modernisation of irrigation systems. Total Channel Control®technology has set a new International benchmark of 90% dist ribution efficiency. The FlumeGate™is t he cornerstone of TCC® and provides accurate flow measurement and control. Rubicon has worked w it h t he University of Melbourne for over 10 years pioneering innovative approaches to water delivery t echnology. Rubicon is Motorola's Distribution Partner, offering MOSCAD and ACE RTU sol utions for management and cont ro l of Water and Wastewat er for urban and rura l Utilities. • • • • •

SCADAConnect'" FlumeGate'" SlipGat e'" SolarDrive'" Total Channel Control® (TCC)~ • FarmConnect'" • Australian Premier Distributor for Motorola MOSCAD and ACE.

The regulator in New South Wales is the Independent Pricing and Regulatory Tr ibunal (IPART). IPART has been the regulator in New South Wales since 1992 and it now has six key functions related to:

--

RUBICON www.ru bicon .com. au

1. price setting for monopoly services; 2. licence administratio n and com pliance; 3. third party access facilitation;

Rubicon Systems Australia Pty Ltd

1 Cato Street Hawthorn East VIC 3123 Austra lia Telephone +61 3 9832 3000

1A Wheeler Street Shepparton VIC 3630 A ustralia Telephone +61 3 5820 8800 www. r ub icon .com.au


water business policy 4. competitive neutrality investigations;

Victoria

5. administration of the Greenhouse Gas Reduction Scheme; and,

The water industry in Victoria covers the whole stat e with all nineteen businesses coming under regulatory control. In the Melbourne metropolitan area there is one bulk service provider and three retail businesses.

6. administration of the Energy Savings Scheme (!PART, 2010). IPART has a regular function reviewing the prices of the regulated businesses on a typical four yearly schedule and also has an annual program of operational performance audits of the regu lated business' operating licence requirements. IPART recently established a formal thi rd party access framework under the Water Industry Competition Act 2006 and has received a number of applications under this framework. To date, there is one licence holder and current applications received relate mostly to small scale non-potable water supply schemes and sewerage infrastructure operations and services. At present, there are no significant changes proposed for the regulatory framework in New South Wales, however some early work is being investigated relating to revenue frameworks for local government.

Australian Capital Territory There is only one water and wastewater service provider in the ACT; ACTEW Corporation, which covers the entire territory. ACTEW is a territory owned corporat ion with two key voting shareholders, the ACT Chief and Deputy Chief Ministers, and a Board which is appointed by the Shareholders. ACTEW is a small organisation with a core team of staff managing high level planning and strategy, capital works contracts, system operations, sustainability, and regulatory compliance. Wat er and wastewater delivery services for capital works and operations is contracted to ActewAGL under the terms of what was, at the time, Australia's first utility joint venture. The regulator in the ACT is the Independent Competition and Regulatory Commission (ICRC) and as a statutory body, it is responsible for the regu lation of prices and access to infrastructure services; and the investigation of competitive neutrality complaints and government-regulated activities. Under the terms of the enabling legislation, the ICRC undertakes regular reviews of the prices charged by ACTEW for water and wastewater services however the ICRC must wait for the release of a ministerial direction to undertake the reviews.

52 MAY 2010 water

In Victoria's regional areas there are th irteen regional water businesses providing retai l or combined retail and bulk services, and there are two rural businesses providing only bulk services. All the water businesses across the state are Stat e Government Business Entities which are guided by skill based Boards of Directors. The businesses provide water and wastewater services to over 4.6 million people across the State. The Essential Services Commission of Victoria (ESC) is, among its other functions, the economic regulator for all water businesses across the state, reg ulating prices and service standards. The ESC has the power and function to make price determinations; regu late standards and conditions of service; develop Codes in relation to its functions and powers; and to require regulat ed businesses to provide information to support the ESC's review processes (ESC, 2010). The ESC has a role undertaking performance comparisons of the urban water and wastewater businesses through the preparation of an annual report. The ESC has also just recently released its final report on its inquiry into the statebased access regime for water and sewerage infrastructure services (ESC, 2009).

Northern Territory The provision of water and wastewater services in the Northern Territory is undertaken by Power and Water Corporation, the first government owned corporation in the Territory. Power and Water Corporation supplies services to 80,000 customers, spread over an area of more than 1.3 million square kilometres. The primary regu lator for the water industry is the Northern Territory Utilities Commission ('the Commission'). The Commission assumed the responsibi lity for regulatory water related services in 2001 with the passage of its enabling legislation. The primary responsibilities of the Commission relating to the water industry are the licensing of water and sewerage service provision. The Commission can be assigned some price or service standard monitoring functions by the relevant Minister

~ r ef e re ed p a per

however their role is a price monitoring role rather than a price setting role. At present, prices for services are still set by the relevant regulatory Minister through the release of a Water and Sewerage Pricing Order. The Commission has had a regular program with regulatory pricing reviews occurring approximately every year. The current pricing order, however, has been set for a period of four years (a similar price path period as those set by other state regulators).

Preparing for a Regulatory Review The previous section provided an overview of the diverse regulatory frameworks in place across Australia and highlighted some of the proposed changes to these frameworks. Even with such a variety of frameworks and requi rements for the regulatory review processes, there are common requirements that if applied prior to a regulatory review might assist in gaining the optimal solution. Regu latory frameworks around Australia are becoming more robust and this is leading to the requirement for water businesses t o publicly report on and justify their expenditure programs, in the process becoming more transparent about how they manage their capital and operating expenditure. Regulatory reviews are taki ng on a new significance in this process and the potential impacts on unprepared businesses can be significant. This section tries to identify these requirements by providing some guidance on how to prepare for a reg ulatory review, taken from the perspective of an independent regulatory reviewer. The regulatory review process operates differently across the various jurisdictions however the fundamental approach is the same. This applies whether the regulatory framework is based on price monitoring and guidance (as currently proposed in Queensland and similar to the system in Western Australia) or price setting (such as the current framework in New South Wales and Victoria). This fundamental approach, at its most basic, involves a review of the prudency and the efficiency of historical and proposed capital and operating expenditure. The assessment of prudence requ ires the reviewer to investigate the reasoni ng behind the levels of expenditure proposed and to determine whether the reasoni ng is appropriate. That is, whether the expenditure is consistent with and clearly

technical features


Q

water business policy

r e fer eed pape r

linked to the water business' obligations, be these legislative driven by a corporate vision or mission statement, or by direct or indirect customer action or advocacy. In essence, the prudency test examines whether there is a need for a project or program of works. The assessment is generally undertaken by reviewing documentation supporting the expenditure under review ranging from corporate level Stat ement s of Corporate Intent, to planning and strategy studies, asset management plans and more specifically to design investigations. In the case of operating expenditure, supporting documentation might include operating budgets and plans, evidence of regu latory obligations, levels of service requirements, or simply evidence of capital assets req uiring ongoing operational expenditure. The prudence of proposed expenditure is generally demonstrated where the following issues can be adequately addressed: (QCA, 2009) 1. the expenditure is required as a result of a legal obligation, new growth, or renewal of existing infrastructure, or where it achieves an increase in the reliabi lity or the quality of service provision that is explicitly endorsed or desired by customers. 2. clear linkages exist between the various levels of planning and strategy documents and the proposed expenditure. A business should be able to easily demonstrate how expenditure on a particular project's assets relates to the achievement of business obligations such as a component of the business' corporate goals and objectives or to legislative requirements such as water quality. 3. sufficient supporting documentation is available for each project to address each stage of the project or asset life cycle. Th is would include documents covering the planning, feasibility, design, construction and operational phases of the life cycle. The provision of these types of documents during the regulatory review process goes a long way to

demonstrating the prudency of expenditure.

endorsement at the Board level demonstrates a robust approach.

In NSW, the Independent Pricing and Regulatory Tribunal defined a similar test for its assessment of capital expenditure. The prudence t est assesses whether, 'in the circumstances that existed at the time, the decision to invest in the asset is one that [the agency], acting prudently, would be expected to make'.

Where the expenditure relat es to an increase in requ ired base service levels, evidence that might be provided could include benchmarking results showing comparatively lower performance in targeted areas.

Further, the prudence test assesses both: 'the prudence of how the decision was made to invest; and the prudence of how the investment was executed (that is, the construction to delivery and operation of the asset), having regard to information available at the time. ' (I PART, 2009) While the assessment of prudence by the regulatory review is intended to provide assurance that the proposed expenditure is appropriate, it does not constitute a fu ll assessment of all the opt ions considered by the proponent. The assessment is reliant on the experience of the reg ulatory reviewer to gain sufficient understanding of the process undertaken and to provide as detailed a review as the available time permits. Demonstrating, for the regulatory review process, that expenditure is required may involve the provision of documents such as correspondence from the relevant Minister outlining t he legislative req uirement, or growth projections showing likely population increases in areas targeted for capital works. It is expected that planning and strategy studies will have been completed which demonstrate a clear need for a solution and that identify all reasonable options that would fulfi l the identified need. A progressive options analysis, that assesses, ranks and prioritises all of the options using a basic triple bottom line or equivalent approach is also expected. A regulatory reviewer will also expect to see evidence of an appropriat e and robust governance framework shadowing the process. The involvement of, and sponsorship from , appropriate executive managers and relevant approvals and

We would expect that the outcomes of the benchmarking and proposed actions would be clearly linked to corporate objectives to achieve higher service performance, supported of course by t he customer's willingness to pay for such an increase in servic e levels. Where an increase in service levels is driven by a change in legislation, the prudence of the expenditure is generally not questioned and the focus of the review shifts to the efficiency of the expenditure proposed to fulfi l the legislative requi rement. Where expenditure relates to improvements that are endorsed or desired by c ustomers, typica l evidence that might be provided includes overwhelming positive responses from customer surveys or outcomes from customer consultative groups. In these cases, the scope of the surveys and the consultation undertaken must include the full impacts of the proposed e xpenditure inc luding the impact on service prices. The efficiency test for expenditure is generally made when the following can be demonstrated: 1. the proposed works are deemed to be the most appropriate solution after having due regard for alternative options, which may include other capital or o perating options or non-expenditure options such as water conservation, demand management or adjustments t o service levels. 2. the cost estimates for the proposed works have been benchmarked against market rates using unit rate dat abases with appropriate consideratio n of issues affecting costs such as market indices and escalation factors. Demonstrating that capital or operating expenditure proposed is efficient may involve the provision of supporting

Ila/crow

Sustaining and improving the quality of people's lives Contact David Martin (03) 8682 3926, or visit www.halcrow.com/australasia water

MAY 2010 53


G

water business policy documents including feasibility studies or options analysis studies where multiple options are considered which detail the options considered , the conditions under which options were assessed, and reasoning for the choice of the optimal solution. It is critically important to provide cost estimates which are as detailed as possible at this stage of the process. The assessment of options needs to be done on estimates which are robust and that allow for all components of the proposed options. This should include items such as risk, contingency, owner's costs, consu ltant's fees, contractor margins, independent reviews, and any other miscellaneous works. An appropriate, or at least a consistent, level of accuracy is required to undertake the comparative analysis of the options. The exclusion of any costs at this stage of the process, should be robustly defended, with any uncertainty taken into account through appropriate risk or contingency allowances. Relevant documents can also include independent peer reviews of costs; evidence of the consistent use, and regular updating, of unit rates databases; and supporting documentation for the determination of contingencies, allowances, and escalation factors. Careful consideration of these latter factors is especially important where market conditions are volatile; whether the volatility be related to increases or decreases in the market or industry. In the majority of cases, the information contained in the types of documents that are listed here can be presented concisely in a detailed business case, a consolidated document that, wh ile alone might not necessarily provide all of the detail required, can provide the structure around which a response to the requirements identified here can be formulated. Extensive referencing of the supporting documents is essential to provide the level of detail required for a robust case. A regulatory reviewer would expect to see business cases for all projects, with the level of detail included reflective of the project size and/or associated risk factors. A number of suitable business case templates exist and this paper does not recommend one particular example over another, however one observation is offered and that is to ensure that the maximum amount of detail possible goes into the business case. This is one example where it is certainly not the case that less is more.

54 MAY 2010 water

The assessment of efficiency is a critical part of the regulatory review process as ultimately this assessment by the regu latory reviewer w ill either confirm the level of expenditure proposed or will recommend w hat the reviewer believes is an efficient level of expenditure to meet the needs first identified at the planning stage. This part of the process provides the greatest opportunity, and the greatest risk, to achieving the optimal outcome from the regulatory review.

re f e r eed paper

So if there is one final statement made in concluding this paper on how to prepare for a water regulatory pricing review, it wo uld be that less is definitely not more in th is area. Businesses need to develop as much detail on their capital and operating expenditure programs as befits each project. What is needed is a clear and logical process supporting the demonstration firstly of the need for the project, and secondly that the proposed works are the optimal method of fulfilling this need.

Conclusion This paper has provided a brief overview of the current water regulatory frameworks in place around Australia. In addition, current and proposed changes to the frameworks have been briefly outlined. The discussion outlines the various charact eristics of the frameworks in each state and highlights a diverse range of processes. From single service providers to multiple large and small scale businesses and from full state regu latory coverage to those with a distinctly metropolitan focus. A range of levels of regu lation were also presented from simple price monitoring and comp liance functions, to guidelines or recommendations for pricing, and finally to enforceable price determinations; from setting operating guidelines to full operational compliance auditing, comparative performance reporting, and benchmarking. With this background, the discussion identifies some common threads in all of the regulatory frameworks; some guidelines that may assist a business in preparing for a review and may help in achieving the optimal outcome. The common requirements identified were two tests used extensively in the regulatory review process, that is, the prudency test and the efficiency t est. In their most basic sense, these tests were defined as: 1. Prudency test - is there a real and demonstrable need for the project? 2. Efficiency test - is the solution / expenditure proposed the optimal way of fulfilling the identified need? To assist in responding to these two tests, various examples of supporting documentation that a regulatory reviewer would expect to examine were identified. One example was highlighted as being one of the most suitable forms of supporting documentation, and that is the detailed project business case.

The Authors David Francis is a Senior Engineer and Market Sector Manager, Water Advisory for Halcrow in Australia providing a range of advisory services to water businesses including assistance in preparations for regulatory reviews. francisdav@halcrow.com. Ana Dijanosic is a Senior Consultant and Market Sector Manager, Water Regulatory for Halcrow in Australia advising independent regulators around Australia. dijanosica@halcrow.com.

References/Further Information ESC [Essential Services Commission]. 2010. Website reference http://www.esc. vie .gov .au/ public/Water/Our+Role.htm ESC [Essential Services Commission]. 2009. Inquiry into an Access Regime for Water and Sewerage Infrastructure Services: Final Report Government of South Australia, 2009. Water for Good: A plan to ensure our water future to

2050 IPART. [Independent Pricing and Regulatory Tribunal). Review of the Department of Water and Energy's (DWE) water management expenditure (RFQ No W03/2009), page 3. IPART [Independent Pricing and Regulatory Tribunal]. 2010. Website reference http://www.ipart.nsw.gov.au/about_us/ functions.asp. NWC [National Water Commission]. 2007. Water Storage and Delivery Charges in the Urban Water Sector in Australia. National Water Initiative Steering Group on Water Charges NWC [National Water Commission]. 2008a. Urban Water Pricing: National Water Commission position. National Water Commission Position Statement 2 July 2008. NWC [National Water Commission]. 2008b. Approaches to urban water pricing. Waterlines Occasional Paper No 7 July 2008. QCA (Queensland Competition Authority). 2009. Draft Report SEQ Interim Price Monitoring Information Requirements for 2010/11. Queensland Competition Authority, OLD, Australia. QWC [Queensland Water Commission]. 2008. http://www.qwc.qld.gov .au/ Commissioning+ the+Water+Grid. Updated November 2008.

technical features


0

project delivery & financing

refereed paper

STRATEGY SELECTION FOR PROJECT DELIVERY G Fox Abstract 1,600

This paper describes the process used by Sydney Water for selecting the most appropriate delivery strategy for a capital project. The delivery st rategy is selected based on the project's characteristics. The process is illustrated with case studies from Sydney Water.

l

1,400 1,200 1,000

~"

..e

800

600

Context

400

Sydney Water has a large asset base with 21 ,000 km of water mains, nine water filtration plants, a 250 ML a day desalination plant, 24,000 km of sewerage pipes and 29 sewage treatment plants. Total assets are valued at over $12 billion (2008-09). It has a large capital investment program to renew, improve and expand these assets.

200 0

2000-2001

2001-42

2002-43

2003-44

• Better allocation of risks • Less prescriptive contracts • More incentives and innovation

Objectives and scope

Since 2000, the enhanced strategic procurement direction has provided overarching guidance in selecting the delivery strategy for each project in the capital program.

Overview A project is successful if it meets the business and project objectives, provides value for money and effectively manages risks. The delivery strategy selected needs to support the achievement of the objectives and deal most appropriately with the risks. It also needs to work

There is no single 'best' delivery model.

2006-47

2007-48

2008-09

impacting on the delivery of the project. Both the project objectives an d the project risks are inputs to decisions on bundling and the project delivery strategy, as described below.

Constraints

Select delivery strategy

Figure 2. Delivery strategy selection process (source: adapted from Beer & Fox).

• Co-operative behaviour with suppliers • Less adversarial.

2005-46

Figure 1. Sydney Water capital infrastructure expenditure.

In 2000 Sydney Water identified that it needed to enhance it s strategic procurement direction to deliver a rapidly increasing capital program (see Figure 1). The outcome was the adoption of the following measures designed to improve the efficiency and reliability of delivery of the capital program: • Reduced supply chai n

2004-45

within the identified constraints and suit the level of project complexity. Industry research by Blake Dawson in 2006 and 2008 confirmed that selecting a contract delivery model aligned to project objectives was essential to the ultimate success of a project. The process used by Sydney Water to select the delivery strategy for a capital project is shown in Figure 2.

Objectives and risks The corporate and/or project objectives are set by stakeholders and are described in the business case. The project will be considered successful if it meets its objectives. Conversely, risks prevent the achievement of project objectives. Project risks need to be assessed to identify the significant risks

The decisions on bundling and project delivery strat egy also need to consider any constraints including time, budget, resources and market conditions. In particular, Sydney Water conducts a market analysis to confirm that the construction industry is capab le of delivering the contract. Factors considered in the market analysis include the market structure, the level of market competition, supply chain arra ngements and its value to the market as a customer. Sydney Water considers if contractors have the capabilit y to deliver (resources, ski lls etc) and whether the

water Future Features JUNE - Desalination, agricultural use, water trading AUGUST - Water treatment, advanced oxidation, demand management, international projects SEPTEMBER - Wastewater treatment, water sensitive urban design, environmental water management

water

MAY 2010 55


project delivery & financing size and type of contract would be attractive to the market.

Bundling Bundling is the aggregation of compatible projects with similar

characteristics into larger contract packages. It has a number of benefits, including: • Faster project delivery • Delivery of a larger capital program

~

refereed paper

• Economies of scale • Reduced supplier interfaces • Reduced management and administration costs • Attracting larger and more mature contractors. Sydney Water considers the following compatibi lity factors when deciding whether to bundle projects into a single contract package: • Dollar value of the project • Required timing for delivery • Funding implications • Type of service being provided • Scope and nature of the work • Complexity of the work • Potential for new technology and innovation (R&D) • Risk profiles

Case study: SPS Upgrade Program Sydney Water operates over 600 sewage pumping stations (SPS). The SPS Upgrade Program comprised civil, mechanical and electrical upgrades of pumping stations. It had previously used multiple, small value contract packages to deliver 19 SPS upgrades per year. Regulatory requirements meant an acceleration of the program to upgrade 240 pumping stations over a three-year period. Each SPS upgrade had similar characteristics, objectives and scope of work. Sydney Wat er decided to bundle the remai ning 240 SPS upgrades into a single $220M contract. This was its first major use of bundling for capital project delivery.

• Environment, commu nity and stakeholder requi rements. Two case studies are provided that illustrate when it is appropriate to bundle projects.

Delivery models Available project delivery models are described in publications from the NSW Construction Agency Coordination Committee (CACC), Queensland Department of Public Works and Australian Constructors Association (ACA). These publications describe the features, applications, strengths and weaknesses of various delivery models, including: • Construct only • Design development and construct (DD&C) • Design novate and construct • Design and construct (D&C) • Design construct and maintain (DCM) • Design build operate and maintain (DBOM) • Managing contractor

Case study: Sewer rehabilitation

• Construction management

Sydney Water has a network of 24,000 km of sewerage pipes. It has an ongoing program of rehabilitation of the pipes' interiors. An assessment of the market indicated that there are few specialist contractors and that Sydney Water is a dominant national client. In the market there is a large ongoing contractor investment in R&D, new technology, equipment and staff training. Although each rehabilitation project has similar characteristics, bundling into a single large contract package would have had a detrimental impact on competition in the market. Sydney Water developed a delivery strategy of bundling small projects into multiple medium-sized contract packages. The size of each contract package is based on contractor capabi lity and is growing as the market develops.

• Engineer procure and co nstruction management (EPCM)

56 MAY 201 o water

• Al liance • Public private partnership (PPP).

Evaluation of delivery models There is no single "best" delivery model that is suitable for all projects. The most appropriate delivery model needs to be

technical features


Q

refereed paper

determined on a project-by- project basis. The approach used by Sydney Water is to bundle projects where appropriate and then to shortlist t he above generic delivery models by considering the follow ing project c haracteristics:

project delivery & financing Project characteristic

Weight%

Complexity and interfacinQ issues Clarity in scope definition and outcomes Ability to identify and quantify risk Opportunity to introduce innovative technoloQy

35

Weighted score

30

25 10 100 Total score Normalised score = total score/50

• Complexity and interfacing issues • Clarity in scope defin ition and outcomes

Score 0-5 (O=low 5=high)

Figure 3. Scoring matrix (source: adapted from ACA).

• A bility to identify and quantify risk • Opportunity to introduce innovative techno logy. Each bundle of w ork is assessed against the above factors us ing the scoring matrix show n in Figure 3. The normalised score from Figure 3 is then plotted on the suitability scale in Figure 4. The range of potential delivery models (i.e. shortlist) for a given bundl e of work is indicated by the position on the suitability scale. The multi -crit eria an alysis method is used to further evaluate the shortl isted delivery models. Project stakeholders participate in a facilitated workshop to

Normalised score

Suitability matrix

10 , - - - - - - - - , 9 8 7 6 5 4 3 2

1 0

Project delivery model Category 3: score > 7 select from:

PPP

• •

Alliancing Managing contractor

Category 2: score 3-7 select from: • Design and construct • Design novate and construct • Construction management Category 1: score < 3 • Select traditional contract

Figure 4. Suitability scale (source: adapted from ACA).

water MAY 2010 57


~ ~ refere e d paper

project delivery & financing consider the shortlisted delivery models. Evaluation criteria are developed based on t he project objectives (e.g . certainty of delivery by t arget date) and risk t reatment measures (e.g. increased operat or engagement, f lexibility to cope w ith variable quantities). Weightings are assigned to the criteria based on thei r relative importance to the project. Each d elivery model is evaluated and scored for its performance against the evaluation criteria (see Figure 5). The d elivery model with the highest w eighted score is selected for the project.

Evalua~ion criteria

Wei htln g g

Deliverv model 1 Deliverv model 2 Deliverv model 3 Raw Weighted Raw Weighted Raw Weighted score score score score score score

Total Ranking

Figure 5. Multi-criteria analysis (source: adapted from CACC).

As a final chec k, the preferred delivery model is reviewed to identify any weaknesses, such as a low score in the mult i- criteria analysis. The generic model can be strengthened by project-specific tactics such as special conditions of contract, the use of incentives etc.

Achievements Sydney Water has successfully implemented its enhanced strategic procurement direction and delivered an increased cap ital program. There has been a reduction in the number of contracts used, with just seven contracts accounting for 90% of the 2008-09 capital infrastruct ure program expenditure. There is a mix of contract models used to deliver t he prog ram as shown in Figure 6. Sydney Water's biennial supplier survey consistently shows that it is considered as a highly valued and trustworthy client and that suppliers want to do more business wit h it.

Conclusions Client organisations need to consider project objectives, risks and constraints when selecting a project delivery strategy. The success of a project will be enhanced if the delivery strategy matches the characterist ics of t he project.

EPCM 1% Design build operate & maintain 33%

Alliance 48%

Design & construct 18%

Figure 6. 2008-09 Capital delivery models.

sa

MAY 2010

water

Case study: North Head STP North Head Sewage Treatment Plant (STP) is the 4th largest sewage treatment plant in Australia. It serves a population of over one million people and treats about 300 million litres of flow a day. Sydney Water identified multiple civil, mechanical, electrical and control improvement projects t hroughout the STP. The projects involved extensive and complex works integrated into a fully operational plant. After considering the characteristics of each project, Sydney Water decided to bundle most projects into a single $SOM contract package. After evaluating delivery options, the price competitive alliance delivery model was selected.

The Author

Beer R 2008, Developing a Comprehensive Contract Selection Methodology & G uidelines, Alternative Contract Selection Conference, IQPC, Sydney. Blake Dawson Waldron 2006, Scope for Improvement: A Survey of Pressure P oints in Australian Construction and Infrastructure Projects.

Greg Fox is Cont racts Manager withi n the Asset Solutions Division of Sydney Water. He is responsible for Sydney Water's procurement and contracting policies, standards and procedures. He was a member of the NSW Construction Agency Coord ination Committee when the Procurement Methodology Guidelines for Construction was developed. Email: greg. fox@sydn eywater.com.au.

References Aust ralian Constructors Association 1999, Relationship Contracting: Optimising Project Outcomes, Sydney.

Blake Dawson 2008, Scope for Improvement: A Report on Scoping Practices in Australian Construction and Infrastructure Projects. Construction Agency Coordination Committee 2005, Procurement Methodology Guidelines for Construction, NSW Department of Commerce. Fox G 2009, Responding Positively to Ne w Market Conditions Through Adaptive Contracts, Contract Selection and Risk for Major Projects Conference, Tonkin Corporation, Sydney. Queensland Department of Public Works 2008, Procurement Strategy and Contract S election, Queensland Government.

technical features


~ refereed pap e r

project delivery & financing

PPPs IN WATER S Johnston Abstract In early 2007 Deloitte released its global research study of the experience with public-private partnerships (PPPs) across different sectors of social and economic infrastructure, including water, wastewater and waste. Titled Closing the infrastructure gap: the role of publicprivate partnerships, it provided a global overview of PPPs including how, why and when PPPs should be considered as a delivery model, which countries have greatest experience with infrastructure and services delivery via PPPs, wh ich sectors of infrastructure have seen greatest deployment of PPPs, and the circumstances under which different types of PPP delivery models have been used and with what success. The follow up, Closing the infrastructure gap: Part II - to be released Q1 2009 - specifically focuses on aiding the public sector deal with the st rategic and operational issues involved with private sector involvement in the delivery of infrastructure projects. Specific quest ions addressed in the research include: • What are my choices in terms of the role of the private sector? • How do I decide what role the private sector should play? • What are the risks and considerations in selecting the role of the private sector? • What is being done in other jurisdictions? • What is the cost of involving the private sector? • What might a deal structure look like for incorporating the private sector? The goal of the research is to provide a framework that helps public officials select the appropriate role or involvement for the private sector in infrastructure projects based on: risk qualification/ assessment, the costs associated with different delivery models, and the desirability of retai ning control over the infrastructure. This paper is an updated version of the paper presented at Ozwater'09 by Roger Black.

Table 1. Water PPP projects by region and type (1991-2007). Region

Total No. Concession of Projects

East Asia & Pacific Europe & Central Asia Latin America & the Caribbean Middle East & North Africa South Asia Sub-Saharan Africa

282 61 193 15

Divestiture

New Projects Service & i.e. Management DBO/B00 Contracts & etc Leases

8 5 12 0 0 0

146

24

116 8 109 0 1 2

42 6 4 2

12 41 30 9 4 20

584

236

25

207

116

9

7

Source: World Bank Group, Private participation in infrastructure database

This paper draws from the findings of bot h Deloitte research papers, including leading examples from around the world , and applies the research to the provision of infrastructure and services in the water, wastewat er and waste sectors.

Introduction: Current Status of PPPs in the Water/Wastewater/ Waste Sector Globally While traditionally the sole province of state and local government, water and wast ewater management has become a growing area for the application of PPPs globally. The drivers for this are many and vary from country to country, although in broad t erms typically differ between the developed and the developing worlds, and the water-rich versus the water-scarce. In the developed world early entrants into various types of PPPs included the Netherlands, a number of Canadian municipal governments and Ireland. • In the Netherlands a 30 year concession - with a total contract val ue of €1.58 billion - was awarded by the Water Board of Delft land in 2002 for the design, construction, and operation of a new wastewater treatment plant, and the refurbishment and operation of an existing wast ewater treatment plant.

There is a strong precedent for water infrastructure projects being delivered via PPPs.

• In Canada, ageing water and wastewater systems requiring renewal expenditure of more than $28 bi llion prompted the municipalities of Moncton, Hamilton & Dartmouth to consider and develop PPP f inancing mechanisms to deliver water services. • In Ireland more than 100 wat er and wastewater PPP projects - m ost of them 'design-build' (DB) - are either operational or in construction and planning. In the developing world private participation in the water secto r began to be hailed as a solution to chro nic failures of coverage and services during the 1990s. Between 1990 and 2005 private investors committed over USO 50 billion to more than 380 water infrastructure projects in developing countries. The PPP experience in provi d ing water and wastewater infrastructure in many ways replicated the PPP experience in other sect ors of social and economic infrastructure. Contracts often reflected excessive optimism by both private investors and governments, an d the socio-political difficulties of raising tariffs to levels covering costs were often underestimated. After a number of major water project contracts proved insufficiently robust to survive the financial crisis, several international operators lost much of their appetite for further investment in developin g cou ntries. More recently, continued necessity combined with the lessons and experience of earlier ventures is resulting in a maturing of the PPP market following the initial boom and subsequent

water MAY 201o 59


project delivery & financing contraction. Stakeholders, public and private, are growing more aware of both the benefits and risks involved, and are looking for contractual arrangements best suited to the nature of the inherent risks and the socio-political context. So before dealing with the specific characteristics of water, wastewater and waste infrastructure - including dams, pipelines, water grids, water treatment plants, water recycling plants desalination plants, waste treatment/sewerage plants - it is worth outlining the range of PPP options and the general learnings t o date regarding the conditions which underpin PPP success.

Private Sector vs Public Sector The defence of PPPs generally is not the focus of this paper. Whi le failures have been headline news - and typically involved the failure of the private operator and the loss of invest or capit al rather than the failure to deliver the contracted infrastructure for public use numerous reports support the effectiveness and utility of PPPs. Whatever the drivers for involving the private sector in greater partnership in the provision of much needed infrastructure and services then - and these can range from the nakedly political to the utterly financial - t he ideal PPP should combine the characteristic and arguably distinctively different skills of the public and private sectors t o maximise the delivery of public good. In short, the regulatory and economic development expertise which is typically the preserve and expertise of the public sector, and the management, innovation , finance raising, budgeting and on-time and on budget delivery skills which, at best, are hallmarks of the private sector.

Common Forms of PPP where New Infrastructure is Required In general terms a PPP refers to a contractual agreement formed between a

~ refereed paper

Table 2. Water PPPs by subsector (1991-2007). Subsector

Segment

Treatment plant

Potable water and sewerage treatment plant Potable water treatment plant

Utility

Project Count

Sewerage treatment plant Sewerage collection Sewerage collection and treatment Water utility with sewerage Water utility without sewerage

12 120 163 1

9 215

64 584

Source: World Bank Group, Private participation in infrastructure database government agency and a private sector entity that allows for greater private sector participation in the delivery of public infrastructure projects. Compared with traditional procurement models, the private sector assumes a greater role in the planning, financing, design, construction, operation, and maintenance of public facilities. Some of the most common PPP models used for new projects include: • Design-Build (DB)/ Build-Transfer (BT): Under this model the public sector contracts with a privat e partner to design, and build, a facility in accordance with the req uirements it sets. Upon completion the public sector assumes responsibility for operating and maintaining the facility. • Design-Build-Maintain (DBM): This model is similar to Design-Build except that the private proponent also maintains the facil ity. The public sector retains responsibility for operations. • Design-Build- Operate (DBO)/ BuildTransfer- Operate (BTO): Under this model, the private sector designs and builds a facility. Upon complet ion, the title for the new faci lity is transferred to the public sector, while the private sector operates the facility for a specified period. • Design-Build- Operate-Maintain (DBOM)/ Build-Operate-Transfer (BOT): Th is model combines the

responsibilities of design-build procurements with the operations and maintenance of a faci lity for a specified period by a privat e sector partner. At the end of that period, the operation of the facility is transferred back to the public sector. • Build-Own- Operate-Transfer (BOOT): The public sector grants a franchise to a private partner to fi nance, design, build and operate a facility for a specific period of time. Ownership of the facility is transferred back to the public sector at t he end of that period. • Build- Own-Operate (BOO): The public sector grants the right to finance, design, build, operate and maintain a project to a private entity, which retains ownership of the project. The private entity is not required to transfer the facility back to the public sector. • Design-Build-Finance-Operate/ Maintain (DBFO, DBFM or DBFO/ M) : Under this model , the private sector designs, builds, fi nances, operates and/ or maintains a new facility under a long-term lease. At t he end of the lease term, the faci lity is transferred to the public sector. (NB In some countries, DBFO/ M covers both BOO and BOOT)

Common Forms of PPP Involving Existing Infrastructure In addition to delivering new infrastructure, PPPs can also be used for

N ew projects

... Existi ng services a n d facilities

Source: The Nationa l Council for Pu blic Private Partnerships

Figure 1. PPPs and responsibilities.

60 MAY 2010 water

technical features


~ refereed paper

existing services and facilities. Models include: • Service Contract: The public sector contracts with a private entity to provide services the public sector previously performed. • Management Contract: A management contract differs from a service contract in that the private entity is responsible for all aspects of operations and maintenance of the facility under contract. • Lease: The public sector grants a private entity a leasehold interest in an asset. The private partner operates and maintains the asset in accordance with the terms of the lease. • Concession: The public sector grants a private entity exclusive rig hts to operate and maintain an asset over a long period of time in accordance with set performance requirements. The public sector retains ownership of the original asset, wh ile the private operator retains ownership over any improvements made during the concession period. • Divestiture/privatisation: The public sector transfers an asset, either in part or in full, to the private sector. Generally it will include certain conditions with the sale of the asset to ensure that improvements are made and the community continues to be served. Figure 1 identifies the differences in the levels of public and private sector responsibility involved in each of the models.

Lessons for the Water Sector from the Global PPP Experience Our research of the PPP experience globally - across all sectors of social and economic infrastructure - holds useful insights for the development and application of PPPs in Australia for water, wastewater and waste. PPPs have proven to be an effective infrastructure delivery tool - under specific conditions! The key to success is the role of the public sector. Most commonly project 'failure' has equated to the financial failure of the private operator. A significant part of designing an enduring market, and therefore private sector willingness to fund and undertake risk, ent ails improving public sector capacity to execute and manage sustainable partnerships. Poor PPP outcomes can be seen to have emanated from:

project delivery & financing • Poor setup. The success or failure of PPPs can often be traced back to the initial design of PPP policies, legislation, and guidance. A common mistake is placing so many restrictions and conditions and expectations of risk transfer on the private sector sponsor and agencies involved that a financially feasible deal becomes impossible to structure. Another is having unrealistic expectations of PPPs-thinking that they provide "free money" or that they're the solution to all problems. • Lack of clarity about project objectives. Sponsors sometimes lack consensus about the purpose of, and expected outcomes for, the project. Public offi cials may then try to compensate for this failure by 'overspecifying' inputs. • Too much focus on the transaction . Public sector may mistakenly view PPPs merely as financing instruments when in fact they represent a very different way of working. This can lead to a poor operational focus and service outcomes. • Inappropriate risk model applied to project. Much of what differentiates the various PPP models is the level and nature of risk shifted to the private sector. A com mon mistake is the attempt to transfer 'demand' risk - i.e. the amount of use the infrastructure will receive - to the private sector, even when the private contractor has no control over demand factors. • Lack of internal capacity. Even when the public sector is supported by external advisers, many tasks cannot be outsourced, and often the agency does not have the skill sets internally to manage complex PPPs or the dedicated team required to address the time intensive upfront structuring needs. • Failure to realise value for money. This failure occurs when the borrowing and tendering costs associated with PPPs are not sufficiently offset by efficiency gains or when government officials don't have a real understanding of how to test value for money. • Inadequate planning. Without taking proper account of the market in the planning phase, the public sector may come out with more projects than bidders creati ng a non-competitive environment. On the flipside, too few projects can result in industry moving on to a more active jurisdiction.

Setting Up Successful PPPs A step-by-step guide to designing and implementing PPPs is beyond the scope of this paper (although further det ail can be found in Closing the infrastructure gap - Parts1 &2). However the lessons learned from water and other infrastructure projects delivered so far suggest several strategies for successful execution of these partnerships. First, the public sector needs a full life-cycle approach or framework that pays adequate attention to all phases of a PPP - from policy and planning, to the transaction phase, and then to managing the facility. The goal of such an approach is to avoid the problems of poor setup, lack of clarity about outcomes, inadequate internal capacity, lack of interest from the private sector, and an overly narrow focus on the transaction. Second, a strong understanding of the range of possible PPP models can help the public sector achieve the proper allocation of risk even in conditions of pronounced uncertainty about future needs. Proper risk allocation allows the public sector to better tailor PPP approaches to specific situations and infrastructure sectors.

A Life-Cycle Approach It is essential to get the fi nancial terms of the deal right. Equally critical however is getting stakeholder buy-in; managing the change process; correctly allocating risk; developing the legislative and regulatory framework; and analysing the long-term effects of the project on the sector as a whole. This means developing a holistic view of the infrastructure project's entire life cycle from the very outset. A life-cycle view helps to get better 'buy in' from all parties involved. It also provides a framework for evaluating whether the solution is the most appropriate for the public over time. Without such a holistic view, public officials will be unable to plan in advance for key considerations that-if not properly accounted for -can stymie efforts to move beyond the transaction stage. A life-cycle approach best ensures the interest of the government agency that retains ownership and ultimate responsibility for the asset throughout the life-cycle. While many experts emphasise the transaction phase of PPP

water

MAY 2010 61


project delivery & financing

Policy and planning phase

Sequential activities for Infrastructure delivery

Key activities

1. Cond1t1on of infrastructure financial situation 2. Leg1slat1on/regulat1on 3. Leadership policy and proJect management 4. Planning· environmental assessments and proJect opportunities 5. Communications internal and external with maior stakeholder groups

[SJ

refereed paper

Transaction phase

1 Transaction process 2. Shortlist qualified bidders 3 Risk transfer and value for money 4. Payment mechanism/performance 5. Request for proposal 6 Finalise proJect agreement 7. Preferred bidder selection and negot1atIons 8. Financial close

1. TransrtIon to construct>on (e g des1gn/bu11d) 2. Construcuon and monitoring 3. Facility operation (contract and relat1onsh1p management) 4. Evaluate whether promised benefits materialised 5. Maintenance· hard and soft service prov1s1on 6. Asset hand back

Establish objectives.

Establish a real istic time frame .

Monitor construction.

The objectives a government establishes for the PPP proiect form the fou ndation for evaluating options and allows it to communicate a consistent message regarding the purpose of the program. Time spent fu lly exploring ob1ect1ves and core values regarding the government's roles and responsibilities will avoid missteps later in the process.

ProJect ob1ec11ves, the budget, market interest, the amount of risk shifting, project size, and the structure of the deal all affect the t1mel1ne for the project delivery.

Many entItIes believe that once they have entered into turnkey contracts w ith concessionaires their responsibility for construction monitoring and oversight has been transferred. The public will continue to hold the public sector accountable for the successful delivery of the project, however, so it is cntJCal to establish sound monitoring programs throughout the construction phase without creating additional project risks.

Evaluate alternative financing structures.

This evaluation should start with an understanding and analysis of the existing debt alternatives within the state. By preparing a range of financial alternatives, the agency can articulate to its stakeholders wha t might be accomplished with traditional financing and what innovative financing structures are available and perhaps necessary for project feasibility.

Secure the best value for money.

A fundamental objective in any project Is to secure the best value for money. Creating comprehensive financial models that allow you to evaluate value for money from both a qualitative and quantitative perspective is a critical component of this process. Establish performance standards.

This often entails using penalties and revvards to achieve the desired behavior. Care must be taken with both rewards and penalties since they can drive unintended consequences. Setting performance standards will also help to develop the best payment approach for each project.

Monitor the concession

Under traditional procurement approaches, monitoring substantially ends at the completion of construction. In the case of a PPP procurement. the contract monitoring needs to be far more soph1st1cated because 1t is required to address a wide range of issues relating to finance, operations and maintenance over an extended period of time.

Communicate the benefits.

A strategic commun1Cations plan that explains the benefits of the program can prevent the discourse from being defined by detractors and focus discussion on economic benefits (such as congestion relief and improved movement of goods) as well as social benefits such as faster and more reliable commute times). Build market interest.

There should be an appropriate number of projects coming into the market at the right pace to ensure that constructors and facility management firms have the capacity and fina ncial ability to keep pace with the potential projects.

Develop a draft proiect agreement.

These agreements are included with the request for proposal (RFP) and help to identify issues bidders may have before the selection of the successful bidder.

Prepare staff

Most jurisdictions are used to undertaking these projects on their own. While PPPs may reduce the need for additional staff to do inhouse design and engineering work, current staff are required to provide proiect management and long-term oversight.

Establish construction governance.

Establish the concession

Large infrastructure construction projects should have effective governance and controls In place before the project begins In order to avoid cost overruns, scheduling delays and litigation.

It's important that effective project governance models are established and that skilled individuals are in place during both the construction and concession phase.

governance model.

Figure 2. Project lifecycle and activities. transactions, the success of the project is actually heavily dependent on a sound policy and legal framework, effective risk allocation, a well-executed procurement process, strong project management, and close attention to the concession phase. A life-cycle perspect ive helps governments understand how decisions made during different phases will affect the long t erm success of the project. For example, the way a project is monitored will be determined largely by how much risk is transferred to the private sector during the transaction construction and concession phases.

62 MAY 2010 water

Figure 2 shows the t hree major phases in the lifecycle of an infrastructure project and the key tasks and activities involved in each stage.

Recent Developments In PPP Delivery Models A number of variation s of the PPP model have been developed in recent years in response to the challenges faced in specific situations and sectors. • Alliancing . Under this model, the public and private sector agree to jointly design, develop, and financ e the project. In some cases they also

work together to build, maintain, and operate the faci lity. • Bundling. Contracting with one partner to provide several small-scale PPP projects in order to reduce the length of the procurement process as well as transaction costs. • Competitive Partnership. Several private partners are selected, in competition w ith each other, to deliver different aspects of a project. The contract allows the public sector to reallocate projects among partners at a later date, depending upon performance. The public partner can also use the cost and quality of ot her

technical features


Q

project delivery & financing

refe r eed paper

Low

Medium

High

The public sector is unsure about the infrastructure it needs (or even what is possible), let alone when or how it wishes to have it delivered.

The public sector knows the kind of infrastructure it needs, but is less certain about the timing and exact extent of work in wishes to undertake.

The public sector knows with confidence either the condition of the assets and/or the future asset and service requirements at a detailed level.

Figure 3. Certainty classifications. partners' outputs as a benchmark for all partners. • Incremental Partnership. The public sector contracts with a private partner, in which certain elements of the work can be called off, or stopped, if deemed unproductive. The public sect or can commission work incrementally, and reserves the right to use alternative partners if suitable. • Integrator. The public sector appoints a private sector partner, the integrator, to manage the project development. The integrator arranges the necessary delivery functions and is rewarded according to overall project outcomes wherever possible, with penalties for lat eness, cost overruns, poor quality, and so on. The integrator has a less direct role in service provision and in some cases is barred from being involved in direct delivery at all. In other cases, the integrator is appointed to carry out the first phase

of work, or specified works but is then barred from carryi ng out subsequent phases of work t o remove the potential for conflict of interest between achieving best value for the public sector and maximising private returns through the supply chai n. • Joint Venture. A joint venture company is set up, a majority of w hich is owned by a private sector partner. The public sector selects a strategic partner through a competitive process that in cludes a bid to carry out the first phase of work. The typical contract is for 20 years. Subsequent phases are commissioned by t he public sector partner, but carried out by the strategic partner using the fi rst phase of work as a benchmark to determine the appropriateness of future costs. The United Kingdom has used a variant of this model, called local improvement finance trust (LIFT), for its hospital PPPs.

Yes

High

Yes

Are the elements of work heterogeneous?

No Low

Can work easily, be separated into

The following key questions should be asked prior to choosing the model : • How confident are you now about the type of infrastructure and services that are needed over the next 10, 15, or 20 years? • How likely is it that the needs of citizens in this area wi ll change? • How likely is significant policy change? • How easy is it to specify what will be needed? • In which sector is the PPP approach going to be employed? • How confident are you in the supplier of the service and how much control do you wish to retain? • Can risks be transferred or wou ld better outcomes be achieved through risk sharing?

The level of certainty the public sector possesses about its infrastructure and service requirements should be a key Private developer determinant in the scheme choice of model. This includes certainty about the external environ ment, including the policy environment, as well as the capacity of contract performance standards and Integrator realities and incentives to higher outputs. Competitive partnership

Selecting an appropriate model Do assets have high residual value?

Choosing the Right Model

Yes

discrete clements?

No )

Joint venture

c::- Yes

Figure 4. Decision tree.

Alliancing

Figure 3 classifies certainty into three cat egories, being low, medium and high. A high level of certainty suggests that the public sector can shift substantial co ntrol and risk to the private sector (the best options are Private Developer Scheme,

water

MAY 2010 63


project delivery & financing Design-Build -Finance-Operate/Maintain, or Conventional Procurement). The integrator, joint venture, or competitive partnership models should be considered where certainty is more limited. The alliancing or incremental partnership models would be more appropriate when a low level of certainty exists. The decision tree in Figure 4 prov ides some guidance regarding t he most appropriate model in particular circumstances . This list of models is by no means exhaust ive; any decision to choose o ne model over anot her shou ld always be derived f rom a robust appraisal of the options, based on the specific circumstances in which the project is being developed.

Characteri stics of Water Sector: Implications for and Current Examples of PPPs While it shares characteristics with some other sectors of social and economic infrastructure t he water, wastewater and waste sector has some unique characteristics which need to be factored into consideration in terms of PPPs.

C hallenges • Market Competition. There are a smal l number of market participants in the wat er sector. Th is is due to the high market concentration, and overspecification. • Substantial procurement costs . High procurement costs and high uncertainty about the availability of technology require a contractual framework with shorter procurement times t hat fosters innovation. •

Uncertainty. The condition of assets in existing facil ities may result in an increase in project costs.

• Scale. The size of the project may not allow for efficient use of private finance. Also, contract s involving water supply of less t han 40 million m 3 suffer from unrealised economies of scale. Conversely, when the level of water supply exceeds 400 million m 3 , the operation wi ll suffer from diseconomies of scale. • Politics. Water and wastewater are often seen as falling squarely under the publ ic sector domain. Public employees may have deep concerns for their welfare under the new management.

64 MAY 2010 water

[SJ

refereed paper

Table 3. Some current water PPPs by type Project

Size

PPP type

Status

Sydney Desalination Plant (value $1 .76 billion)

250 megalitre/day DBO

Contract awarded 22/06/07

Bahrain, Tubli WWTP expansion (value $133 million)

230,000m3 to 350,000m 3

080 & 5 year operating contract

RFP issued 31 /12/08

Egypt, New Cairo WWTP

250,000m3

BOO - 20 years

RFP issued 01 /12/08

Egypt, Abu Rawash WWTP

800,000m3

BOO 20 or 25 years

In progress

China, Harbin Qunli WWTP (value $56.8 million)

150,000m3

DBOM 30 years

Contract awarded 30/11 /08

China, Wenzhou Xipian, Zhejiang Province WWTP

100,000m3

DBOM 26 years

Contract awarded 20/11 /08

Mexico, Guadalajara, Agua Prieta WWTP (value $150 million)

734,000m3

DBOM

RFP issued 01/11 /08

Source: Global Water Intelligence, volume 9; issue 12, December 2008. Because of th is, most water projects tend to be leases or operating contracts as t hey allow the private operator to concentrat e on improving t he utility's operational efficiency and viability while leaving t he public authority in charge of raising investment financing. Greenfield project s are still common in treatment plants however.

Solutions Applying the analytical assessment of PPP models outlined in brief in the foregoing (and in more detail in Closing the infrastructure gap: parts 1 & 2) can help overcome some of the challenges in the water sector. For example, the public sector can reduce the length of the procurement process and attract companies with stronger financial and operat ional capacity by using a bundl ing approach. This saves procurement time and effort as t he public sector is no longer required to contract w ith d ifferent private partners in delivering individual small-scale projects. A key challenge in t his sector is that t he consumer is generally not exposed t o the f ull cost of water. Movi ng to full cost pricing of water utilities before moving to a PPP approach can help to avoid rate shocks that may derail the project.

Conclusion PPPs are an effective way of delivering infrastructure projects, and draw upon the strengths of both the public and private sect ors. There are a number of different PPP models that can be utilised, and

sufficient plan ning and investigation shou ld be undertaken t o ensure the correct model is implemented. There is a stro ng precedent for wat er infrastructure projects being delivered via PPPs. Key learnings from previous projects must be taken on board and thinking creatively about the delivery model can aid in the successful procurement of this very much needed sector of infrastructure.

Acknowledgment This document was derived from the author guidelines used for al l AWA conferences and events.

The Author

Steve Johnston leads infrast ruct ure services within Deloitte's Economics practice. sjohnston@deloitte.com.au.

References Allen Consulting Group, November 2007, Performance of PPPs and traditional procurement in Australia. Booz & Company, 2008, Public-Private Partnerships: A New Catalyst for Economic Growth. Deloitte Touche Tohmatsu, 2006, Closing the Infrastructure Gap: The Role of PublicPrivate Partnerships.

Global Water Intelligence, volume 9; issue 12, December 2008. Philippe Marin and Ada Karina Izaguirre, September 2006, Gridlines - private participation in water. The World Bank & PPIAF, Private Participation in Infrastructure Database, http://ppi.worldbank.org/explore/ ppi _exploreSector.aspx?sectorlD=4.

technical features


~ refereed paper

water treatment

REMOVAL OF CYANOBACTERIAL METABOLITES: LABORATORY EVALUATION OF GRANULAR ACTIVATED CARBON L Ho, K Craig, H Bustamante, G Newcombe Abstract Cyanobacteria can prod uce metabolites which are problematic for the water industry as they can impair water quality. Of concern are metabolites which impart tastes and odours and those which are toxic. Granular activated carbon (GAC) filters can be an effective barrier for the removal of these metabolites; however, like all adsorption processes there is a finite adsorption capacity. Furthermore, the GAC filters can harbour organisms within them which may biologically degrade these metabolites, potentially extending the lifetime of the filters for removal of these contaminants. Consequently, tools which can be employed to ascertain the effectiveness of GAC filters during various operational stages would be of value for water authorities. This study utilised laboratoryscale GAC column experiments to evaluate the performance of a full-scale GAC filter for the removal of a range of cyanobacterial metabolites and selected water quality parameters. The laboratory GAC column was shown to be able to mimic the full-scale GAC filter for the removal of dissolved organic carbon and geosmin, validating its application as a successful monitoring and performance testing tool.

Introduction The onset of climate change coupled with drought and a general rise in temperatures has resulted in more frequent detection of cyanobacteria (blue-green algae) in water sources wh ich were not previously significantly impacted by these organisms. The prevalence of cyanobacteria in water supplies is problematic for water authorities as they can impair water treatment processes and subsequently the quality of drinking water. Of major concern are the metabolites that these organisms produce, some of wh ich impart taste and odours which can result in consumer complaints and the general

perception that there is a problem with the treatment process. In addition, another class of metabolites of a toxic nature can be produced by these organisms, and these have the ability to severely compromise human health. In dissolved (extracellular) form, cyanobacterial metabolites are somewhat recalcitrant to conventional water treatment practices and require more advanced treatment regimes (Lalezary et al., 1986; Rittmann et al., 1995; Nerenberg et al., 2000). Activated carbon is generally the treatment option of cho ice in Austral ia, in particular, powdered activated carbon (PAC) due t o its ease of process control (Cook et al., 2001; Ho and Newcombe, 2005). However, when cyanobacterial blooms become a frequent occurrence, dosing PAC may not be economically viable due to the high costs associated with continuously dosing this material. Therefore, long term treatment options must be considered to cater for the forecasted increase in proliferation of cyanobacteria. Granular activated carbon (GAC) has long been considered a viable treatment option for the removal of organic compounds from water, including cyanobacterial metabolites (Pirbazari et al., 1993; Lambert et al. 1996; Newcombe et al. , 1996, 2003; Ho and Newcombe, 2007; Elhadi et al. , 2006). This is because GAC can remove compounds via two mechanisms, adsorption and biological degradation. GAC filters can effectively remove cyanobacterial metabolites by both mechanisms, although removal by biological degradation may be site specific and dependent upon the indigenous organisms present. Like all adsorption processes, GAC has a finite adsorption capacity. Several studies have reported varying

Successfully mimics full-scale GAC filters.

performances of GAC filters, in particular, the length of their effectiveness. Newcombe et al. (1996) reported that with an empty bed contact time (EBCT) of 20 mins, 18 months was the maximum time that their GAC could be expected to reduce 2-methylisoborneol (MIB) concentrations to below odour threshold. On the other hand, other studies have reported GAC bed lives in the order of years for the effective removal of MIB (Yagi et al., 1988). At present, there is no definitive way to accurately predict how effectively a full-scale GAC filter is functioning for the removal of cyanobacterial metabolites. Therefore, a simple tool which could evaluate the performance of a GAC filter for the adsorption of these contaminants would be useful in determining when t he GAC should be regenerated or replaced. Research conducted at the Australian Water Quality Centre (AWQC) has shown that small laboratory-scale GAC columns can be successfully applied to estimate the removal of organics (including cyanobacterial metabolites) from water by GAC fi lters at pilot- and fu ll-scale water treatment plants (Newcombe, 2002; Ho, 2004; Ho and Newcombe, in press). The basis is that the laboratoryscale GAC columns are operated at the same EBCT as that of the pilot- or fullscale GAC filters. This paper describes the application of these laboratory columns t o assess the

OZWATER1 l OBEST POSTER The poster entitled "Characterising Treated Secondary Wastewater For Drinking Purposes Following Reverse Osmosis Treatment", from a wide ranging team in Western Australia, was awarded the best poster. It is our intention to publish a paper based on t he poster, but it will not be ready until the June issue.

water MAY 2010 65


~ re f ereed p ape r

water treatment performance of a full-scale GAC filter at the lllawarra water filtration plant (WFP) in New South Wales, Australia, for the removal of the cyanobacterial metabolites MIB, geosmin, microcystin and saxitoxins. Pertinent water quality parameters such as dissolved organic carbon (DOC) and UV absorbance at 254 nm (UV254) were also monitored. GAC was sampled routinely from the full-scale GAC filter and evaluated in laboratoryscale column experiments over a period of one year (April 2008 to April 2009). The lllawarra WFP is a 210 ML per day plant operated by Veolia Water Australia under a build, own and operate (BOO) contract for Sydney Water Corporation. The plant practices contact filtration using ferric chloride, a cationic and nonionic polymer coagulation before filtration on deep bed dual media filters. During January 2007 a taste and odour incident occurred in the lllawarra distribution system that appeared to be associated with the source water. Apart from optimising the chemical doses the only short-term treatment option involved the use of GAC in the existing dual media filters. The media (Australian filter coal/sand) was partially replaced in one of the WFP's ten rapid gravity filters with GAC media. The purpose of this was to determine if the GAC filter was more effective than the dual media filters in removing a range of water quality parameters, in particular, mitigating issues caused by cyanobacteria and their metabolites. Performance testing of GAC filters has the ability to allow water utilities to monitor the trends in removal of cyanobacteria metabolites. This will give operators knowledge of the appropriate time for GAC replacement, prior to a severe cyanobacteria challenge that may compromise water quality. Overestimating the operational lifetime of the GAC filters poses a significant risk to drinking water quality during episodes of high cyanobacterial numbers in raw water, as there is the potential for breakthrough of these contaminants through the filters. It is envisaged that by utilising the information from this study, GAC filter media could be replaced over a period of time ensuring the security of water quality while minimising disruption to the plant and supply.

Experimental Methods Materials and reagents MIB and geosmin were purchased from a commercial supplier (Ultrafine Chemicals, 66

MAY 2010

water

Table 1. Characteristics of the GAC and dual media filters at lllawarra WFP. Parameter

Dual media

GAC (coal based) Australian filter coal

Effective size (mm) Uniformity coefficient

Sand

2.0

1J

0~

<1.4

<1.3

<1.4

UK) and separately dissolved in ultrapure water (Millipore Pty Ltd, USA) to prepare stock solutions. Aliquots from the stock solutions were then dosed into waters at a target concentration of 25 ng L·1 of each compound. Samples for M IB and geosmin analyses were pre-concentrated using a solid phase microextraction syringe fibre (Supelco, Australia) and analysed on a 7890 Gas Chromatograph System with 5975C VL Series Mass Selective Detector (Agilent Technologies, Australia) against quantified labelled internal standards (Ultrafine Chemicals, UK). Full details of this method have been documented by Graham and Hayes (1998). The detection limits for MIB and geosmin were 4 and 2 ng L·1 , respectively. Purified microcystin-LR (MCLR) was purchased from a commercial supplier (Sapphire Bioscience Pty Ltd, Australia) and dissolved in ultrapure water to prepare an MCLR stock solution. Aliquots from this stock solution were then dosed into sample waters at a target concentration of 5 µg L·1 • Prior to analysis MCLR was concentrated from sample waters by C18 solid phase extraction according to the methods described by Nicholson et al. (1994). A high performance liquid chromatographic (HPLC) system consisting of a 600 pump controller, 717 plus autosampler and 996 photodiode array detector (all from Waters Pty Ltd, Australia) was employed for MCLR analyses. Sample volumes of 50 µL were injected into a 150 x 4.6 mm Luna C18 column (Phenomenex, Australia) at a flow rate of 1 ml min-1 • Concentrations of MCLR were determined by calibration of the peak areas (at 238 nm) with that of an external reference standard (Sapphire Bioscience Pty Ltd, Australia). The method has a detection limit of 1 µg L· 1 . Five saxitoxin variants were used in this study: C1, C2 , GTX2, GTX3 and STX. These were extracted and purified from a natural bloom of Anabaena circinalis that occurred in Myponga Reservoir, South Australia. The purified toxin solution had a profile characteristic of Australian strains of A circinalis where the variants C1 and C2 were predominant with smaller quantities of GTX2, GTX3 and STX variants (Velzeboer et al., 2000).

Analyses were conducted using a HPLC system comprising a 600 pump co ntroller and 71 ?plus autosampler with postcolumn derivatisation and detection using a 2475 multi "J.. fluorescence detector (Waters Pty Ltd, Australia). Full details of the method are given in Rositano et al. (1998). Concentrations of the saxitoxins were determined by calibration of the peak areas with that of certified reference standards (Institute of Marine Biosciences, National Research Council, Canada). Conversion factors (Oshima 1995) were used to express the toxicity of the sum of the variant s as saxitoxin toxicity equivalents (STX-eq) due to t he differing toxicities and concentrations of the individual saxitoxin variants. The method has a detection limit of 0.5 µg L· 1 (as STX-eq). DOC measurements were conducted on an 820 Total Organic Carbon Analyser (Sievers Instruments Inc, USA). UV25 4 measurements were carried out on a UV-1201 UVNIS Spectrophotometer (Shimadzu Corporation, Japan). Prior to DOC and UV analyses, samples were filtered through 0.45 µm membrane filters. Full-scale GAC monitoring Samples were taken over a period of a year (April 2008 to Apri l 2009) from t he influent and filtered water of the full-scale GAC and dual media (DM) filters at the lllawarra WFP and analysed for MIB, geosmin, MCLR, saxitoxins, DOC and UV254 . Characteristics of the GAC and DM are presented in Table 1.

Laboratory-scale GAC column trials Settled water (post coagulation, p rior to filtration, DOC= 3.2±0.2 mg L·1, UV254 = 0.059±0.007 cm· 1) and GAC samples from the fu ll-scale filter were obtained from the lllawarra WFP on the fol lowing dates: 7th Apri l 2008, 5th August 2008,

Table 2. Conditions of the laboratoryscale GAC column trials. GAG Bed Height Column I.D. GAG Volume

EBGT Flow Rate

15 cm

2.5 cm 78.5 cm 3

6 min 12.5 ml min·1

technical features


G

water treatment

r e f ereed pap er

24th Oct ober 2008, 15th January 2009, and 15th April 2009. For each trial, the respective GAC sample was packed into glass columns (length of 30 cm, I.D. of 2.5 cm) at a bed height of 15 cm. The column was fed with the settled water sampled from the respective dates of the GAC. The settled water was spiked with MIB, geosmin, MCLR and saxitoxins and passed through the GAC column to correspond to an EBCT of 6 min. Influent and effluent samples were routinely taken from each of the col umns over a period of two weeks commencing soon after arrival of the GAC and water samples and analysed for MIB, geosmin, MCLR, saxitoxins, DOC and UV absorbance. Conditions of the columns are described in Table 2.

100

GAC

- • -DOC - • - UV2s. Dual Med ia - • - DOC - • - UV2s.

90

80 70 "O Q)

a:

50

C: Q)

40

l)

Q) [J_

30 20

Results and Discussion

60

> 0 E Q)

10

'. _·-~-W.--- -~- __,,_ \

_.,,;:::;::/\:~•~1 ....

A_.

· -

·-

Full-scale GAC monitoring Influent and effluent water samples were obtained over a period of one year from t he full-scale GAC and OM fi lters at the lllawarra WFP. Initially the GAC filter was effective for the removal of DOC and UV254 ; however, after approximately 3 months, the removal of these parameters was similar between t he GAC and OM filters, indicating the adsorption capacity of the GAC had been red uced for these natural organic material (NOM) components (Figure 1). The GAC filter was able to initially remove 40% and 70% of the DOC and UV254 , respectively, and as expected th is removal efficiency decreased with time due to the exhaustion of adsorption sites. At the end of the study (April 2009) the GAC filter was removi ng approximately 20% and 35% of DOC and UV254 , respectively, which is exactly half that of what it was removing initially in Apri l 2008. These findings complement those by Newcombe (2002) where the authors showed similar removals through pilot-scale GAC filters located at the Hope Valley and Myponga WFPs in South Australia. No MI B, MCLR or saxitoxins were detected in the influent water of the filters during the study. Only trace levels of geosmin were detected in the influent water during the study except on December 15, 2008, where 12 ng L-1 of geosmin was detected. At this time the GAC filter removed 58% of the geosmin (effluent of 5 ng L- 1) while the OM filters removed 25% (efflu ent of 9 ng L-1) .

Laboratory-scale GAC column trials Removal of DOC and UV254

Five separate trials were cond ucted between April 2008 and April 2009 where GAC was sampled from the fu ll-scale filter at lllawarra and tested in the laboratory-scale col umns for the removal of cyanobacterial metabolites, DOC and UV254 • Figure 2 shows the removal of DOC and UV254 during the five trials. The removal trends of DOC and UV254 in the laboratory trials reflected that of the full-scale GAC filter (Figure 2) where higher removals were initially apparent followed by a gradual decrease and plateauing of the removal efficiency. Similar removals of DOC were observed between the laboratory colum n and the fu ll-scale filter; however, lower removals of UV254 were apparent in the laboratory columns compared with the fullscale filter (approximately 10% lower). It is unclear why this is the case, although it may be possible that the higher removal in the full-scale filters was due to a greater level of biological degradation of the UV absorbing components of NOM. Thomson et al. (2004) showed a white rot fungus was able to remove high molecular weight chromophores and attributed this to a depolymerisation reaction where enzymes such as peroxidises are able to oxidise specific activated aromatics. During the transport of the GAC samples to the AWQC , some

Figure 1. Per cent removal of dissolved organic carbon (DOC) and UV absorbance at 254 nm (UV254 ) through the full-scale GAC and dual media filters at lllawarra. 100

- • - DOC - • - UV254

90 80 70 "O

Q)

> 0 E Q)

a:

C: Q)

60 50 40

~

Q)

[J_

30 20

, , 1\

0

Trial 1

Trial 2

Apr 2008

Aug 2008

1Jt--·

.,A•

. !"····=·-·

10

Trial 3

Trial 4

Trial 5

Oct 2008

Jan 2009

Apr 2009

Figure 2. Removal of dissolved organic carbon (DOC) and UV absorbance at 254 nm (UV254) during the five individual laboratory-scale GAC column trials. 50

- • - Influent - • - Effluent 40

47% ~

.

,•I\•

' :•, .

0)

.s ~

46%

. . \ . . . . . ../. .. I •• . ..... rJ

30

-'....J CD

47%

:' :

20

\

10

I

:

:''

I

I

• /••:'•

\

e, I

0

Trial 1

Trial 2

Apr 2008

Aug 2008

I

• •

Trial 3

••••••\

Trial 4

Oct 2008 Jan 2009

Trial 5 Apr 2009

Figure 3. Removal of 2-methylisoborneol (MIB) during the five individual laboratory-scale GAC column trials. The average percentage removals during each trial are listed.

water

MAY 2010 67


GJ

water treatment of the organisms responsible for biological degradation of UV absorbing chromophores may have been compromised.

laboratory GAC filters, highlighting the applicability of the laboratoryscale col umn to simulate full-scale GAC filters. Ho (2004) also previously demonstrated that the laboratory column was able to successfully simulate a pilot-scale GAC filter for the removal of both MIB and geosmin.

Table 3. Comparison of removals by GAC during trial 1 using EBCTs of 6 and 15 mins. Average % removal by GAC during trial 1 EBCT 6 min EBCT 15 min

Parameter

Removal of MIB and geosmin

DOC 38 Although the GAC appeared to have 30 UV2s4 been rapidly exhausted for NOM MIB removal, based on the DOC and Geosrnin 77 UV254 results, removal of MIB was still evident (Figure 3). albeit at lower concentrations, that is, ng L-1 compared with mg L-1 . It is likely that DOC removal in the ng L1 (and even µg L- 1) range was occurring, although this would not have been detected due to the limit of detection of the DOC analyser (0.1 mg L-1) . An anomaly occurred in trial 1 where no MIB was detected in the influent or effluent of the GAC column . This was rectified in trial 2 where an average of 70% MIB removal was observed. The average removals in the subsequent trials (3-5) remained steady at approximately 47%, implying that the main mechanism of MI B removal was adsorption, since removal had decreased from trial 2. The low removals of M IB from trials 3-5 may be attributed to the low EBCT employed (6 min) which was used to simulate the EBCT in the full-scale filters. Previous studies have shown that high levels of MIB adsorption could occur at higher EBCTs, in general between 10-20 mins (Newcombe et al. , 1996; Ho, 2004). Figure 4 shows the removal of geosmin through each of the trials. Greater geosmin removals were observed com pared with MIB, which is consistent with previous studies (Newcombe and Cook, 2002; Ho, 2004). This has been attributed to the structures of the respective molecules where geosmin exhibits a flatter, more planar structure than MIB, wh ich would be conducive for favourable adsorpt ion. The removals of geosmin were reasonably consistent through the trials averaging approximately 69% over the duration of the study. As mentioned previously, during the full-scale monitoring geosmin was detected at 12 ng L-1 in the raw water in December 2008. A similar geosmin influent level was applied to the laboratory column during trial 4 in January 2009. Similar percent removals were observed between the full-scale and

refereed paper

63 57 86 86

A sample of the GAC in trial 1 was also evaluated at an EBCT of 15 min. The removals of DOC, UV254 , MIB and geosmin were compared with the experiment cond ucted at an EBCT of 6 min (Table 3). The resu lts clearly show the advantage of operating at a higher EBCT, with greater removals of t he measured parameters. This supports the previous contention that greater removals of cyanobacterial metabolites could be achieved with higher EBCTs. Removal of microcystin

MCLR concentrations in the influent and effluent of the laboratory GAC col umn during the trials are shown in Figure 5. Removals of MCLR were reasonably consistent during the first four trials, but decreased in trial 5 where greater breakthrough of MCLR was observed. Some form of MCLR degradation occurred in the influent water during trials 3 and 4 which has also been reported in previous stud ies and attributed to biological degradation (Ho et al., 2006). This translated to enhanced removals of MCLR during these two trials where MCLR levels in the effluent were below t he limit of detection towards the latter stages of both trials. However, this enhanced removal of MCLR was not evident in trial 5 where greater MCLR breakthrough was observed, implying a loss in adsorption capacity and/or biological degradation during this trial. These results suggest that the removal of MCLR was predominantly through adsorption. Previous studies by our research group have shown that if biological degradation processes were occurring, efficient removal of MCLR would have been observed (Ho and Newcombe, 2007; Wang et al., 2007). Moreover, Wang et al. (2007) discriminated the adsorption and biodegradation processes during GAC filtration of microcystin toxins and showed that complete removal of 10

30

25

.,~ 20 _J

Cl

.s c: .E

15

II)

0

77%

67%

.i\i·. \ •j\-. •

~ 10

5

70%

••••••

I•

9

. . ' ...

Apr2008

Trial 2

••••••

Trial 3

Aug 2008 Oct 2008

/\•

1• •

•••

Trial 4

Trial 5

Jan 2009

Apr 2009

Figure 4. Removal of geosmin during the five individual laboratoryscale GAC column trials. The average percentage removals during each trial are listed. The percentage removal shown in parenthesis in trial 4 is the removal observed in the full-scale GAC filter.

68 MAY 201o w ater

47%

68%

~'....J Cl

6

2:

5

(.)

4

er: _J

::E

3

2

0

Trial 1

51%

70%

29%

53%

7

i\ .. . • •• j \ • ._ I •

- • - Influent - • - Effluent

8

••• •j \ • •

/ •

- • - Influent - • - Effluent

61% (58%)

:.

...

• • •• •• \ I •• •• ••• • t· I 'e \ • , . ' •••• ,., I •• • • \_;\ \\ ••• ••• • I.. I •• -.• •

.

. ..

\

\

I

••

\

\

.. •....

Trial 1

Trial 2

Trial 3

Trial 4

Trial 5

Apr 2008

Aug 2008

Oct 2008

Jan 2009

Apr2009

Figure 5. Removal of microcystin-LR (MCLR) during the five individual laboratory-scale GAC column trials. The average percentage removals during each trial are listed.

technical features


~ refereed paper

MCLR was evident after 6 months operation of a GAC filter and that the adsorption process accounted for 70% of this removal. If adsorpti on was the only mechanism of microcystin removal then the GAG would have failed t o adequately remove MCLR to below the Australian Drinking Water Guideline (ADWG) level of 1.3 µg L·1 at an average inlet concentration of approximately 5 µg L-1 . This illustrates the importance of the biological degradation mechanism withi n GAG fi lters for the ability t o remove MCLR.

water treatment 20 18

59%

60%

47%

- • - Influent - • - Effluent

-:::;- 16

114 48%

Removal of saxitoxin

Figure 6 shows the removal of saxitoxin toxicity, as STX-eq, by the GACs in each of the trials. Similar STX-eq removals were observed in trials 1-3 where approximately 60% of the toxicity was removed. The removals decreased to below 50% in trials 4 and 5, implying a loss in adsorption capacity for the saxitoxins. This strongly suggests that the removal of saxitoxins through the GAC column was predominantly through adsorption processes. The literature is sparse with respect to any biological removal of saxitoxins, although Kayal et al. (2008) documented biotransformations of saxitoxi ns through anthracite filters, where an increase in toxicity was evident through the filters. This was attributed to the less toxic Ctoxins being biologically transformed into the more potent GTX variants. Evidence for the biotransformations of saxitoxins t o more toxic variants has been documented in marine studies (Kotaki et al. , 1985; Bricelj et al., 1991; Cembella et al., 1993). Newcombe (2002) conducted similar laboratory column trials operating at an EBCT of 15 min and showed that the GAC column was able to remove : 100% of STX-eq on commissioni ng (using vi rgin GAC); approximately 95% of STXeq after 1 month operation; and approximately 70% of STX-eq after 5 months operation. These removals are greater than what was observed in our study and may be attributed to a number of factors including the lower EBCT used in our study (6 min). It is intuitive to expect greater removals with longer contact times, and this was the case with the results shown in Table 3 for geosmin, DOC and UV254 . Other factors which may have contributed to the difference in removals between our study and that of Newcombe (2002) include the type of GAG used and water qual ity, as studies have shown that these factors can influence GAG adsorption capacity (Lambert et al., 1996; Newcombe et al., 1996; Elhadi et al., 2006).

Trial 1

Trial 2

Trial 3

Apr 2008 Aug 2008 Oct 2008

Trial 4

Trial 5

Jan 2009

Apr 2009

Figure 6. Removal of saxitoxin toxicity (as STX-eq) during the five individual laboratory-scale GAG column trials. The average percentage removals during each trial are listed. in particular, MCLR and saxitoxins. A higher EBCT may be required to reduce effluent cyanotoxin levels t o the desired concentrations. The resu lts obtained at the higher EBCT of 15 mins support this co ntention, albeit with different parameters being measured (DOC, UV absorbance, MIB and geosmin).

Summary and Conclusions Results from the fu ll-scale monitoring indicated that the GAG filter was initially advantageous for the removal of a range of water quality parameters; however, as the GAC filter became saturated with organics (after approximately 3 months), the removals through this filter became equivalent to that of the OM filters. The laboratory-scale GAC col umn was shown to successful ly mimic the full-scale GAC filter for the removal of DOC and geosmin where comparisons cou ld be made. The only anomaly was the removal of UV254 where slightly higher removals

There were instances where the effluent concentration of STX-eq exceeded the ADWG provisional health alert level of 3 µg L· 1 , indicating that the GAG was ineffective in efficiently removing saxitoxins under the prescribed conditions. This is in accordance with the MCLR results. Consequently, the results suggest that the GACs in the full-scale filters may not be able to adequately cope w ith relatively high concentrations of cyanotoxins,

water

MAY 2010 69


water treatment (approximat ely 10%) were observed for the fu ll-scale filter. Th is was presumed to be due to greater biological removal of the UV absorbing components of NOM within the full-scale filter. The laboratory GAC columns were more effective in removi ng geosmin than MIB which is consistent with previous studies and attributed to the structural differences of both compounds. Removals of both compounds were predominantly attributed to adsorption, rat her than biodegradation. An average of approximately 70 and 50% removal of geosmin and MIB, respectively, was apparent in the last four trials. In some instances, the levels of both compounds in the effluents of the GAC columns were above 10 ng L-1 , which may be detected by co nsumers. Trial 1 was also conducted at an EBCT of 15 min which resulted in greater removal of geosmin compared with an EBCT of 6 min. Inadequate removals of the cyanot oxins, MCLR and saxitoxins were apparent throug h the laboratory GAC columns implying that the full-scale filters may not be able to cope with these toxins at relatively high concentrat ions. This was thought to be due to the low EBCT employed (6 min) as previous studies have shown greater removals at higher EBCTs. Other factors such as the GAC type, NOM concentration and character would also influence the removals of the cyanotoxins.

Acknowledgments The authors acknowledge Nawal Kawai and Najwa Slyman of the AWQC for their assistance in operating the laboratory GAC col umns.

The Authors

0

refereed paper

Ho L. (2004) The removal of cyanobacterial metabolites from drinking water using ozone and granular activated carbon. PhD Dissertation, University of South Australia, Adelaide, Australia. Ho L., Meyn T. , Keegan A. , Hoefel D., Brookes J., Saint C.P. and Newcombe G. (2006) Bacterial degradation of microcystin toxins within a biologically active sand filter. Water Research 4 0, 768-77 4. Ho L. and Newcombe G. (2005) Effect of NOM, turbidity and floe size on the PAC adsorption of MIB during alum coagulation. Water Research 39, 3668-3674. Ho L. and Newcombe G. (2007) Evaluating the adsorption of microcystin toxins using granular activated carbon (GAC). Journal of Water Supply: Research & Technology - Aqua 56, 281-291. Ho L. and Newcombe G. Granular activated carbon adsorption of 2methylisoborneol (MIB): Pilot- and laboratory-scale evaluations. Journal of Environmental Engineering (in press, doi: 10.1061 /(ASCE)EE.19437870.0000231 ). Kayal N. , Newcombe G. and Ho L. (2008). Investigating the fate of saxitoxins in biologically active water treatment plant filters. Environmental Toxicology 23, 751-755. Kotaki Y, Oshima Y, Yasumoto T. (1985) Bacterial transformation of paralytic shellfish toxins. In Anderson, D.M., White, A.W., Baden, D.G. (Eds.), Toxic dinoflagellates, Elsevier Science Publishers, New York, USA, pp. 287292. Lalezary S., Pirbazari M. and McGuire M.J. (1986) Oxidation of five earthymusty taste and odor compounds. Journal of the American Water Works Association 78, 62-69. Lambert T.W., Holmes C.F.B. and Hrudey S.W. (1996) Adsorption of microcystin-LR by activated carbon and removal in full scale water treatment. Water Research 30, 1411 - 1422. Nerenberg R., Rittmann B.E. and Soucie W.J. (2000) Ozone/biofiltration for removing MIB and geosmin. Journal of the American Water Works Association 92, 85-95. Newcombe G. (2002) Removal of algal toxins from drinking water using ozone and GAG. AwwaRF Report 90904, American Water Work Association, Denver, CO, USA. Newcombe G., Collet A., Drikas M. and Roberts B. (1996) Granular activated carbon pilot plant studies. Water: Journal of the Australian Water Association 23, 29-31. Newcombe G. and Cook D. (2002) Influences on t he removal of tastes and odours by PAC. Journal of Water Supply: Research & Technology - Aqua 5 1, 463-474.

Lionel Ho (Senior Research Officer) and Gayle Newcombe (Research Leader, Appl ied Chemistry) are with the Australian Water Quality Centre, South Australian Water Corporation (email: lionel.ho@sawater.com.au). Keith Craig is a Technical Director with Veolia Water Australia.

Heriberto Bustamante is a Project Manager (Science and Technology, Sustainability Division) with Sydney Water Corporation.

References Bricelj V.M. , Lee J.H. and Cembella A.D. (1991) Influence of dinoflagellate cell toxicity on uptake and loss of paralytic shellfish poison toxins in the Northern Quahog Mercenaria mercenaria. Marine Ecology Progress Series 74, 33-46. Cembella A.D., Shumway S.E. and Lewis N. I. (1993) Anatomical distribution and spatio-temporal variation in paralytic shellfish toxin composition in two bivalve species from the Gulf of Maine. Journal of Shellfish Research 12, 389-403. Chin Y.P. , Aiken G. and O'Loughlin E. (1994) Molecular-weight, polydispersity, and spectroscopic properties of aquatic humic substances. Environmental Science & Technology 28, 1853-1858. Cook D., Newcombe G. and Sztajnbok P. (2001) The application of powdered activated carbon for MIB and geosmin removal: Predicting PAC doses in four raw waters. Water Research 35, 1325-1333. Elhadi S.L. N., Huck P.M. and Slawson R.M. (2006) Factors affecting the removal of geosmin and MIB in drinking water biofilters. Journal of the American Water Works Association 98, 108-119. Graham D. and Hayes K.P. (1998). Application of solid phase microextraction for the analysis of off-flavours in water. In Proceedings of the WaterTECH Conference, Brisbane, Australia.

70 MAY 2010 wat er

Newcombe G., Cook D. , Brooke S., Ho L. and Slyman N. (2003) Treatment options for microcystin toxins: Similarities and differences between variants. Environmental Technology 24, 299-308. Nicholson B.C., Rositano J. and Burch, M.D. (1994). Destruction of cyanobacterial peptide hepatotoxins by chlorine and chloramine. Water Research 28 , 1297-1303. Oshima Y. (1995) Postcolumn derivitization liquid chromatographic method for paralytic shellfish toxins. Journal of AOAC International 78, 528-532. Pirbazari M., Ravindran V., Badriyha B.N., Craig S. and McGuire, M.J. (1993) GAC adsorber design protocol for the removal of off-flavors. Water Research 27, 1153-1 166. Rittmann B.E., Gantzer C.J. and Montiel A. (1995) Biological treatment to control taste-and-odor compounds in drinking water treatment. In Advances in taste-and-odor treatment control. Eds. I.H. Suffet, J. Mallevialle and E. Kawczynski , American Water Works Association Research Foundation, Denver, CO, 209-246. Rositano J., Nicholson B.C. , Heresztyn T. and Velzeboer R.M.A. (1998) Characterisation and determination of PSP toxins in neurotoxic cyanobacteria and methods for their removal from water. Urban Water Research Association of Australia, Research Report No. 148, Australia. Thomson J., Parkinson A. and Roddick F. (2004) Depolymerization of chromophoric natural organic matter. Environmental Science & Technology 38, 3360-3369. Velzeboer R.M.A., Baker P.D. , Rositano J. , Heresztyn T. , Codd G.A. and Raggett S.L. (2000) Geographical patterns of occurrence and composition of saxitoxins in the cyanobacterial genus Anabaena (Nostocales, Cyanophyta) in Australia. Phycologia 39, 395-407. Wang H., Ho L., Lewis D.M. , Brookes J.D. and Newcombe G. (2007) Discriminating and assessing adsorption and biodegradation removal mechanisms during granular activated carbon filtration of microcystin toxins. Water Research 41 , 4262-4270. Yagi M., Nakashima S. and Muramoto S. (1988) Biological degradation of musty odor compounds, 2-methylisoborneol and geosmin, in a bioactivated carbon filter. Water Science & Technology 20, 255-260.

technical features


GJ

wastewater treatment

refereed paper

COMPARISON OF TERTIARY FILTERS OPERATIONAL PERFORMANCE MP Thomas Abstract The upgraded Maroochydore STP was commissioned in February 2007 to produce Class A recycled water. Cloth fi lters consisting of a woven polyester fabric with 10Âľm pore size provide tertiary filtration at th is plant. Th is paper reviews the results from 3 years of operation of these filters and compares it with the performances of a conventional sand filter at Noosa STP, and a continuous backwash sand filter followed by an ozone-SAC f ilter at Landsborough STP. The conventional sand filters and the ozone-SAC filter produced recycled water that complied with the turbidity requi rements of the Queensland Water Recycling Guidelines. The cloth filters at Maroochydore were adversely affected by fouling, wh ich required the development of a chemical cleaning regime. The most recent resu lts from Maroochydore comply with the maximum Class A Recycled Water turbidity requirement, and the rol ling annual 95% ile turbidity has been improved t o 2.2 NTU, only slightly above the Guideline value. Operation and maintenance costs are compared for the different types of filters.

Keywords Cloth filter; Fouling; Operation; Recycled water; Sand filter; Te rtiary filtration; Turbid ity.

Introduction Tertiary filtration is an essential process element for achieving high quality recycled water. This paper compares the performance of fou r different types of tertiary filters in operation in the Sunshine Coast region. The overall operating performance of tertiary filters is determined by: (1) effluent quality, (2) hydraulic capacity, and (3) operation and maintenance (O&M) costs. These three aspects are inter-related, particularly in the case of cloth fi lters. For example, when cloth fi lter fouling occurs that cannot be removed by the normal backwash sprays then fi lt er headloss increases and ultimat ely filter hydraulic capacity is reduced. The increase in filter headloss causes partial bypass of

Table 1. Details of tertiary filters at Sunshine Coast Water STPs. Location of STP

Maroochydore

Noosa

Landsborough

Woven polyester fabric, 10Âľ m pore size

Mono-media sand, 1.7 mm ES

Mono-media sand, 0.9 mm ES followed by ozone-BAG, 0.85-1.1 mm ES

2

4

3 sand filters + 1 BAG filter

420 filter segments x0.16m 2 x 50% submergence i.e. wetted area = 33.6 m2

36 m2 X 1.8 m depth

Sand filter = 4.9 m2 x 2.3 m depth; BAG filter = 3.8 m2 x 2.4 m depth

Hydraulic loading rate @ average flow

14. 7 m3/m2/h

3.2 m3/m2/h

Hydraulic loading rate @ peak flow

15.6 m3/m 2/h

17.5 m3/m2/h

4.2 m3/m2/h (sand filters) 16.1 m3/m2/h (BAG filter) 4.2 m3/m 2/h (sand filters) 16.1 m3/m 2/h (BAG filter)

2.6 NTU

1.2 NTU

Filter media

Number of filter units Filter dimensions (per filter)

Feed water turbidity (50%-ile)

clarifier effluent around the filters. Consequently, final effluent quality is adversely affected by the bypass around the fi lt ers. Filters are essentially a physical process, consequently it is their solids removal capabilities that have the most impact on effluent quality. Turbidity is the relevant water quality criterion in the Queensland Water Recycling Guidelines (Old EPA, 2005). For both "Class A" and "Class A+" the recommended requirements are 95 percentile (95%-ile) turbid ity < 2 NTU and maximum turbidity < 5 NTU. Maroochydore Sewage Treatment Plant (STP) was upgraded in 2007 to achieve high levels of nutrient removal and Class A recycled water (Thomas et al, 2009). The relatively high capital cost of traditional deep-bed sand filters has resulted in water authorities and designers investigating alternative technologies to satisfy recycled water quality requirements. This was the case at Maroochydore STP. The lower capital cost of cloth filters makes them an

The cloth filters' operational performance depended on regular chemical cleaning.

1.8 NTU

attractive alternative, and this benefit apparently derives from their compact process footprint. Consequently woven "cloth" media fi lters were installed at Maroochydore STP. The performance of these fi lters has generated frequent enquiries from within the industry and this paper reviews the results from 3 years of operation of these filters.

Description of Filters Tertiary filters are installed at three Sunshine Coast STPs, at Maroochydore, Noosa and Landsborough. The dimensions and operational loading rates for the filters are detailed in Table 1. Although the hydraulic loading rates vary between the different filters, they were all operating within a comparable range of their design envelopes. The filters at both Maroochydore STP and Landsborough STP were designed to treat average dry weather flow (ADWF) only, however, the Noosa STP filters were designed to treat peak wet weather flows of 5 x ADWF. Furthermore, Landsborough STP was designed as a recyc led water demonstration plant and incorporated 3 parallel sand filters in series with a single biological activat ed carbon (SAC) fi lt er with ozone dosing. The hydraulic loading on the Landsborough filt ers was limited by the downstream capacity of the ozone-SAC filter facility.

water MAY 2010

11


wastewater treatment

~ refereed paper

Table 2. Effluent quality from tertiary filters at Sunshine Coast Water STPs. Feed Water Turbidity (NTU) Type of Filter

50%-ile

Woven polyester cloth media filter Deep-bed sand filter Continuous backwash sand filter BAC filter

2.6 1.2

Filter Effluent Turbidity (NTU)

95%-ile

Max

50%-ile

6.2

15.8

(Limited Data) (Limited Data)

95%-ile

Max

1.3

3.5

10.6

0.5

0.9

1.5

1.8

2.8

7.0

1.6

2.2

9.2

1.6

2.2

9.2

0.7

1.8

4.9

A further difference between the t wo sand filter installations was that the Noosa filters are backwashed interm ittently, after approximately 48 hours of run-time, whereas the Landsborough sand filters feature continuous backwash.

Samples of filter feed water and filtered effl uent were collected from each STP. Samples were t ested for turbidi ty and all analyses were performed at Sunshine Coast Water laboratories usin g Standard Methods. Sample collection was a mix of 24-hour composites and grab samples. However, the low variabi lity in secondary effluent solids on an hour-tohour basis meant that grab samp les

2.00 ~ - - - - - - - - - - - - - - - - - - - - - - - - - - - ~ 1.80

5 ....

• •• • •

..

1.20

~

1:- 1.00

:2 ~ ....

0.60 0.40 0.20 0.00 ~ - - - - - - - - - - - - - - - - - - - - - - - - - - - ~

Figure 1. Deep-bed sand filters effluent quality. Rolling annual 95%-ile turbidity< 1 NTU. 10

8

5 ....

-e ....:,

• • • •• ••

5

4 3

, 2

r--.,~.

Regular chemical clearwng

6

~

~ :2

Grab samples only ,n 2008

7

• . ~· .... , ..' ..

. ....-:... .... ·-.·~ ...,,... ... . . t

• • 411!: • • • •

~ •

Results Effluent turbidity

The effluent quality from the variety of filters types at Sunshine Coast STPs is summarised in Table 2 and also shown graphically in Figure 1, Figure 2 and Figure 3, which show the rolling annual 95 percentile effl uent quality for the periods stated above.

• The traditional deep-bed sand filters produced the best effluent quality. They complied with the recommended turbidity requirements for Class A Recycled Water with 95% -ile turbidity < 2 NTU and maximum < 5 NTU (Qld EPA, 2005).

0.80

9

• Landsborough STP: data period = 3 months; frequency = at least 5 grab samples per week.

The followi ng observations are highlighted from the data presented above:

1.60 1.40

• Maroochydore STP: data period = 3 years; frequency = 1 grab and 1 composite sample per week. • Noosa STP: data period = 5 years; frequency = 1 composite sample per week - effluent only.

Methods

The feed water for all the filters was secondary clarifier effluent from biological nutrient removal activat ed sludge processes, and the typical range of turbidity was from 1.0 to 5 NTU.

were representative of the effluent qualit y under the majority of conditions. The data report ed in this paper was for different periods and at different freq uencies for the three STPs.

..

••

-~

..

••

•+

•A•••• "•.,,

0

Figure 2. Woven cloth media filters effluent quality. In the last 12 months since regular chemical cleaning has been implemented the rolling annual 95%-ile turbidity = 2.2 NTU. 72 MAY 2010 water

• The conti nuous backwash sand filters showed unexpectedly poor turbidity removal and did not comply with the recommended turbidity requirements for Class A Recycled Water. These filters requ ire further investigation to determine the cause of the poor performance. • Despite the unexpected poor performance of the continuous backwash sand filters, the filtered effluent from the downstream ozoneBAG filter com plied with the recommended 95% -ile and maximum turbidity requirements for Class A Recycled Water. The potential implication is that the ozone-SAC filter may have been used alone to achieve the req uired recycled water quality. A similar concl usion was suggested by McAuliffe et al (2009), although this concept would req uire pi lot testing to confirm its feasibility. • A further benefit of the ozone-SAC process is its ability to provide other treatment functions such as dissolved organics removal and better

technical features


~ refereed paper

wastewater treatment

disinfection t han filtration alone (Reungoat et al, 2010).

5.0 4.5

• The filtrate from the cloth filters did not comply with the recommended 95%-ile or maximum turbidity requirements for Class A Recycled Water when considering the complete 3 year dataset. However, the implementation of regular chemical cleaning of the cloth filters has improved their performance. The most recent annual results comply with the maximum Class A Recycled Water turbidity requirement , and the rolling annual 95%-ile turbidity has been reduced to 2.2 NTU, only slightly above the Guideline value. Whether further improvements can be obtained to comply with the 95% -i le Guideline val ue wi ll require on-going monitoring. • Based on the 50 %-ile t urbidity data, the conventional sand filters, cloth filters and BAC filter all had similar turbid ity reduction of approximately 50-60%. However, the Recycled Water Guidelines for turbidity are based on specific values rather than percent removal. Although this observation is not relevant to achieving compliance with the Guidelines, it is useful for understanding these filters' behaviour.

4.0 3.5

...

3.0

~

2.5

5' ~

;'!2

...~ 2.0 1.5 1.0 0.5 0.0 O>

9

"':, ~

0

O>

O>

9

0,

9

"' ~

.,

0,

i.,

9

~

~

9

~

~

0

0,

.,a.

0,

i.,

"' ~

"I

"I

0

Cloth filter performance with high levels of feed water turbidity was monitored during a peak wet weather flow event on 13 and 14 April 2009. Over a period of 12 hours the clarifier effluent turbidity increased from 4 to 16 NTU. Over the same time the cloth filter effluent turbidity increased from 1 to 6 NTU. These resu lts demonstrate that

t 'i'

t 'i'

...

~

N

O>

~

"'

BA C Filter Effluent

Figure 3. Effluent quality from continuous backwash sand filters and BAC filter operating in series. Although the data-set is <1 2 months, the rolling 95%-ile turbidity = 2.2 NTU for the sand filters and 1.8 NTU for the BAC filter. 200

2.0

I

180

I

160

,A

-

140

e 120

.s ~

80

-

I

/

-

V

40

,/

20

0

50

100

/j

t

.,./

:;; ~

-

/

~ 100

60

Operating results for the cloth fi lt ers showed that effluent turbidity was not affected by increasing flows, but it was affected by increasing feed water solids conc entrations. Similar results were reported by McAuliffe et al (2009). The first test of the cloth filters involved incrementally increasing the flow and holding it steady for an hour. Then the feed water, filtered wat er turbidity and headloss across each filter were recorded. The resu lts in Figure 4 show that the effluent turbidity remai ned constant at 0.85 NTU with increasing flows and consistent feed water solids levels (average feed water turbidity = 2.4 NTU). The backwash sprays operated continuously so there was no reason for the headloss to increase due to solids accumulation on the cloth media during this test , and the sole cause of increased headloss was related t o increased flows.

~

~

O>

0,

9

"I

- - s and Filter Effluent

-+- Clarifier Effluent

0

Cloth filter performance with high flows and high feed water turbidity

~

.,a.

,-----%

150

- I

-I

1.8 1.6 1.4

...5'z

1.2 ~ 1.0 0.8

;'!2

-€

...

:,

~

= ii: 0 .6 w 0.4 0 .2

200

250

300

350

40 0

0 .0 450

Flow (Us) -

Filter 1 Headloss

-a- Filter 2 Headless

Effluent Turbidity

Figure 4. Effect of increasing flow on the performance of the cloth fi lters. The critical headloss limit is 300 mm. effluent quality is dependent on the feed water quality. This would imply that conservative secondary clarifier design is req uired to ensure good feed water quality.

Discussion of O&M Aspects Cloth filters The main challenge associated with the operation of the cloth filters at Maroochydore STP was to develop effective methods for controlling and cleaning fouling compounds on the filter material to ensure both hydraulic capacity and effluent quality were maint ained. Filter fouli ng was due to a combination of iron (Fe) scaling and biofilm growth. The formation of the iron scale was

interesting because the apparent mechanism of scale formation wou ld appear to be relevant to many STPs despite the low levels of iron present. Maroochydore STP clarifier effluent averaged Fe= 0.09 mg/L, which is not particularly high, and the fi lter effluent averaged Fe= 0.04 mg/ L. Therefore, the removal of iron across the filters was 0.05 mg/ L. This may seem like an insignificant amount but with a typical daily flow through the cloth fi lters of 20 MUd, this corresponded to a scale formation rate of 1.0 kg Feld, which is equivalent to 1.9 kg/ d of Fe(OHb scale. Scale accumulation typically caused excessive headloss after a period of 2 months, which indicated that the scale accumulation was 0.8 kg/m 2 of filter segment area.

water MAY 2010 73


wastewater treatment The mechanism for iron scaling appeared to be related to chang ing levels of dissolved oxygen (DO) and pH, and their effect on the oxidation state of iron. The bioreactors were operated with low DO levels, in the range of 0.5 to 1.0 mg/ L to optimise biochemical nitrogen removal to achieve effluent total N = 2 mg/ L (Thomas et al, 2009). Alum dosing was used to achieve effluent t otal P = 0 .2 mg/ L (Thomas et al, 2009). Due to the low alkalinity, the alum dosing resulted in a lowering of the pH to 6.5 to 6.6. Under these conditions, the pH-redox (Pourbaix) diagram for iron indicates that the predominant form of iron would be as sol uble ferrous (Fe2+) (Stumm & Morgan, 1970). The soluble ferrous iron remained in the effluent and when it flowed to the cloth filters the backwash spray nozzles caused an increase in DO to 5 mg/L and an increase in pH to 7.2. The DO increased as water droplet s were sprayed th rough the air, and in parallel, carbon dioxide stripping caused the pH to increase. These changes wou ld cause oxidation of the ferrous (Fe2+) to ferric (Fe3+). Ferric is significantly less soluble than ferrous and so it wi ll precipitate and produce a scale on all areas exposed to the backwash sprays (see Figure 5 and Figure 6). Analysis of the iron species has not been undertaken to confirm this proposed mechanism, however the observations closely match the theory. The first strategy for control ling the iron scaling was related to the fundamental redox chemistry of iron in water. The objective was to oxidise the ferrous to ferric in the bioreactors so that the ferric would precipitate and adsorb to the activated sludge floes and not flow

Figure 5. Cloth filter with cover removed showing the extent of the rust-coloured iron scaling. through to the filters. Increasing the DO, or the pH , or both was trialled. The DO levels were increased slightly to the range of 1.0 t o 1.5 mg/L, and lime dosing was re-commenced with the aim of increasing the pH to 6.8. However, this strategy was unsuccessful due to lack of reliability in achieving continuity of lime dosing due to blockage in the lime powder dosing syst em. In addition , it was known that further increases in the DO levels higher than 1.5 mg/L would increase aeration power costs and compromise denitrification. The second st rat egy to control filter fouling was to use chemical cleaning. Initially all 840 filter segments were removed from the filter and manually scrubbed in a 10% hydrochloric acid solution. Although th is was successful for

Figure 6. Cloth filter segments removed from the filter showing a comparison of a new segment (left) and a segment fouled with iron (right). 74 MAY 2010

wat er

~ refereed paper

removing the majority of t he fouling it was clearly going t o be too labour intensive and potentially dangerous for on-going use. A modified approach was implemented that involved stopping the filter rotation and spraying the 10% acid solution onto the cloth fi lter segments in-situ. Then after several minutes of "soaking" time, the fi lter was rotated through 90° and the procedure repeat ed. This procedure remains in use at present, although further automation is being considered. Acid cleaning is required approximately every 2 months. Wh ilst manually cleaning the cloth filter segments the operators also observed that a "slimy" biofilm was causing fouling on the "inside" surface (feed water side) of the filter segments, whereas the iron fouling was predominantly on the "outside" surface (effluent side) of the filter segments. The biofilm formation appeared to be related to the use of methanol for supplementing nitrogen removal. Methanol over-dosing is a known risk associated with achieving effluent total N levels of 2 mg/L. The normal operating performance of the bioreactors is characterised by effluent nitrate plus nitrite (NOx-N) concentrations of > 0.5 mg/ L and BOD5 < 3 mg/ L. On several occasions it was noted that Bioreactor A effluent NOx-N was< 0.2 mg/L and this coincided with effl uent BODs > 5 mg/L (Figure 7 and Table 3). This indicates that methanol was overdosed and that the excess methanol was not oxidised in the final Re-Aeration Zone of the bioreactors. Methanol breakthrough cou ld subseq uently stimulate rapid biofilm growth downstream of the secondary process. Improved control of the methanol dosing

technical features


~ refereed paper

wastewater treatment

was achieved using a feed-forward control strategy linked to on-line nitrate monitors, coupled with vigilant review of routine effluent testing results. However, even with improved methanol dosing control, biofilm growth was still observed on the filters, albeit at a slower rate . Removal and control of biofilms on the filters was achieved using 13% sodium hypochlorite solution. The sodium hypochlorite was intermittently added to the inlet chan nel of the filters and to t he suction zone of the backwash spray pumps. Hypochlorite cleaning is required approximately every month. The overall cost for cloth filter cleaning was $400 p.a. This was based on usage of 960 Uyr of 13% sodium hypochlorite plus 48 Uyr of 32% hydrochloric acid. In addition to the man-hours required for these cleaning tasks, there are risks associated with manual cleaning using t hese chemicals. However, the current cleaning regime is effective for maintaining hydraulic capacity. It's important to not e that the fouling causes a gradual increase in headless (at equivalent flows) with time. The headless curve in Figure 4 was recorded 2 weeks after a hypochlorite clean and 2 months after an acid clean i.e. the filter was partial ly fouled at t he time of t he test but showed acceptable headless. The major consequence of f ilter foul ing was that it increased filter headless, which consequently limited filter hydraul ic throughput. One of the characteristics of Maroochydore STP is that the hydraulic grade-line from t he secondary clarifiers through t he tertiary treatment process is very f lat. If filter headless increases to 300 mm then a bypass weir in the filter inlet channel is

Table 3. Frequency of methanol break-through for Maroochydore Bioreactor A. Description of Operating Conditions

Characteristic Effluent BOO5

"' .ยง.

"'C 0

a,

Normal operation

::: 3 mg/L

2: 0.4 mg/L

58%

::: 3 mg/L

15%

Methanol break-through

> 3 mg/L > 3 mg/L

< 0.4 mg/L < 0.4 mg/L 2: 0.4 mg/L

17%

Poor performance

over-topped, directing unfiltered effluent to the disinfection process. Under high headless conditions the risk of physically damaging the filter segments increases significantly and the additional cost of torn filter segments also needs to be considered. The cloth filter segment panels were designed to withstand a maximum differential hydraulic level of 400 mm. However, it appears that when the headless exceeds 300 mm t hen t he risk of teari ng a filter panel increases. Prior to the filter fouling being effectively controlled by chemical clean ing, filter headless of 250 to 300 mm was common at flows of 300 Us. Consequently, the design hydraulic capacity of 420 Us could not be attained. Over the 3 year period since fi lter commissioning, the average number of filter panels t hat have needed t o be replaced because t hey were torn was approximately 20 per annum. This corresponds to an annual replacement rate of the filter segments of 2%, at a cost of $4,000 p.a. Furthermore, the replacement cost for a full set of filter panels at Maroochydore STP will be approximately $200,000, and based on experience to date, the demonstrated service life of the filter panels will be at least 3 years, and may possibly be 5 years.

13

5.5

12

5.0

11

4 .5

10

4 .0

9

3.5

8

3.0

7

2.5

6

2.0

5

1.5

~

.ยง.

4

1.0

3

0.5

2

Frequency of Occurrence

Perfect optimisation

14 . - - - - - - - - - - - -- - - -- - - - - - - - - - - - - 6 .0

~

Characteristic Effluent NO.-N

'i

5 z

9%

Sand filters Effluent quality from the deep-bed sand fi lters at Noosa STP has been extremely consistent. During dry weather flow conditions the filter run-times are typically 48 hours. However, during peak wet weather flow conditions (5 x ADWF) run-times are reduced to between 6 and 12 hours due to rapid increases in headless. There are two major reasons for this high headless. Firstly, strands of algae occasionally slough off the secondary clarifier launders and cause blinding of the filter surface. Secondly, high headless develops because backwashing cannot be carried out at the requ ired short int ervals due to limited capacity in the backwash waste water recycle system. Therefore, the overall capacity of the fi lters is limited by the backwash system capacity, which results in red uced filter throughput and bypassing of secondary effluent to the disinfection process. The unexpected poor performance of the continuous backwash sand filters at Landsborough is currently under investigation. The fi nal aspect of sand filters t hat should be considered for completeness is thei r media life. Filters may suffer from some loss or carry-over of sand media during backwashing. Apart from sand loss during backwashing, the service life of filter sand is expected to be 15 to 20 years before requ iring replacement. The traditional deep-bed sand filters have had negligible sand loss over t heir 12 years in operation to-date. The costs for media replacement are significant, but they occur at long intervals, and t hey wou ld normally be considered as asset replacement capital costs . Nevertheless, the O&M costs for media replacement were calculated and spread over t he service life for both sand filters.

Operating costs and power consumption

r-- r-- r--

'!., '!a. 'i ., ~ 1~ 0 0 0 -+- 8005

-.- NQx-N

Figure 7. Maroochydore Bioreactor A effluent BOD and nitrate concentrations.

The specific power consumption and direct operati ng costs (OPEX) for each type of filter are presented in Table 4. The power consumption results exclude effluent pumping if this was req uired at

water MAY 201 0 75


wastewater treatment a particular site, and also excludes pumping of the dirty backwash water back to the process. Both these costs are highly dependent on the site specific hydraulic profile and can be significant at certain sit es. However, they are not an inherent characteristic of these filters. The OPEX results are costs directly associat ed with filter operation, and include electricity, cleaning chemicals and filter media replacement. The OPEX excludes overheads and labour costs for O&M tasks. The labour input for O&M tasks was less than 5% of tot al operator man-hours, although the cloth filters did have the highest O&M labour requirements. Automation of the chemical cleaning of the Maroochydore cloth fi lters is being considered to reduce the labour requirements. The deep-bed sand filters clearly had the lowest specific power usage and OPEX. Whereas the ozone-BAC filter had the highest specific power usage and OP EX, although its higher costs can be attribut ed to its capability to provide additional dissolved organics removal and better disinfection than other filters (McAuliffe et al, 2009; Reungoat et al, 2010). Electricity costs are the dominant factor for the ozoneBAC fi lter and the continuous backwash sand filter. The service life of the filter media that was adopted in Table 4 was 20 years for the sand filters, 8 years for the activated carbon and 5 years for the cloth filter. Filter media replacement costs are the dominant cost factor for the cloth filters and conventional sand filt ers. The power usage of the Maroochydore cloth f ilters is higher than some other installations because the filter disk drive motors and backwash spray pumps ru n continuously due to the flat hydraulic grade-line. However, in the typical cloth filt er installation these motors would normally operat e intermittently. The range of operating costs for different types of filters highlights the importance of undertaking whole-of-life cost assessments for determining the optimum type of filter in each application. However, such an evaluation would be site specific and this paper focuses on operational aspects rather than capital costs.

Conclusions 1. Deep-bed sand filters prod uced extrem ely consistent high quality effluent that com plied with the recommended turbidity requirements for

76 MAY 201 0 wat er

[iJ

refereed paper

Power OPEX Chemicals OPEX ($/ ML) ($ / ML)

Media OPEX Total OPEX ($ / ML) ($/ ML)

Table 4. Power consumption and OPEX for tertiary filters. Type of Filter

Power Usage (kWh/ ML)

Deep-bed sand filter

0.9

0.13

0.00

1.99

2.12

Woven polyester cloth media filter

19

2.90

0.04

5.06

8.00

65 225

9.80 33.75

0.00 0.00

1.51 2.20

11.31 35.95

Continuous backwash sand filter Ozone-BAG filter

Queensland Class A Recycled Water. Furthermore, the deep-bed sand filters had the lowest specific power consumption and operating costs of the types of filters reviewed. 2. The cloth f ilters operational performance was dependent on regu lar chemical cleaning to control fouling. The most recent annual result s comply with the maximum Class A Recycled Water turbidity requiremen t, and the rolling annual 95%-ile tu rbi dity has been improved to 2.2 NTU, only slightly above the Guideline value. Effluent turb idity was dependent on the feed water quality, and wi th elevated feed water solids levels it did not comply with the recommended turbidity requirements for Queensland Class A Recycled Water. Furthermore, the operational costs were higher than co nventional sand filt ers.

3. The continuous backwash sand filters showed unexpectedly poor effluent turbidity results that did not comply with the recommended turbidity requirements for Class A Recycled Water, and they require further investigation.

4. The ozone-BAC filter complied with the recommended 95%-ile and maximum turbidity requirements for Class A Recycled Water. The potential implication is that the ozone-BAC filter may have been used alone without upstream sand filtration, although pilot t esting would be necessary to confirm this. A further benefit of the ozone-BAG process is its ability to provide other treatment functions such as dissolved organics removal and better disinfection than other filters. However, th is additional treatment capability comes at a cost, and the ozone-BAG filter had the highest specific power consumption and operating costs of the types of fi lters reviewed.

Acknowledgments The author would like to acknowledge the operators at Maroochydore, Noosa

and Landsborough treatment plants for their on-going diligence in the operation of the STPs and for assisting with this investigation ; and to the Sunshine Coast Water laboratory staff for their on-going efforts with the sampling and analysis necessary to generate the data reported in this paper. Finally, the author would like to acknowledge the contribution of Ken Hartley in elucidating the mechanism for iron fouling on the cloth filters at Maroochydore STP.

The Author

Michael Thomas is the Process Engineer for STP operations at Sunshine Coast Water. He has 20 years experience in the field of wastewater treatment operations, including STP operations, process optimisation, techn ical support to operations, operator traini ng and mentoring, and applied research and development. Email: mike.thomas@sunshinecoast.qld.gov.au

References McAuliffe, C., Mieog, J. , Williams, C. & Currie, J. (2009). "Planning for advanced tertiary treatment: Melbourne's ETP", AWA Water Journal, Vol. 36, No. 6, pp. 46-55, Sept 2009. Queensland Environmental Protection Agency (2005). Queensland Water Recycling Guidelines , Queensland Government document, December 2005. Reungoat, J., Macova, M. , Carswell, S., Escher, B.I. , Mueller, J.F., Gernjak, W. & Keller, J. (2010). " Effective removal of pathogens and micropollutants by ozone and GAC", AWA Water Journal, Vol. 37, No. 1, pp. 69-72 , Feb 2010. Stumm , W. & Morgan, J.J. (1970), Aquatic Chemistry, Wiley-lnterscience. Thomas, M., Hartley, K. , Gloag, G., Walpole, R. & Jensen, P. (2009). "Healthy River, Happy People: How Maroochydore STP achieved exceptional nutrient removal N = 2 mg/ L and P = 0.1 mg/ L", AWA Ozwater'09 conference proceedings, paper 276, Melbourne, Vic, 1618 Mar 2009.

technical features


G

re fereed paper

wastewater treatment

SLUDGE SETTLEABILITY IN BNR PROCESSES K J Hartley, P A Lant Abstract This paper develops an empirical correlation between specific sludge volume index (SSVI) and primary anoxic mass fraction in biological nutrient removal (BNR) processes. Two separate studies are described involving eight fullscale BNR plants covering a wide range of process format s. These include nitrification denitrification (ND), enhanced biological phosphorus removal (EBPR) and NDEBPR processes in continuous flow or intermittently decanted formats, with or without primary treatment and prefermentation. None of the plants had chemical dosing to the liquid stream processes. In Study 1 routine operating data from six continuous f low, compartmentalised BNR plants were analysed. In St udy 2 the operat ing regime of two intermittently decanted plants was systematically varied and process behaviour monitored. For all plants SSVI increased with primary anoxic mass fraction in a consistent manner. It was demonstrated in one of the intermittent plants t hat secondary anoxic mass fraction had no effect on SSVI.

Introduction Sludge settleability is one of the most important design and operat ing characterist ics of the activated sludge process. It affects both the sizing of tankage and the process capacity actually achieved. Yet, nearly one hundred years after the activated sludge process was invented (Ardern and Lockett, 1914), positive prediction and control of this parameter remains elusive. Four sl ightly different tests are in use for measurement of sett leability, SVI (sludge volume index, employing simple unstirred cylinder settling, Mahlman, 1934), sSVI (stirred SVI, in which the cyl inder is fitted with a stirrer, APHA et al, 1998), DSVI (diluted SVI, using serial dilutions of t he sludge, Koopman and Cadee, 1983) and SSVl 3 _5 , or simply SSVI (stirred specific volume index, in which the sludge concentration is standardised at 3.5 g/L, White, 1975; White originally called his test the Stirred Specific Volume, SSV, but it is now commonly referred to as SSVI). The values of these settleability

parameters can be related to the settling properties of the sludge, and to each other, through published statistical correlations (Daigger, 1995; Ekama et al, 1997; Ozinsky and Ekama, 1995). In this paper the sSVI and SSVI of a sludge are assumed to be essentially equal. A range of factors affects settleability in act ivated sludge processes, including feed quality, SRT (solids retention t ime), mixing characteristics, oxygen regime, pH and nutrient availability (Jenkins et al, 2004; Henze et al, 2008). For biological nutrient removal (BNR) processes (both nitrification denit rification, ND, and nitrification denitrification excess biological phosphorus removal, NDEBPR), the Water Research Group at the University of Cape Town (UCT} conducted a comprehensive sett leability study which they published in a lengthy series of papers and summarised in Ekama et al (1996), Casey et al (1999) and Tsai et al (2003). They found that t he low F/M group of filaments (such as Microthrix parvicel/a and 0092) cause practically all bulking problems in long SAT nutrient removal processes and that t hey cannot be controlled by anaerobic, anoxic or aerobic selectors. Whether or not the process includes biological phosphorus removal, it is the operating nitrogen regime which governs sludge settleability . According to the model they developed, in a two-stage anoxic-aerobic system utilising mainly slowly biodegradable COD, if denitrification in the primary anoxic zone is incomplete and nitrite is present, t he intracellular intermediate nitric oxide persists and inhibits the denitrifying floe-formers in the aerobic zone. The fi lamentous organ isms, which do not compete well in t he anoxic zone because t hey can on ly denitrify nitrate to nitrite, are not inhibited in the aerobic zone where t hey proliferate. Intracellular nitric oxide does not accumulat e when the COD is readily biodegradable (RBCOD) so a significant fraction of RBCOD relieves t he competitive pressure on the floe-formers.

SSVI increased with primary anoxic mass fraction.

Tsai et al (2003) observed that Microthrix parvicel/a growth could be stimulated or inhibited by varying the concentration of ammonia (the organism 's requi red nitrogen source) to high or low values respect ively. They t herefore postulated that deterioration of settleability at low aerobic fraction could be due to increasing ammonia concentration (decreasing nitrification efficiency). In a study of seven fu ll-scale BNR plants in South Africa, Casey and Alexander (2001) found a significant inverse correlation between average DSVI and process aerobic mass fraction. In these compartmentalised plants the DSVI increased from 61 mUg at an aerobic mass fraction of 76% to 162 mU g at an aerobic fraction of 44%. In a study of full-scale oxidation ditch plants, Hartley (2008) used the effluent ammonia:nitrate ratio as a surrogate for the operating anoxic mass fraction. He found that SSVI was a minimum at an optimum ammonia:nitrate ratio and increased at higher and lower ratios. The optimum ratio varied between plants wit hin the range 0.1-10 and SSVI could be controlled by appropriate aeration control. Despite t hese studies, there is still no reliable method of predicting t he SSVI to be expected with any given BNR process format except the oxidation ditch. While the UCT model provides insight into possible mechanisms affecting settleability, it is not currently useful for design or operational purposes. This paper examines the relationship between anoxic mass fraction and sludge settleability using operating data gathered from full-scale Australian BNR plants in two separate studies. Both continuous flow and intermittently decanted process formats are included.

Methods Study 1 The first study examined settleability data from six Australian contin uous flow, compartmentalised BNR plants, two being separate t rains with slightly different formats at one facility. Data from two of the plants have been published

wat er

MAY 2010 77


~ refereed paper

wastewater treatment previously (Hartley, 1995). Process formats of the six plants included ND, EBPR or NDEBPR, with or without primary treatment and prefermentation. No plant had chemical dosing to it s liquid stream processes.

Phase J1 I

Feed Aerate Settle

Study 2 In the second study, evolutionary operation (EVOP) programs (Box and Draper, 1969) were run at two small intermittently decanted ND plants, Jimboomba and Flagstone (capacities 1500 and 2000 EP respectively}. At each plant the operating regime was systematically varied and the process behaviour monitored . The Jimboomba plant utilised a single tan k, co ntinuous feed, interm ittently decanted process with diffused air aeration. The inlet was located near the surface and during extended air-off periods t here was no positive mixing between influent and settling sludge. During aeration the air flow was constant with no dissolved oxygen (DO) control. The program was run over 15 months split into eight operating phases, J1 -8. In Phases J 1-3 the DO concentration measured manually during aeration periods averaged less than 1 mg/L. Due to a change in the aeration syst em, in Phases J4-8 the average DO increased t o about 3 mg/ L. The initial operating format (Phase J1) is shown in Figure 1. During subsequent phases the air-off fraction of the aerate period was varied. In Phase JS continuous aeration was used in the aerate period. In Phase J6 the air-off (primary anoxic) period was consolidated into a single initial period preceded by fou r minutes of aeration to resuspend the settled sludge. However, significant sludge settlement during the anoxic period prevented mixi ng of influent with sludge. Therefore in Phase J7 short bursts of air (air off 11 minutes, air on 1 minute) were used to mix the reactor during the primary anoxic period. The primary anoxic period was extended by 12 minutes in Phase J8 (shown in Figure 1 ). The Flagstone plant utilised two identical tanks in series with variable water level and a pumped return activat ed sludge (RAS) recycle from the second tank to the first. Sewage was fed contin uously to the first tank (the anoxicaerobic tank or AAT) which was intermittently aerated. Mixed liquor transferred continuously to the second tank via a submerged connection. The second tank (the intermittently operated

78 MAY 2010 water

Decant Waste

Phase JS I

Feed Waste Aerate

I I I I I I

-

Time

Settle Decant Time

Figure 1. Jimboomba - cycle formats. I

Feed

-

Aeratio n AAT Aeration IOC

Settle IOC Decant IOC

Time

Figure 2. Flagstone - generic cycle format.

-

700 , - - - - - -- - - - - - - - - - - - - - - - - - - , - - - - , 600 500 ~

400

...J

.§.

> "'

300 200 100

0L--- - - - - - - - - - - - - - - - - - -- - - - - ~ o 01. Jan 01 -Feb 01·Mar 01 -Apr 01 -May 01 -Jun

01·Jul 01-Aug 01-Sep 01-Oct 01-Nov 01-Dec 01 -Jan 2002

Figure 3. SVI and MLSS trends for Loganholme STP.

clarifier or IOC) was intermittently aerated for a period, then allowed to settle and effluent decanted. The settle and decant period was initiated by reactor water level, not by time. As for Jimboomba, there was no DO control and DO concentrations measured manually during aeration periods in the AAT and IOC varied from 1-6 mg/L. Six operating phases, F1 -6, were run over an eleven month period. Figure 2

shows the generic cycle format used. Various combinations of air-on and air-off were used in the AAT and IOC. During air-off periods the tanks were mixed by periodic bursts of aeration .

Results Study 1 Operating data from the six plants are summarised in Table 1. The plants were notionally operating under long-term

technical features


~ r e fereed paper

wastewater treatment

Table 1. Study 1: Results. Ballarat South

Bendigo

Goodna

Loganholme Stages 5-6

Maroochydore A Plant

Maroochydore B Plant

69,000

120,000

36,500

100,000

50,000

65,000

Yes

No

No

No

Yes

Yes

Yes

No

No

No

Yes

Yes

AO: EBPR partial nitrification

Modified UCT: NDEBPR

MLE: ND

A20: NDEBPR

PROCESS FORMAT Capacity, EP Format PST 1 Prefermenter BNR

West Bank: NDEBPR West Bank: NDEBPR

Mass fractions:2 Pre-anoxic

0

0

0

0

0.09

0.09

Anaerobic

0.20

0.13

0

0.15

0.07

0.07

0

0.36

0.45

0.20

0.22

0.26

Aerobic

0.70

0. 41

0.45

0.55

0.52

0.48

Sec anoxic

0.10

0.1 0

0.10

0.10

0.10

0.10

1993

1993

2001-02

2002

2001 -02

2001-02

3.5

24

13

11

15

17

Mixed liquor (a)

2.7

20

3.0

4.0

20

Anoxic (r)

1.0

Primary anoxic

OPERATING DATA (Medians) Data period SRT3, d Recycle ratios 1.0

RAS (s) 4

0.15

0.8

0.9

0.7

1.0

1.0

MLSS, g/L

2.7

3.8

3.2

4.4

2.9

3.7

NH3-N, mg/L

_5

0.2

0.6

0.5

0.3

1.6

N03¡N, mg/L P0 4-P, mg/L

3.6

4.0

0.7

3.0

5.0

3.2

2.3

0.8

3.7

5. 2

4.8

Parameter

SVI

DSVI

sSVI

SVI

sSVI

sSVI

Value, mUg Equivalent SSVl6,

69

199

143

195

88

93

51

109

143

88

93

1.41

1.16

1.45

77 1.41

1.20

1.18

Effluent quality:

Settleability:

mu g

90:50%ile

1. Primary sedimentation tanks plus anaerobic sludge digestion with lime treatment of return flow 2. Nominal 10% of sludge assumed to be in secondary clarifiers. Maroochydore pre-anoxic zones fed only with RAS. 3. Solids retention time 4. Return activated sludge 5. Not routinely monitored; partial nitrification 6. Equivalent SSVI calculated using the conversions of Ozinsky & Ekama (1995)

steady state c onditions and operating dat a for one or two year periods are listed. As an example, Figure 3 trends the raw data from the Loganholme plant. None of the plants had formal sec ondary anoxic zones ahead of the secondary clarifiers and it was assumed that a nomi nal 10% of the sludge mass resided in the clarifier underflow zones. The two Maroochydore plants had pre-anoxic zones to remove nitrate from t he RAS recycles but during the data period no sewage was fed to those zones. Figure 4 shows the excel lent correlation between median SSVI and primary anoxic fraction for the six plants.

160 y= 51 .11 e 2.24x

140 120

~ 100

...J

.s

80

en

60

> en

40 20

0 0

0.1

0.2 0.3 Primary Anoxic Fraction

0.4

0.5

Figure 4. Study 1: correlation between primary anoxic mass fraction and SSVI for continuous flow plants.

w ater MAY 201o 79


G

wastewater treatment Study 2

refereed paper

(a) Jimboomba

Operating data for the Jimboomba and Flagstone plants are summarised in Tables 2 and 3. The SVI and MLSS data are trended in Figure 5. These plants operated with significant secondary anoxic fractions associated with their settle and decant periods and, in the case of Flagstone, intermittent aeration in the IOC. Effluent phosphate was monitored at Flagstone but no EBPR was observed (Table 3). Figure 6 shows the good relationships between SSVI and primary anoxic fraction for the two plants. In contrast, correlation with total anoxic fraction (not shown) is poor.

360 ~--------~--~--~---------~ s 300 ~

240

4

::;

.ยง.

180

~

3

120

2

"':i;-'

~

60 Phaao

oL=====L--==::::......-L_.:=:.-1-==--L---'==----_JL.:=____J o 01.Jul

01-Sep

01-Nov

01-Jan

01-Mar

01 -Jul

01-May

01-Sep

2008-09

(b) Flagstone

300

~ .ยง. .,>

"'"'-'

200

:i;

s

100

Discussion

O ' -_ _ __.__ _ __.__ _ _ _ _L , , __

Operating and settleability data have been presented from six conti nuous-flow BNR plants notionally operating at steady state (Study 1) and from two intermittently decanted plants whose operating regimes were systematically varied over fifteen and eleven month periods (Study 2). To clarify terminology, in this paper primary anoxic refers to an anoxic zone

"'

01 -Sep

01-Nov

_ .. _ __ _ _. __ _ _

01-Mar 2008-09

01 -Jan

_, O

01-Jul

01 -May

Figure 5. SVI and MLSS trends for Jimboomba and Flagstone STPs.

or period in which the biomass is mixed with raw sewage or sewage mixed with biomass which has passed through an anaerobic zone on ly. Secondary anoxic refers to an anoxic zone or period fed

only with biomass which has already passed through an aerobic zone or period . The Study 1 results provide an excellent correlation between median

Table 2. Study 2: Jimboomba Results. Operating phase

J1

J2

J3

J4

J5

J6

J7

JS

4.00 1.67

4.00 1.67

4.80 1.55

4.80 1.55

4.80 1.55

4.80 1.55

4.80 1.55

4.80 1.55

0.14 0.44 0.42 0.56

0.14 0.44 0.42

0.17 0.51 0.32

0.17 0.51 0.32

0 0.68 0.32

0.19 0.49 0.32

0.23 0.45 0.32

0.56

0.49

0.49

0.32

0.18 0.50 0.32 0.50

0.51

0.55

1/7-14/8 16 4.0

15/8-16/9 24 1.9

17/9-26/11 100 2.8

27/11-15/12 44 2.7

16/12-8/2 91 2.8

9/2-30/3 55 2.4

31 /3-6/8 56 2.6

7/8-30/9 41 2.7

31 0.7

43 0.2

25 1.0

0.5 21

0.6 32

0.9 43

1.5 6.3

1.2 4.3

SVl5, mUg

192

193

221

93

88

91 57

80

85

343 132

219

Equivalent SSVl6, mUg

55

96

244 102

PROCESS FORMAT Cycle time: Total, h Settle & decant, h Mass fractions: Primary anoxic, fx11 Aerobic, f.2 Secondary anoxic, fxi Total anoxic, fxt OPERATING OATA (Medians) Period (2008-09) SRT4, d MLSS, g/L Effluent quality: NH3-N, mg/L N03-N , mg/L Settleability:

1. Primary anoxic mass fraction, fx1 = 1 - fa - fx2 2. Fraction of the cycle time with air on 3. Settle & decant fraction of the cycle time 4. Nominal value calculated from wasting and effluent SS data 5. When new level had been reached

6. Equivalent SSVI calculated using the conversions of Ozinsky & Ekama (1995)

so

MAY 2010 water

technical features


wastewater treatment

~ refereed paper

Table 3. Study 2: Flagstone Results.

F1

F2

F3

F4

F5

F6

Of AAT

0.52

0.56

0.65

0.65

0.83

0.45

Of process (primary anoxic)

0.23

0.25

0.28

0.32

0.39

0.21

0.46

0.46

0.46

0.1 2

0.1 2

0.34

Of process (secondary anoxic) 0.25

0.25

0.26

0.05

0.06

0.18

0.48

0.50

0.54

0.37

0.45

0.39

Operating phase

PROCESS FORMAT Anoxic mass fractions: 1 AAT

IOC Of IOC Process total anoxic OPERATING DATA2 (Medians) 1/9-28/10

29/10-14/12

15/12-22/2

23/2-31 /3

1/4-22/5

23/5-22/7

Cycle time, h SRT3, d

8.8

6.9

8.1

9.2

6.5

7.4

52

47

30

21

25

26

RAS recycle ratio4

3.6

2.9

3.9

12

8.4

7.8

MLSS (AAT), g/L

3.4

3.2

3.0

2.6

2.5

2.3

NHrN, mg/L

0.1

0.1

0.1

0.2

0.3

0.2

N03-N, mg/L

6

13

11

32

9.5

Period (2008-09)

Effluent quality: 28

9.9

8.3

9.2

11

9 8.8

SVl5, mUg

196

215

182

250

266

196

Equivalent SSVl6, mUg

85

93

84

104

108

89

P04 -P, mg/L Settleability (AAT):

1. Biomass-proportioned fraction of time with aerators off in AAT and IOC 2. Manually measured DO concentrations during aeration periods 1-6 mg/L 3. Nominal value calculated from wasting and effluent SS data 4. RAS flow I sewage flow 5. When new level had been reached 6. Equivalent SSVI calculated using the conversions of Ozinsky & Ekama (1995) SSVI and process primary anoxic mass fraction (Figure 4). Variability of t he SSVI at the various plants as expressed by the 90:50%ile ratio ranged from 1.2 to 1.5 (Table 1). Process formats included ND (one plant), EBPR (one plant) and NDEBPR (fou r plants). Three plants wit h EBPR had primary treatment and prefermentation. In these cases the expected higher influent RBCOD levels did not produce lower SSVI (Casey et al, 1999), possibly because most of the RBCOD was utilised in the anaerobic zones ahead of the anoxic zones. Inclusion of an anaerobic zone at the head of an ND f lowsheet decreases the size of the primary anoxic zone required for denitrification thro ugh enhancement of the denitrification rate (Henze et al, 2008). The resu lts of this study suggest that reduction in primary anoxic fraction should also result in a decrease in the process SSVI. The Study 2 intermittent plant results also show consistent relationships between SSVI and primary anoxic fraction , except for Jimboomba Phases

J2 and J6 (highlighted in Figure 6). Phase J2 is the latter part of J 1, redesignated J2 because the SVI suffered a short-term spike for unknown reasons (Figure 5). However, the Phase J6 behaviour is reveal ing. Phase J6 followed JS when the

process operated w ith no primary anoxic period and low SSVI (57 mUg). In Phase J6 a consolidat ed primary anoxic period (mass fraction 0.18) with on ly an initial short mix was introduced. The sludge settled rapidly during this period and

140 120 100 ~

...J

.s

>

80 60

Cl) Cl)

40 â&#x20AC;˘ Jimboomba

20

â&#x20AC;˘ Flagstone

0 0

0.1

0.2

0.3

0.4

Primary Anoxic Fraction

Figure 6. Study 2: correlations between primary anoxic mass fraction and SSVI for the Jimboomba and Flagstone intermittent plants.

water MAY 2010 s1


Q

wastewater treatment mixing with the influent was minimal. The SSVI remained low (55 mUg). When mixing was introduced in Phase J7 effluent nitrate fell significantly and SSVI increased to 96 mUg. In J6 the apparent primary anoxic period was really a continuation of the preceding secondary anoxic settle period. This demonstrates that SSVI was governed by primary, and not secondary, anoxic fraction. It is unlikely that EBPR in the settled sludge played a part as the effluent nitrate in Phase J6 was very high at 43 mg/L.

160 y= 59.33e 1 ' 81 x

140 120 ~

100

...I

E.

80

> en en

60 40

The trends for Jimboomba and Flagstone in Figure 6 are not exactly aligned. One possible explanation for this is the differing effects of DO decay after mixing (by aeration burst s) on the actual anoxic fractions. Casey and Alexander (2001) correlated settleability with aerobic fraction. However, this study shows that settleability correlates better with primary anoxic fraction and is not affected by anaerobic or secondary anoxic fractions. Tsai et al (2003) postulated that poor settleability at low aerobic fraction (high anoxic fraction) cou ld be due to high ammonia concentration stimulating fi lamentous growth. This is not borne out by the data for Jimboomba Phases J3 and J4 which had the same primary anoxic fractions (0.17). When the average aerobic period DO concentration was raised from less than 1 mg/L to 3 mg/L, the effluent ammonia concent ration fel l from 25 to 0.5 mg/L, with an associated nitrate increase from 1 to 21 mg/ L. SSVI remai ned essentially the same (93 and 88 mUg respectively), consistent with the

• Continuous Flow o lntermittent-Jimboomba

20

1::,. Intermittent-Flagstone

0 0

0. 1

0.2 0.3 Primary Anoxic Fraction

show a consistent relationship between SSVI and primary anoxic fraction .

It could be argued that the SSVI remained poor because the driver switched from incomplete nitrification (Tsai et al, 2003) to incomplete denitrification (Casey et al, 1999). However, all the Study 1 plants had relatively low effluent ammonia and nitrate concentrations (Table 1) and the plant with the lowest nitrate concentration (Goodna, 0. 7 mg/L) had the highest SSVI. Overall, the results are consistent with the hypothesis of Casey et al (1999) that sludge settleability is governed by nitrogen removal biochemistry. Whatever the specific mechanisms, the data from both studies

Figure 7 consolidates the results from the two studies. This empirical correlation is useful for both design and operational purposes.

100% Australia owned and manufactured

• • • •

high surface area encourages excellent biological film growth custom modular design for easy maintenance and handling ideal for commercial & domestic waste-water treatment cost effective, efficient, reliable chemically inert used in trickle filters , aeration and degassing systems , submerged filtration systems

Acacia Filtration Systems Pty Ltd Tel: 02 9756 6077 Fax: 02 9756 0173 sales@acaciaproducts.com.au www.acaciaproducts.com.au

82 MAY 201 o water

0 .5

constant primary anoxic fraction rather than the falling ammonia concentration.

The ideal filter medium for wastewater treatment

0.4

Figure 7. Consolidated data from the two studies.

BIO MESH TUBING •

refereed paper

Conclusions Reliable prediction of sludge settleability in the activated sludge process has been an elusive goal since the process was invented in 1914. Based on two studies involving eight full-scale BNR plants, this paper develops a useful empirical correlation between SSVI and process primary anoxic mass fraction. Conclusions are as follows: 1. The plants studied had a wide range of process formats including: ND, EBPR or NDEBPR; continuous flow or intermittent decant; with or without primary treatment and prefermentation. None of the plants had chemical dosing to their liquid stream processes. For all plants SSVI increased with process primary anoxic mass fraction in a consistent manner. 2. It was demonstrated in one of the intermittent plants that secondary anoxic mass fraction had no effect on SSVI. 3. Inclusion of an anaerobic zone at the head of an ND flowsheet decreases the size of the primary anoxic zone required for denitrification through enhancement of the denitrification rate. The

technical features


red uction in primary anoxic fraction should also result in a decrease in the process SSVI.

The Authors

Ken Hartley is an independent consulting engineer and an Adjunct Professor in the School of Chemical Engineering at The University of Queensland. Email: kenhartleyp l@bigpond .com.

Paul Lant is Head of the School of Chemical Engineering at The University of Queensland.

References APHA, AWWA, WEF, (1998). Standard methods for the examination of water and wastewater. 20th ed. Ardern, E., Lockett, W.T. , (1914). Experiments on the oxidation of sewage without t he aid of filters. Society of Chemical Industry Journal 33 (1 0), 523-539. Box, G.E.P. Draper, N.R. , (1969). Evolutionary operation. John Wiley, New York. Casey, T. , Alexander, W.V., (2001). Design and operating strategies to minimize bulking by anoxic-aerobic filamentous organisms in nutrient removal activated sludge plants. WRC Report No. 775/1/01 , Wat er Research Commission, Pretoria. Casey, T.G., Wentzel, M.C., Ekama, GA, (1999). Filamentous organism bulking in nutrient removal activated sludge systems Paper 11: A biochemical/microbiological model for proliferation of anoxicaerobic (AA) filamentous organisms. Water SA 25(4), 443-451. Daigger, G.T., (1995). Development of refined clarifier operat ing diagrams using an updated settling characteristics database. Water Environment Research 67(1), 95-100. Ekama, GA, Barnard, J.L., Gunthert, F.W. , Krebs, P. , McCorquodale, JA, Parker, D.S. and Wahlberg, E.J. (1997) Secondary settling tanks: theory, modelling, design and operation. /WA Science & Technology Report No. 6. Ekama, GA, Wentzel, M.C., Casey, T.G. and Marais, G.v.R. (1996) Filamentous organism bulking in nutrient removal act ivated sludge systems, Paper 6: Review, evaluation and consolidation of results. Water SA, 22(2), 147-152. Hartley KJ (1995) Biological nutrient removal plants: review of full-scale operation. Urban Water Research Association of Australia, Research Report No. 94. Hartley, K.J. (2008) Controlling Sludge Settleability in t he Oxidation Ditch Process. Water Research, 42, 6-7, 1459-1466. Henze M, van Loosdrecht MCM, Ekama GA & Brdjanovic D (2008), Biological Wastewater Treatment: Principles, Modelling and Design. IWA Publishing, London. Jenkins, D., Richard, M.G. and Daigger, G.T. (2004) Manual on the causes and control of activated sludge bulking, foaming, and other solids separation problems. 3rd ed., CRC Press, Boca Raton, and IWA Publishing , London. Koopman, 8. and Cadee, K. (1983) Prediction of thickening capacity using the diluted sludge volume index. Water Research, 17, 427. Mahlman, F.W. (1934) The sludge index. Sewage Works Journal, 6(1), 119-122. Ozinsky, A.E. and Ekama, GA (1995) Secondary settling tank modelling and design, Part 2: Linking sludge settleability measures. Water SA, 21 (4), 333-349 . Tsai, M.-W. , Wentzel, M.C. and Ekama, GA (2003) The effect of residual ammonia concentration under aerobic conditions on the growth of Microthrix parvicella in biological nutrient removal plants. Water Research, 37(12), 3009-3015. White, M.J.D. (1975) Settling of activated sludge. Technical Report TR11, Water Research Centre, UK.

WASTE AND RAW MATERIALS MANAGEMENT

13-17 SEPTEMBER 2010,. MUNICH

-+ -+ -+ -+ -+

Future-oriented innovations, technologies and trends All application fields, all products, all services Tech nology leaders, specialists and newcomers from aro und the world The right solution for every application and every use First-rate related-events program for information and networking

Detailed information and presentation schedule: www.ifat.de

Information: German-Australian Chamber of Industry and Commerce Tel. (03) 9602 2320 gccmel@germany.org.au

ENTSORGA A WORLD OF ENVIRONMENTAL SOLUTIONS

water

MAY 2010 83


wastewater treatment

FAULT PREDICTION FOR WASTEWATER PUMP STATIONS S Beckwith, N Ning, J Adeane This article is a very brief summary, prepared by the Editor, of the paper presented at Ozwater'10 by Beckwith, Ning, Adeane. consistent asset operation, but alarms alone convey little information about cause of degraded asset performance. For example, a build up of "ragging" or foul ing of a pump's impellor would deteriorat e the pump performance and an operational breakdown could happen at any t ime without notification. Abrupt pumping dysfunct ion disrupts the normal operations of a station and precipitates urgent work orders to repair the faulty pumping unit so as to ensure the integrity of the network. Such urgent repairs may be required at times outside normal working hours which incur greater cost versus pre-scheduled corrective maintenance during working hours.

Abstract The project was to derive a signal which wou ld predict pump fai lure and to log it into a monitoring dashboard in the Operations Centre, thereby driving a shift from reactive to proactive control.

Introduction The Water Corporation of Western Australia operates throughout the state, covering not only the City of Perth and over 100 towns, but also outlying communities in remote areas. It operates over 1300 wastewat er pumping stations. All wastewater pump st ations are linked to the operations centre in Leederville. A t ypical wastewater pump station consist s of a wet well, two pumps and level alarms. The duty pump operat es between set levels and if these are exceeded, the st andby pump cuts in. Each station is provided with a SCADA system which relays data on water levels and current draw as shown in Figure 1, 2. However this data does not provide long term performance information. They serve adequately for automatically actuating the stand-by pumping unit to maintain generally contin uous and

The Project

The project was to combine both parameters into a signal wh ich would increase the overall prediction accuracy for individual pumps, and to log it into a monitori ng dashboard in the Operations Centre, thereby driving a shift from reactive to proactive control.

The Predictive System Figure 3 shows the arch itecture of the performance prediction system. The baseline model comprises two sub-modules: wet well level performance baseline and current draw performance baseline, wh ich are based on normal operating performance data obtained during the trai ning stage. Real time raw data obtained from SCADA is then compared with these baseline patterns.

This project was developed to detect early inception of anomalies to predict impending faults, such as shown in Figures 1 and 2.

Current Draw Behaviour Description

Wet well level is a station-level parameter, whereas current draw is based on the pump's operation.

Each time a pump st arts, there is a current draw associated with it. For a pump performing normally, it can be observed that its cu rrent draw for each

Each is a possible indicator for performance of the stat ion, but not an absolute indicator and relying on one of them cou ld result in many false positives.

Using advanced pattern recognition techniques.

Co mpl ete Failure: 19/02/2009 16:30:00 40 00 An

2~ 10 0 14/02/2009 12:00:00 AM

1s.oa9days Stroe•

01/03/2009 12:00:00 AM

FSWWC, f'

r

Figure 1. Current Draw readings that involve an abnormal performance before an incident of failure occurred. It can be observed that the current draw trend starts to deviate from the normal pattern and build up before reaching the complete failure point. Deviatio n Starts: 18/02/2009 23: 11:27

3.6

t---~-

I

-I.~ :I . ;1·~:·.

Complete Failure: 19/02/2009 16:30:00

5:2

14/02/2009 12:00:00 AM

t,01 W

Leve

~=

01/03/2009 12:00:00 AM

WWC P&11. fiea on

Figure 2. Wet Well Level readings in accordance with the time duration in Figure 1.

84 MAY 2010 water

technical features


wastewater treatment burst of start exhibits a consistent pattern. On the other hand, if a pump is experiencing physical det er ioration caused by ragging, for example, it is expected that current draw would start deviating from its normal behavioural pattern and building up for certain amount of time before reachi ng practical complete failure point, as shown in Figure 1. The abnormal functioning capability would also affect the station's standard wet well level empty-fill cyc les as illustrated in Figure 2.

Raw Data (Wet Well Level & Current Draw)

Normal Performance Data

Training

Performance Baseline Model Prediction Model

Wet Well Level Performance Baseline (Station level, frequency domain)

Wet Well Level Performance Pattern Extraction and Prediction

&

&

Current Draw Performance Baseline (Pump level, time domain)

Current Draw Performance Pattern Extraction and Prediction

A signal a day or so in advance would provide opportunity to pre-schedule remedial maintenance to avoid the impending failure.

Dashboard (Alarms)

Combining the Performances The final performance of the station is determined by combining the current draw and the wet well level performance. The combination is done by standardising both performance coordinates by their sample variances and forming a 2D performance feature point. In this way, we can express the pump performance using a single parameter and employ statistical distance to classify the feature points. The complex mathematics of the process are outli ned in the original paper (Ozwater 136). Each day the program automatically calculates the parameter representing the current operation and compares it with the baseline. If the deviation is significant, a red alarm is logged onto the dashboard for that particular station in the Operations Centre, so that relevant action can be taken before complete failure occurs.

The Monitoring Dashboard We have designed and developed an operational dashboard for our operators to monitor the current performance of each pump station. A snapshot of the dashboard is shown in Figure 4. It has two parts. The left part shows the list of pump stations that are monitored and their performance. The red colour denotes a

Figure 3. The Architecture of WWPS Performance Prediction System. warning of impending failure and the green colour denotes normal performance. By clicking the red or green buttons, the detailed cu rrent draw and wet well level performances are shown. The right part of the dashboard shows the graphical plots of the moving average of current draw trends (Unit 1 and Unit 2) and the wet well level performance trends (wet well level performance deviation from baseline) of a selected station (by clicking the station's button on t he left part). The default date range was set to cover the previous 15 days. The date range of the trends can be reset.

These graphs are usef ul for operators to examine the past and current behaviours of pump station as an empirical validation of the prediction results. A major benefit of using the monitoring dashboard is it enables th e operators to focus on those stations with red warnings. This could save significant time and manpower for monitoring our extensive wastewater pump station network as the majority of pump stations should perform normally during normal situations.

C&S BRAND AUSTRALIAN FILTER COAL GRANULAR MEDIA FILTRATION FOR DESALINATION PRETREATMENT Reliable Quality and Service from a Company with more than

l 00 years of expertise in the manufacture of Coal and Carbon Products for Australian Industry

C&S BRAND GRANULAR & POWDERED ACTIVATED CARBONS JAMES CUMMING & SONS PTY LTD 319 Parramatta Rd, AUBURN NSW 21 44 Phone: (02) 9748 2309 Fax: (02) 9648 4887 Email: jamescumming@jamescumming.com.au Website: www.jamescumming.com.au

I Quality

1S09001 . ... °'"'"""-

Lie No. 1628

QUALITY ENDORSED COMPANY AS/NZS ISO 9001 SAi GLOBAL

water

MAY 2010

as


wastewater treatment Testing Results We tested our system on 11 pump stations in a selected suburban d istrict of Perth Region. We used one year (2008) data for train ing the baselines of wet well level trends and current draw trends and their correlated feature points. After that, we used data from 01/01/2009 to 01 / 10/2009 fo r test ing and validation. Different threshold values were explored in order to attain the best prediction results.

1- &-..-&..i&. s.....

5,.;..s..;;;.s.. - .w- - m1

EIHl St Stad.,., Unit 2 CWTNC Dr- Pffftnunn Tren4

We define the terms "Hit " and " False Alarm " in the context of our application as follows: â&#x20AC;˘ Hit (or True Positive) - The predicted failure is an act ual fail ure; â&#x20AC;˘ False Alarm (or False Positive) - The predicted fai lure is not an actual fai lure. Table 1 summarises t hose which could be t ermed successful. For stations that have steady performance trends, the system could reach 100% precision with no false alarms. However, there were some aberrant cases, as shown in Table 2. These varied considerably for different pump units and catchments. For some stations t hat have unstable performance trends, false alarm rates tend t o be high. Taking Essex St U1 as an example, the predictive system fired off 9 times. But when the operators opened up t he station, they found raggi ng 4 t imes, but

Figure 4. Snapshot of Dashboard. one of these failures had not reg istered on the predictive system. This could be caused by situations such as self-derag (i.e. the pu mp builds up rag and then resolves t he rag by itself during running time) or derag during regular maintenance without a work order record. In these cases, t he validation of the actual false alarm rate is complicated as a false positive could be a true positive. A possibly usefu l view of the high false alarm rate is t hat it could

Table 1. Successful operation. Pump Unit

Actual Failure

Hit

False Alarm

Adrian St

U2

Irwin St

U1

Irwin St

U2

Marine Tee

U1

0 0 0 0 0 1 4

0 0 0 0 0 1

0 0 0 0 0 0

Station

Preston pt Rd

U1

Holland St

U2

Marine Tee

U2

Holland St

U1

Samson St

U1

3

2

Table 2. False alarms.

In general, our prediction system can be verified as a useful tool for wastewater pump station performance prediction with high prediction accuracy and acceptable false positive rate.

Future Work Further case studies and testing are needed and planned to explore the Receiver Operating Characteristic curves, investigate the hit rate vs. false alarm rate, and to set the best alarm t hresholds (i.e. deviat ion levels). We will then move into real time operations and measure t he real benefit, and refi ne t he system to accommodate more complex situations.

Conclusions By applying advanced data mining and statistical analysis techniq ues, we are extracting performance information from raw data, enabling us to shift from reactive to proactive maintenance.

The Author

Station

Pump Unit

Actual Failure

Hit

False Alarm

Essex St

U1

3

6 3 17 5 7

Adrian St

U1

4 1

Curedale Mews

U2

5

1 4

0 0 0

0 0 0

Curedale Mews

U1

Essex St

U2

Preston Pt Rd

U2

Samson St

U2

86 MAY 2010 w ater

indicate an underlying unstead iness of the performance and the necessity for attent ion and f urther investigation.

1

4

Dr Stephen Beckwith is Operations Technology Manager, Operations Centre, WA Water Corporation. Email Stephen. beckwith@watercorporati on. com.au. Nan Ning and Dr Jaime Adeane are Operations Data Analysts, Operations Centre, WA Water Corporation. Email: {nan.ning, jaime.adeane} @watercorporat ion.com.au

technical features

Profile for australianwater

Water Journal May 2010  

Water Journal May 2010