Water Journal November - December 1995 by australianwater

AUSTRALIAN WATER & WASTEWATER ASSOCIATION

NOVEMBER/DECEMBER 1995

AUSTRALIAN WATER & WASTEWATER ASSOCIATION

Volume 22, No 5 November/December 1995

Editorial Board F R Bishop, Chairman

CONTENTS

B N Anderson, G Cawston, M R Chapman P Draayers, W J Dulfer, G A Holder M Muntisov, P Nadebaum,J D Parker AJ Priestley,] Rissman

ASSOCIATION NEWS From the Federal President From the Executive Director Association Meetings

2 4 5

MY POINT OF VIEW A Journey Through Environmentalism

Professor David Shearman

AWW A Federal Office PO Box 388 Artarmon NSW 2064 Level 2, 44 Hampden Road Artarmon NSW 2064 Tel (02) 413 1288 Fax (02) 413 1047

Features Editor

FEATURE - WATER QUALITY CRC for Water Quality and Treatment

DBursill Drinking Water Quality and Treatlflent Requirements: A Risk-Based Approach

M Stevens, S McConnell, PR Nadebaum, M Chapman, S Ananthakumai', J McNeil Risk Assessment in Microbial Water Quality Criteria:

CNHaas Epidemiology: the Safety of Australian Water Supplies

R M Douglas, L S Pilotto

MANAGEMENT Organisational Restructuring of Water Authorities

B A Nosworthy Can Water Trading Achieve Environmental Goals?

H Bjornlund,J McKay

Advertising & Administration

â&#x20AC;˘

Chichester Trunk Gravity Main: An Asset Management Application

31 35

E A (Bob) Swinton 4 Pleasant View Crescent Glen Waverly Vic 3150 Tel/Fax (03) 9560 4752

Branch Correspondents ACT - Ian Bergman Tel (06) 248 3133 Fax (06) 276 1997 New South Wales - Mitchell Laginestra Tel (02) 412 9974 Fax (02) 412 9876 Northern Territory - Graeme Reed Tel (089) 82 7346 Fax (089) 82 7221 Queensland - Ted Cusack Tel (07) 383 1 7316 Fax {07) 3832 1625 South Australia - Peter Martin Tel (08) 303 8723 Fax (08) 303 8750 Tasmania - Dao Norath Tel (002) 332 596 Fax (002) 347 559 Victoria - Mike Muntisov Tel (03) 600 I 100 Fax (03) 600 1300 Westem Australia - Alan Maus Tel (09) 420 2465 Fax (09) 420 3178

G Pjke, G Hales WATER (ISSN 0310- 0367)

ENVIRONMENT Western Australian Seagrass Transplant Experiments

is published six times per year January, March, May,July, September, November by

Australian Water & Wastewater Inc

CJ Walker Role of Sediments in Phosphorus Cycling in Aquatic Systems

W Maher, I Lawrence, T Donnelly

ARBN 054 253 066

Federal President

DEPARTMENTS International Affiliates From th@ Battam of the Well Books Industry News Product News Meetings

Richard Marks

9 9 45 45

46 48

OUR COVER

Will Ju just play with it, or drlttk it? The quality ofAustralian water supplies has been the subject of much discussion over the years, with some arguing that 'WHO rules, OK?' The Draft Australian Guidelines are pragmatic, recognising that some Australian water sources are virtually pristine, unlike the usual sources in the First and the Third Worlds. However, it seems inevitable that any values quoted in the guidelines will be legally enforced by the regulators. Limits should be based on real assessment of risk rather than 'slavishly' adopting limits derived from different circumstances. The research necessary to accomplish this will only be achieved by cooperation. The new Cooperative Research Centre for Water Quality and Treatment, featured in this issue, involves every sector of the water industry together with public health research groups.

Executive Director Chris Davis Australian Water & Wastewater Association assumes no responsibility for opinions or statements of facts expressed by contributors or advertisers and editorials do not necessarily represent the official policy of the organisation. Display and classified advertisements are included as an informational services to readers and are reviewed by the Editor before publication to ensure their relevance to the water environment and to the objectives of the Association. All material in Water is copyright and should npt be reproduced wholly or in part without the written permission of the Editor.

Subscriptions Water is sent to all members of the AWW A as one of the privileges of membership. Non members can obtain Water on subscription at an annual subscription rate of $35 (surface mail).

WATER QUALITY

DRINKING WATER QUALITY AND TREATMENT REQUIREMENTS: A RISK-BASED APPROACH M Stevens*, S McConnell, PR Nadebaum, M Chapman, S Ananthakumar,j McNeil Abstract The use of a risk-based approach to the management of water quality focuses attention on the key sources of risk, and allows informed decisions regarding the requirements to treat drinking water to protect public health and aesthetic quality. This paper outlines work taking place to develop risk-based water quality and treatment guidance for Australian waters. The work involves establishing and linking together algorithms which relate water quality and risk, and which relate water quality to factors associated with the catchment, reservoir, treatment system, disinfection, distribution system, and the consumer's supply system. These algorithms are being developed for the important water constituents (including microbiological and chemical constituents), drawing on the considerable information base of the major water authorities and linking this with epidemiological, risk assessment and engineering thinking. The ultimate objective of this work is to define the requirements for cost effective management of particular water supplies, and to define treatment requirements in terms of achieving an acceptable level of risk; rather than relying on empirical rules and indicators of risk.

Keywords Water quality, water treatment, risk assessment, environmental management.

Introduction The

1995

NHMRC/ ARMCANZ

Australian Drinking Water Guidelines and the 1993 WHO Guidelines for Drinking Water Q,uality are intended to define 'good' and 'safe and acceptable' water quality respectively; reflecting consideration of both health and aesthetic effects. The philosophy underlying both sets of guidelines involves a "risk-based" approach to setting the individual healthbased criteria; that is the guideline values reflect toxicological and aesthetic considerations, rather than, say, detection limits. For many constituents there is sufficient toxicological information available on 12

which to base maximum acceptable concentrations, however, for some constituents, the toxicology is uncertain and change in guideline values for these constituents can be expected as data is accumulated. The WHO guidelines have the objective of providing guidance of general applicability, but with the suggestion that water quality criteria can be varied to suit local requirements. The NHMRC/ ARMCANZ guidelines include provision for the development of catchment-specific water quality objectives which recognise the risk of the water quality supplied to consumers and, in principle, some variation from the generic guidelines is possible. A riskbased approach of this type is consistent with the philosophy adopted by regulatory authorities in setting other environmental guidelines. However, to date, the tools required to develop and apply risk-based water quality criteria have not been available. The work reported in this paper seeks to draw together research from Australia and overseas to develop an integrated and consistent risk-based framework for the management of drinking water supplies. It has been funded by a Landcare grant from the Department of Primary Industry and Energy, with support from Melbourne Water, and will be a continuing project in the CRC for Water Quality and Treatment. Several terms used in risk assessment and management are often used loosely and need to be defined: • Risk: the probability that an adverse outcome will occur in an individual or a group that is exposed to a particular dose or concentration of the hazardous agent (Langley and Van Alphen, 1993) • Probability: a measure of the likelihood expressed as the ratio of the number of times a given outcome occurs to the total number of outcomes. In the context of probabilistic analysis, it is used to reflect the probability that an individual will be subject to a given level of risk, given the variation throughout the population. • Hazard: the capacity to produce a par'ticular type of adverse health effect eg. one hazard of Giardia is gastro-enteritis.

Project Objectives The objectives of the project are to provide information for water managers on: • The preferred methodology for riskbased assessment of both microbiological and chemical water quality aimed at determining the characteristics governing risk in particular situations, and the target values consistent with a safe, aesthetically acceptable water in accordance with the current formulation of Australian water quality guidelines • The significance of catchment protection and the prioritisation of management approaches with respect to health risk • The requirements for water treatment, considering in particular the need to balance long-term 'chemical risk against short-term microbiological risk. Water treatment in this context includes water source protection, filtration, disinfection and prevention of contamination in pipework. The process for the development of microbiological quality criteria is generally less developed than for chemical characteristics. The concept of acceptable risk, and criteria based on achieving a numerical value of acceptable risk, has not generally been extended to microbiological characteristics. Rather, criteria have been set on the basis of simple "rules of thumb" for indicator organisms thought to be of general applicability across a wide range of water supplies. The application of these simple rules of thumb can have significant cost implications in assessing whether water supply systems require full treatment, and do not, for example, distinguish between catchments where pathogenic organisms are likely to be prevalent, and those where such organisms are less likely to be present. The use of indicator organisms only, does not adequately reflect risk to people drinking the water. Further, treatment processes which eliminate indicator organisms may, in some cases, result in long-term health risks due to residuals left after treatment. • CMPS&F, 12th flr, 390 St. Kilda Road, Melbourne Vic 3004

WATER NOVEMBER/DECEMBER 1995

Figure 1 Flow Sheet for Microbiological Risk

Assessing Total Risk Overview. A framework or methodology for the assessment of the total risk to water consumers involves a large number of elements, some of which may require further detailed consideration (eg. dose-response relationships for various pathogens). The information necessary to complete many of the elements of the overall framework is available from the literature. However, a methodology that allows each element to be combined in such a way as to estimate the total risk to water consumers has yet to be developed. Further, most risk estimates completed to date have focussed on narrowly defined scenarios that may adequately describe one set of conditions within a part of the system, but not the impact of the entire water "production line" on the risk to consumers. One aim of this project is to develop a framework that allows the effects of a realistic range of variability of conditions within the elements of this water quality "production line". This involves the defining of algorithms that relate the various parameters which determine the risk associated with water quality in the catchment, reservoir, filtration system, disinfection and distribution systems and the impact of the consumer's plumbing on water quality. These algorithms are being developed for the important water constituents (including both microbiological and chemical constituents) drawing on the information base of the major water authorities and linking this with epidemiological data, health risk assessment and engineering risk and reliability principles.

• Failure Mode Effects Analysis (FMEA) principles used in risk and reliability engineering • Human health risk assessment techniques, such as those employed by agencies such as the WHO , NHMRC and USEPA, including consideration of both acute and chronic health effects, and employing techniques such as Monte Carlo simulation • Epidemiological techniques, as a basis for validating predictions. Consideration is given to each of the elements in the "production line" that may cause poor water quality and risk to consumers. Microbiological contamination, chemical contamination and aesthetic aspects are considered. While aesthetic quality of water may not directly present a risk to consumer health, poor aesthetic quality can result in customers using less safe, but more aesthetically pleasing sources, or installing treatment systems in the home possibly increasing the risk to consumers. In addition, aesthetic quality of water is important when assessing the requirements for treatment. The framework for risk analysis has been developed for a generic system and involves the following key processes: • Hazard Identification • Exposure Assessment • Dose Response Assessment • Risk Characterisation. In application of these risk assessment

principles, one of the most important steps is identification of all significant "failure modes" . Once a failure mode (e.g. poor disinfection due to turbidity associated with a storm event) has been identified, the risk assessment process outlined above can be applied. One of the principle outcomes of this risk assessment framework is a methodology which will allow water quality managers to look from the customer's tap, back up through the water distribution, treatment and collection systems to identify how events, say, in the catchment, affect the water quality seen by consumers. An outline of the overall risk assessment framework for microbiological constituents is shown in Figure 1. This figure has been drawn to encompass as many possible risks as can be envisaged. For a given system the majority of these may not be relevant, in which case, for a preliminary assessment most could be deleted, thus focussing attention on the important elements.

Probabilistic Analysis Most of the risk estimates undertaken to date have been based on point estimates of exposure, usually reflecting either the "average" or "worst case" conditions to which a consumer is likely to be exposed. In practice, many of the parameters of relevance in risk assessment vary with time (e.g. seasonally, diurnally) or across the exposed population or, alternatively, are uncertain, reflecting limited knowledge. Use of point estimates for the assessment of risk often produces a false sense of precision, and fails to account for the inherent variability in many factors. Further, risk assessments using point estimates often focus on the average consumer, which can obscure the risk to sensitive sub-populations. Where the evaluation is based only on a worst case

....__

Framework The framework for the assessment of total risk has drawn on:

Source Water Concentration

Treatment and Disinfection

Treatment Plant

Tap Water

Effcctiveneu

Reliabilit y

Concentrat ion

Figure 2 Probabilistic Analysis

'""'F====:~===~::::::::====, C

800/o

t _! f.

l+,---

~%- - - -- -- - - - - - - - - - - <

l0%1 - - -- - - - - - - - - - - - - - - < 0% 1 - - - . - - - - - - - - - - - 0

L.12 .5

1.688

Daily Risk

U .5 o(

U IIJ

JJ7.5

J,9J8

~ ..5

Illness x I0~

Figure 3 Risk ofIllness from

Cryptosporidium scenario, which carries a high degree of conservatism, the risk to most individuals is overstated. Probabilistic analysis is one of several tools that allow for the inherent variability ar,u uncert..inty in many parameters to be considered. As part of this work, MonteCarlo simulation techniques have been used. Rather than assigning typical or worst case point estimate values to each of the parameters, e.g. water consumption rate, a probability distribution is assigned to each parameter. Monte Carlo analysis involves calculation of risk estimates using values selected at random, for each contributing parameter in accordance with the distributions specified. The process is repeated for, say, several thousand iterations, until a consistent probability distribution describing the risk to consumers is generated. This approach allows a more realistic estimate of risk and the variation of risk across the population, given that all individuals are not necessarily exposed to the same level of risk. Figure 2 illustrates the use of probability distributions to describe variables in the risk modelling. In the above framework, consideration has been given to the total system and its impact on consumer risk, rather than starting with the quality of water at the consumers tap. For example, consideration has been given to the effectiveness and reliability of chlorination, the correlation between chlorination effectiveness and turbidity, the impact of catchment management and storage design on raw water turbidity and the dieoff of pathogens, and regrowth of microorganisms in the distribution system. Care must be exercised in the application of Monte-Carlo and other probabilistic techniques since many of the factors relevant to the assessment of risk, e.g. Cryptosporidium concentration and turbidity, are co-related.

Event-Based Variations It is important that the estimation of risk can take into account the variation in water quality that can occur both through "normal" variations, and through particular events occurring. In practice, this can 14

be considered by applying fault tree principles, whereby all of the major events or situations which can apply are identified, and the probability of the condition occurring is identified. Fault trees for the situations or events which determine the various water quality conditions can be derived from considering the situations which are embodied in the framework outlined in Figure 1. Thus, fault tree analysis can consider situations such as the following: â&#x20AC;˘ Where a chlorinator is working normally, and when it has failed â&#x20AC;˘ Where a storm has occurred in a catchment and has resulted in a slug of dirty water entering the reservoir or â&#x20AC;˘ Where a reservoir is low or high, and there may or may not be short circuiting. This type of analysis gives rise to a large number of situations which contribute to the overall risk, and enable the combinations of situations (or "pathways") to be identified which are most significant in determining the level of overall risk. This approach is being used in the modelling work which is being undertaken and, when the most significant pathways are determined, these are assessed in more detail with Monte Carlo techniques as described above .

Applying the Framework The abcve framework is being used to assess the most important sources of risk associated water quality as water moves along the "production line" from the catchment to the tap. This can allow. identification of those areas where the greatest reduction in overall risk can be achieved for a given expenditure limit. Such cost-benefit analysis, focussing on the risk reduction achieved by improvements, is an essential element of "best practice" management of water supply systems. In most Australian water supplies sourced from surface water, the primary health concern centres on the microbiological quality of water, rather than on chemical constituents, although byproducts of disinfection and treatment residuals can sometimes be of concern. Disinfection by-products may be of concern particularly where coloured, unfiltered waters of poor quality require high doses of chlorine for disinfection prior to supply to the consumers. The framework outlined above is designed to provide the basis for estimation of risk associated with both pathogens and disinfection by-products, allowing the relative risk from each of these sources to be balanced and an optimum water quality management strategy to be developed for each supply system.

Risk Modelling Cryptospor1dium. Cryptosporidium parvum is an amoebic flagellate that

reproduces in the intestine of the host and produces infective oocysts that are excreted in the host's faeces. Waterborne transmission of Cryptosporidium was not documented until 1974 when the parasite was identified by a veterinary scientist as the causative agent of gastroenteritis in a three year old girl. International interest on this subject has intensified since an outbreak of Cryptosporidiosis in Milwaukee in 1993, affecting approximately 400,000 people. The source of the outbreak was contaminated drinking water, attributed to storm events in the catchment resulting in a large plume of dirty water entering the reservoir directly above the treatment plant inlet, combined with less than optimal treatment plant operation due to a change in the flocculant used (Fox and Lytle, 1994). Cryptosporidium is considered by the USEP A to be one of the three most common disease-causing organisms in the world and to be responsible for 25% to 30% of diarrhoeal outbreaks where there is no other known cause (Milwaukee Journal, 1993). Cryptosporidium has been selected to demonstrate some of the factors which need to be considered in development and application of the risk assessment framework to assess the total risk to the consumer, and in , particular, to demonstrate some of the likely interactions within the water supply system. For the purposes of this illustration, a simple closed catchment, dam storage (long detention time) and disinfection only system has been assumed. The risk assessment process for Cryptosporidium has been simplified for the purposes of illustration; some complex factors that effect microorganism concentration have not been incorporated due to a lack of data. The cumulative frequency distribution for the estimated risk to the consumer associated with Cryptosporidium is shown in Figure 3. The Cryptosporidium concentrations used in this case study are based on monitoring data from a protected catchment. Using Monte-Carlo analysis, the resultant estimate of risk to water consumers is expressed as a probability distribution, rather than as a simple point estimate or single value. For the example presented in Figure 3, the 50th percentile or median daily risk was estimated to be in the order of one in one thousand. There are few published estimates of the risk to consumers in the United States associated with similar scenarios for Cryptosporidium

Disinfection By-Products The first evidence that disinfection of drinking water resulted in the formation of by-products was reported in 1974 when the formation of trihalomethanes (THMs) was observed in water that had WATER NOVEMBER/DECEMBER 1995

been chlorinated (Bellar et al., 1974; Rook, 1974). It is now recognised that THMs are just one of many by-products of disinfection. Chlorination of drinking water is the major source of THMs to the consumer and the predominant THM formed is chloroform. Other THMs are dibromodichloromethane, dibromochloromethane and bromoform. THM formation depends on raw water quality (presence of precursor species), pH and temperature of the water, chlorine dose and contact time. There is only limited evidence oflong-term effects of exposure to chlorinated by-products, although epidemiological evidence suggests they may be responsible for an increase in certain cancers in the community. Research conducted using experimental animals show individual by-products can be carcinogenic when ingested in large amounts (Bull, 1992). The International Agency for Research on Cancer has concluded that chloroform and bromodichlormethane are possibly carcinogenic to humans (Group 2B, inadequate evidence in humans, but sufficient evidence in animals), and that bromoform and dibromochloromethane are not classifiable as to their carcinogenicity to humans (Group 3, inadequate evidence in humans and limited evidence in animals) (IARC, 1991). The risk framework discussed in Section 3 was applied, using information on THM production in a rural Victorian water supply of generally good quality (colour of25 to 50 HU) disinfected using 1.9 mg/L chlorine. The THM concentrations in the production measured in the order of 74 µg/ L, consisting of 70 µg/L chloroform and 4 µg / L bromodichloromethane. These levels fall below the maximum concentration recommended by both the NHMRC and WHO guidelines. Probability distributions were used to account for water consumption, body weight and exposure time. The incremental lifetime cancer risk estimated from consumption of the above water is shown in Figure 4, with the 95th percentile risk estimate in the order of three in 1,000,000 (based on WHO dose-response factor for chloroform).

Key Issues and Data Gaps Dose-Response Relationships for Pathogenic Organisms. Assessment of the health risks associated with microbiological contaminants requires knowledge of several parameters that determine the likelihood that an individual will become ill from ingestion of water containing the microorganism, including the virulence and concentration of the pathogen, and susceptibility of the host. Dose-response curves may be generated, and can be used to predict the likelihood that an individual will become infected given a certain level of exposure to a pathogen. Infection and illness rates

WATER NOVEMBER/DECEMBER 1995

are determined by the virulence of the microorganism, the concentration of the microorganism, and the resistance to infection by the host. Some sub-populations have an increased susceptibility to waterborne illness, especially immuno-compromised individuals such as those suffering from HIV or cancer, the very young and the elderly. For example, gastrointestinal illness, which is the most common form of waterborne microbial illness, is not generally considered life-threatening in healthy adults. However, research has shown that mortality rates for gastro intestinal illness in the elderly and in very young infants, range from 3 to 5%, (Glass, 1991). Similarly, Cryptosporidiosis is a leading cause of death in AIDS patients in the US. In Australia, Cryptosporidiosis was the AIDS determining condition for 2.3% of AIDS patients (National AIDS Database, 1995). Attempts have been made to define dose response curves for infective agents in water principally by exposing healthy adults to known concentrations of the pathogen or from data gathered during epidemics of water-borne disease. These are the only two sources of dose-response information and neither method is entirely adequate. Information gained from feeding studies has the advantage that a known population is exposed to a quantifiable concentration of microorganisms, however, the infection of a restritted portion of the community does not allow the study of the effects of infection on a range of sub-populations. The use of epidemic data allows the observation of the effect of infection across the whole community, although the number of persons exposed and the concentration of the microorganisms is not known. Dose-response curves have been defined for a limited number of waterborne disease-causing microorganisms. This limits the applicability of the assessment of risk associated with drinking water and focuses effort on a few reasonably well defined microbes, for example Cryptosporidium and Giardia. Other important issues in the assessment of risk associated with microbiological contamination of drinking water include the extent of cross infections and the effects of simultaneous exposure to multiple pathogens. While some information is available on this subject, further research focussing on the incidence of low-level gastrointestinal disease is required. The intervention approach adopted in Canada (Payment, 1991) would provide a sound basis for such research in Australia.

microorganisms in the raw water. For those microorganisms of primary concern, for example, Cryptosporidium, Giardia, enteric viruses and bacteria such as Campylohacter, only limited data are available for Australian waters. While considerable overseas information on the occurrence of some pathogens in water supply systems is available, particularly from the United States, the relevance of this information for Australia is uncertain. Many of the pathogens detected in other qmntries have also been detected in Australian waters, however the reasons for differences in the range in concentrations are yet to be assessed. Some information is available from the major water authorities regarding the concentrations of specific pathogens, rather than indicator organisms, however its use is limited due to the relatively high detection limits and variable recovery rates. The collation of high quality data on Australian waters is a priority, as is the standardisation and improvements in analytical techniques for pathogen detection and enumeration.

Fate of Contaminants A significant amount of qualitative information is available regarding the fate of chemical and microbiological contaminants within water supply systems. However, in order to reliably assess the risk to the consumer it is important to establish quantitative relationships . At present there is reasonably good information on the change in concentration along the "production line", however, the reasons for the observed changes are often unknown. One advantage of the risk assessment framework is that it helps focus attention in a systematic way on the processes within the water supply system that affect water quality and the risk to consumers (e.g. bacterial regrowth). Further, the framework outlined in this paper facilitates the assessment of risk reduction associated with various water quality improvement strategies. However, in order to achieve this, it will be necessary to further develop the currently available predictive fate models for chemical and microbiological contaminants in the water supply system.

S ~Monk:C&ito

,.... f===,@l=="=="'='=·==. ,.1=""======· = =·==-=·= ·"""======j g

IO¾ /

.... 1 . / - - - - - - - - - - --

l. ~

---i

IJ___ _ _ _ _ _ ___,,

,ov. L-- - - - - - - - - - - - 1

··· L - - - - - - - - - - - - - '

Occurrence of Pathogens One of the principal input parameters in the assessment of the microbiological risk to consumers is the concentration of

locnmmtal Ufetime Risk of Cane« 1 10'6

Figure 4 Risk of Cancer from T rihalomethanes

Comparison of Dissimilar Risks When the overall risk to a. consumer is assessed, consideration needs to be given to all forms of risk, including: • The risk of cancer or other chronic health effects associated with long term exposure to chemical contaminants in water • The risk of acute health effects associated with major spills, leaks or other similar events (e.g. blue green algae toxins, risk of chemical overdosing or high doses of microorganism due to water-sewer cross connections follow repairs of bursts) • The occurrence of endemic illness, such as gastroenteritis, associated with microorganisms present at relatively low levels in the water supply due to ineffective treatment • The risk of an epidemic, associated with failure of treatment systems when there is a significant excursion in raw water microbiological water quality, (e.g. associated with a storm event) • The risk of adverse tastes, odours, laundry staining or other detrimental aesthetic effects. Each of these sources of risk result in differing frequencies and severity of effect as well as costs to the community. A means to weight each risk and potential cost is therefore difficult and often subjective judgement is the best answer. The value of the proposed framework and modelling work is that it provides better evidence upon which to make such judgement. For example, the acceptable lifetime risk of cancer associated with chemical contamination is often cited as being between one in 1,000,000 and one in 10,000, whereas the acceptable risk of gastrointestinal illness associated with consuming pathogens may be in the order of one in 10,000 per year. Judgements regarding the acceptability of risk require the input of both the community and the professionals working in the field, balancing the risk of illness from the many other sources and the social and economic cost of waterborne illness.

Management Implications The principal advantage associated with the framework outlined in this paper is that it provides a tool which will permit water quality managers to identify and semi-quantitatively assess the major sources of potential risk within a water supply system, from the catchment to the tap. It enables water managers to make more informed decisions regarding: • The basis for improving water quality in terms of reduction in risks and overall costs to customers • Negotiations with stakeholders about acceptable risk • The most appropriate (and lowest overall cost) means by which such improvements in water quality can be achieved • The contaminants or sources which pose the greatest risk. 16

When these tools are correctly applied, a water supply authority is more likely to be able to balance, for example, the cost of constructing a water filtration plant against the risk reduction likely to be achieved, and to compare the relative benefit of such a strategy with less costly alternatives. Similarly, application of the framework outlined in this paper may help to identify deficiencies in the water quality "production line" and help define the priorities, or order of work, needed to redress these deficiencies. One of the potential difficulties with the development and application of catchment-specific water quality objectives is the cost involved, particularly the cost associated with the collection of sufficient system-specific data on water quality, and the on-going water quality monitoring requirements. Such costs can be readily justified for very large urban supplies where the costs of treatment and management are large. However, for smaller supplies the cost may not be justified, and in such situations the managers of smaller systems may draw on information obtained from the assessment of large urban supplies with similar characteristics. In this respect, the application of risk-based techniques to larger systems should lead to an improved understanding of what water quality information is most necessary for the management of smaller supplies.

Conclusions A framework which will allow the semi-quantitative assessment of the total risk to consumers associated with drinking water quality is being developed and used to• identify the most important sources of risk in Australian public water supply systems. The objective of this work is to provide a basis for the development of catchment-specific water quality objectives, and to assist water quality managers in making more informed decisions about the requirements for management of their supplies and the level of treatment required. The risk assessment framework is designed to identify the sources of risk and their significance, and the benefits which can flow from appropriate management of the water "production line" including catchment and storage management, filtration, disinfection, and distribution system maintenance. This should provide more cost effective management of water quality by identifying the actual sources of risk within a system and the requirements for management and treatment, rather than defining these requirements on the basis of indicators of contamination and empirical rules of thumb. It is expected that information gained using this approach on larger urban systems will be transferable to smaller systems with similar catchments and treatment and distribution systems.

Reference Bellar TA, Lichtenberg] J, Kroner RC (1974) The occurrence of organohalides in chlorinated drinking water.] Am Water Works Assoc. 66: 703 Bull RJ (1992) Toxicology of drinking water disinfection. p. 184-230. In: Lippmann, M. (ed.)

Environmental Toxicants: human exposures and their health effects. Van Nostrand Reinhold, New York. Fox KR, Lytle D A (1994) Cryptosporidium: The Milwaukee experience and relevant research. Proc Am. Water Works Assoc : 505-516. Glass RI , Lew J F, Gangarosa RE, et al. (1991) Estimates of morbidity and mortality rates for diarrhoeal disease in American children. ]. Pediatr. 118: 27-33. IARC (1991). IARC monographs on the evaluation of carcinogenic risks to humans: chlorinated drinking water; chlorination by-products; some other halogenated compounds; cobalt and cobalt compounds. World Health Organisation International Agency for Research on Cancer, 52, Lyon. Langley A, Van Alphen M (1993) Eds. Proc. 2nd. Natl Workshop on th e Health Risk Assessment and Management of Contaminated Sites. South Australian Health Commission, Adelaide. Milwaukee Journal (1993) A comp endium of reports relating to the Milwaukee outbreak of cryptosporidiosis. Milwaukee Journal Lobby, Public Service Bureau. Milwaukee. National AIDS Data Base (1995) National Centre for HIV Epidemiology, St. Vince nts Hospital, Sydney. NHMRC-ARMCANZ (1995) Australian Drinking Water Guideli;es National H ealth and Medical Research Council - Agricultural and Resource Management Council of Australia and New Zealand. 2nd Draft. March. Payment P, Richardson L, SiemiatyckiJ, Dewar R, Edwardes M, Franco E (1991 ). A randomised trail to evaluate the risk of gastrointestinal disease du e to co nsumption of drinking water meeting current microbiological standards. Am Jnl of Public Health 81 (6):703. Rook] J (1974) Formation of haloforms during ch lorination of natural waters. Water Treat Exam 23 (2):234. WHO (1993) Guidelines far Drinking Water Q,uality. 2nd Edition. World Health Organisation, Geneva.

Authors As outlined in the introduction, all authors are from agencies which are partners in the CRC Water Qyality and Treatment. The CMPS&F Environmental team comprises the Project Manager, Stuart McConnell, a chemical engineer specialising in risk assessment, Dr Melita Stevens, a microbiologist, specialising in water and wastewater aspects and Dr Peter Nadebaum, who is the Principal in the Environmental Management Group. Melbourne Water fielded Mike Chapman, a chemist/chemical engineer, who is Senior Engineer, Water Technology and Dr Sri Ananthakumar, a civil engineer, responsible for water treatment. John McNeil is Professor of Social and Preventative Medicine at Monash University. WATER NOVEMBER/DECEMBER 1995

WATER QUALITY

RISK ASSESSMENT IN MICROBIAL WATER QUALITY CRITERIA C N Haas* Abstract Risk assessment methods can be used to develop microbial guidelines or assess microbial water quality. This paper reviews the genesis of the quantitative microbiological risk assessment methodology, and provides an example of the development of risk-based water quality guidelines for virus, Giardia and Shigella in finished drinking water and marine recreational waters used for contact recreation.

Key Words Water quality, risk assessment, epidemiology, indicator organisms

Introduction Historically, the acceptable microbial levels for drinking water, contact recreational waters, and shellfish harvesting waters have been set using indicator organisms, most often the coliform (either total or fecal} group. While there have been some attempts to develop criteria using indicator organisms based on epidemiological evidence, particularly in recreational waters (4, 9), such correlations are far from universally accepted (11, 12, 36, 37}. In recent years, there have been strong incentives to reconsider disinfection processes used to attain acceptable water quality so as to minimise potential adverse ecological or human health effects particularly associated with chemical disinfection processes (1, 2, 5- 7, 19, 21 - 23, 28, 39,41}. With the advent of better methods for direct measurement of pathogens in water (15, 16, 24, 27, 30, 31), and the development of risk assessment paradigms for setting environmental standards {26, 35, 38), it has become time to consider the application of these methods for development of microbial standards of acceptable water quality with an eye towards supplementation or replacement of traditional indicator measurements. This paper summarises the use of a risk assessment framework to develop such standards for microbial quality of water used for direct potable consumption, contact recreation, or shellfish consumption.

Framework The quantitative microbiological risk assessment (QMRA) approach follows 18

the framework proposed for chemical risk assessment (26}, and broadly includes the following steps: • Hazard assessment • Exposure assessment • Dose-response analysis • Risk characterisation • Risk management. For microorganisms, hazard assessment (i.e., identification of a pathogen as an agent of potential significance} is generally a straightforward task, except for opportunistic agents. The major tasks of QMRA are focused on exposure assessment, dose analysis, and risk characterisation. The task of risk management is one of deciding the necessity of any action based on the risk characterisation outputs, and incorporates significant policy and trans-scientific concerns. However, a risk analyst needs to be aware of the levels of significance that might require an action to be taken. One outcome of the hazard analysis is a decision as to the principal consequence(s} to be quantified in the formal risk assessment. With microorganisms, consequences may include infection (without apparent illness}, morbidity or mortality; furthermore these events may occur in the general population, or at higher frequency in susceptible sub-populations. Although mortality from infectious agents, even in the general population, cannot be regarded as de minimus (20}, the general tendency (in water microbiology} has been to regard infection in the general population as the particular hazard which should receive protection. This has been justified based on a balancing between the degree of conservatism inherent in using infection as an endpoint and the {current} inability to quantify the risks to more susceptible sub-populations (25}. The purpose of an exposure assessment is to determine the microbial doses typically consumed by the direct user of a water {or food}. In the case of water microbiology, this may necessitate the estimation of raw water microorganism levels followed by estimation of the likely changes in microbial concentrations with treatment, stora~e and distribution to the end-user (29, 32). In the case of shellfish water quality, additional information on the bioaccumulation of pathogens, and possible loss due to depuration and other processes is necessary in the case of

wastewater discharges to receiving water. A second issue arising in exposure assessment is the amount of ingested material per "exposure". As a default number, 2 Uperson-day is used to estimate drinking water exposure (25}, although this may be conservative (33}. For contact recreational exposure, 100 mU day has often been assumed as an exposure measure{l 7}, however actual data to validate this number is lacking. For shellfish exposure, market surveys may be used (34), although it is believed that there may be sub-populations with consumption levels substantially higher than average. In QMRA, for many microorganisms, human dose-response studies are available which can be used to estimate the effects of low level exposure to microorganisms. In prior work, it has been found that these studies may be adequately described by one of two semi-mechanistic models of the infection process. In the exponential model, which may be derived from the assumption of random occurrence of microorganisms along with a constant probability initiation of infection by a single organism (r}, the probability of infection (Pi) as a function of the ingested dose (d} by: (1) Pi= 1- exp(-rd} For many microorganisms, the doseresponse relationship is shallower than reflected by equation (1), suggesting some degree of heterogeneity in the microorganism-host interaction. This can be successfully described by the beta Poisson model, which can be developed from (1) if the infection probability is itself distributed according to a beta distribution (13, 14, 18}. This model is described by two l'arameters, a median infectious dose (N50) and a slope parameter{~ :

Figure 1 depicts the effect of the slope parameter on the dose-response relationship; in the limit of a-too, equation (2) approaches equation (1). Given a set of dose-response data, i.e., exposure of populations to various doses of microorganisms and measurement of response (such as infection), the best fit• Professor of Environmental Engineering, Drexel University, Philadelphia, PA 19104 USA

WATER NOVEMBER/DECEMBER 1995

Figure 2. Best fit

0.75

(solid line) and confidence limits (dashed lines) of beta-Poisson model to dose-response data (points) ofHuman Rota virus (Ward, 7968). Not shown are doses of 0.09 and 0. 009 pfu, both of which had zero response.

0 .5

0.25

I I 0.000 I

0.00 I

0.0 I

0. 1

dosc/ N 50

100

Figure 1. Comparison of exponential

and beta-Poisson dose response models

10Âˇ3 ...,..............................._.......w.,,._._.......................~ ~ - - -

ting parameters of a dose-response relationship may be computed via standard maximum likelihood techniques. The method has been illustrated for human rotavirus (20, 29) and protozoans (32) . Confidence limits to the parameters can then be found, and used as a basis for low- dose extrapolation. The process of risk characterisation combines the information on exposure and dose-response into an overall estimation of likelihood of an adverse consequence. This may be done in two basic ways. First, a single point estimate of exposure (ie number of organisms ingested) can be combined with a single point estimate of the dose-response parameters to compute a point estimate of risk. This may be done using a "best" estimate, designed to obtain a measure of central tendency, or using an extreme estimate, designed to obtain a measure of consequence in some more adversely affected circumstance. An alternative approach, which is receiving increasing favour, is to characterise the full distribution exposure and dose-response relationships, and to combine these using various tools (for example, Monte Carlo analysis) into a distribution of risk. This approach conveys important information on the relative imprecision of the risk estimate, as well as measures of central tendency and extreme value (3, 10). One important outcome of the risk characterisation process using a Monte Carlo approach is the assessment of the relative contribution of uncertainty and variability to a risk estimate. Variability may be defined as the intrinsic heterogeneity that leads to differential risk among sectors of the exposed group, perhaps resulting from differential sensitivities or differential exposures. Uncertainty may be defined as the factors of imprecision and inaccuracy which limit the ability to exactly quantify risk. Uncertainty may be reduced by additional resources, for example devoted to characterisation of the dose-response relationship . Variability represents a lower limit to the overall risk distribution. The results of a risk characterisation are used in risk management. The understanding of appropriate action levels for decision making with respect to microorganisms is still at an early stage.

WATER NOVEMBER/DECEMBER 1995

0.01

0.1

100

dose

(pfu)

However, in the case of waterborne protozoa, it has been suggested than an annual risk of infection of 0.0001 (ie one in ten-thousand) is appropriate for drinking water (25) . In the case of contact recreation, correlations between highly crediduring ble gastroenteritis rates epidemiological surveys with microbial levels (4, 9) may be used along with a presumption that classical (in USA, 200 fecal coliforms/ 100 mL) water quality standards yield an acceptable level of risk to determine that daily illness rates of 19/ 1000 in marine waters and 12/ 1000 in fresh waters are appropriate benchmarks.

Application We have been collecting and analysing dose-response data for pathogens of importance in water quality management. Figure 2 provides an example of a fit to experimental ds se-response data, in the particular case of human rotavirus, using the data of Ward et al (40). Table 1 abstracts several of these. These parameters can be used to estimate the risk associated with a given exposure, or alternatively, given a measure of acceptable risk, the microbial quality standard that is necessary to attain

that objective. Note that the above numbers are maximum likelihood, or best estimators. It may also be appropriate to consider confidence limits in order to introduce a margin of safety. There is a growing literature on the use of uncertainty analysis in risk assessment (3, 10), and the future application of these techniques to microbial quality assessment is anticipated to grow. Table 2 illustrates the results of these calculations for an acceptable water quality associated with drinking water exposure (consumption 2 L/day) and exposure via marine bathing water (intake 100 mL/day) . So, for example, a rotavirus level of 0.46 per million litres would yield acceptable risk levels in drinking water, and a Shigella level of 3.64 per litre would yield acceptable risk levels in marine waters. When multiple pathogens are present, then as a first approximation, additivity may be assumed, although data to validate this assumption are lacking. These computations have implications for monitoring programs. Given the very low level of pathogens that would present a barely unacceptable level of risk in finished drinking water, use of pathogen monitoring in this circumstance is inappropriate. Instead, the quality of

Table 1. Summary of Selected Dose-Response Parameters

Organism

Model

rotavirus

beta-Poisson

Giardia

exponential

Shigella

beta-Poisson

a 0.2531

Reference (20) (32)

0.2099

(8)

Table 2. Computation ofAcceptable Water Qjtality Levels far Two Exposure Scenarios

Drinking

Marine Swimming

0.1

Daily Intake (L) Acceptable daily risk

2.74

10- 7

0.017

(Ml 000 000 L)

(#/L)

rotavirus

0.46

0.030

Giardia

0.86

Shigella

3.64

finished drinking waters might best be assessed by monitoring source water, and inferring finished water quality based on previously documented pathogen removal capacity. Indeed, this is the approach that is underlying the US Surface Water Treatment Rule (SWTR) (25). In recreational water, however, the pathogen levels that would pose an adverse risk are high enough to be detected, if they were present, using ordinary or molecular genetic methods (although issues of viability, sensitivity and selectivity for these latter methods remain to be completely understood) (30).

Summary and Conclusions

IÂˇ I

I I

Although the use of indicators in water microbiology will undoubtedly continue to play an important role, it is becoming possible to develop criteria for acceptable quality using the same risk assessment framework that is being applied for chemicals. There are some unknowns, including issues of relative susceptibility of diverse sub-populations, effects of multiple exposures, secondary cases, possible immunisation by prior infection, and intake/ingestion quantification. However, the framework is clearly amenable to use. While in drinking waters it will likely continue to be relatively difficult to verify attainment by direct finished water pathogen monitoring, in recreational waters this possibility may exist. Current coliform standards date from the first decade of this century; perhaps we may approach the future century with a more rigorous application of the tools of microbial risk assessment to public health protection.

Acknowledgements The work of the author has been partially supported by funding from the American Water Works Association Research Foundation. The author appreciates many useful interactions with Charles P Gerba and Joan Rose, and with his students, including Chris Crockett and Aamir Fazil.

References I. Bean RM (1983) Recent Progress in the Organic Chemistry of Water Chlorination, p 843- 850. In R Ljolley, WA Brungs,J A Cotruvo, RB Cumming,] S Mattice, and V AJacobs (ed) ,

Water Chlorination: Environmental Impact and Health Effects, vol 4, Book 2. Ann Arbor Science, Ann Arbor, Michigan. 2. Bull RJ, Kopfler F C (1991) Health Effects of Disinfectants and Disinfection Byproducts. AWWA Research Foundation and A WWA, Denver. 3. Burmaster D E, Anderson P D (1994) Principles of Good Practice for the Use of Monte Carlo Techniques in Human Health and Ecological Risk Assessment. Risk Analysis. 14 (4) 477-481. 4. Cabelli VJ (1983) Health Effects Criteria for Marine Recreational Waters. US EPA. 5. Craun G F, Bull R, Clark R, DoullJ, Grabow W, Marsh G, Okun D, Regli S, Sobsey M, Symons J (1994) Balancing Chemical and

Microbial Risks of Drinking Water Disinfection, Part I. Benefits and Potential Risks.jnl WaterSRT - Aqua. 43 (4) :192-9. 6. Ibid. Part II. Managing the Risks. 7. Craun G F (1991) Epidemiologic Studies of Organic Micropollutants in Drinking Water. In 0 Huntzinger (ed), "The Handbook of Environmental Chemistry " vol SA - Water Pollution. Springer-Verlag, Berlin. 8. Crockett CS, Haas C N, Fazil A, Rose] B, and Gerba C P. in press. Prevalence of Shigellosis in the US : Consistency with Dose-Response Information. lntljnl ofFood Microbiology. 9. Dufour AP (1984) Health Effects Criteria for Fresh Recreational Waters. US EPA. 10. Finkel A M (1990) Confronting Uncertainty in Risk Management Resources for the Future, Center for Risk Management, Washington DC. 1 I. Fleisher J M (1990) The Effects of Measurement Error on Previously Reported Mathematical Relationships Between Indicator Organism Density and SwimmingAssociated Illness: A Quantitative Estimate of the Resulting Bias. lntljnl of F,pidemiology. 19 (4) 1100-6. 12. Fleisher J M (1991 ) A reanalysis of data supporting US federal bacteriological water quality criteria governing marine recreational waters. Research journal of the Water Pollution Control Federation. 63 (3): 259. 13. Furumoto WA, Mickey R (196 7) A Mathematical Model for th e InfectivityDilution Curve of Tobacco Mosaic Virus: Experimental Tests. Virology. 32: 224. A 14. Furumoto WA, Mickey R, (1967) Mathematical Model for the lnfectivityDilution Curve of Tobacco Mosaic Virus: Theoretical Considerations. Virology. 32: 2 16. 15. Gerba C P, Rose] B (1990) Viruses in Source and Drinking Water. In G A K McFeters (ed), "Drinking Water Microbiology". SpringerVerlag, New York. 16. Gregory J (1994) Cryptosporidium in Water Treatment And Monitoring Methods. Filtration and Separation. 31 (3):283-289. 17. Haas C N (1983) Effect of Effiuent Disinfection on Risks of Viral Disease Transmission via Recreational Exposure. ]WPCP 55: 1111 -6. 18. Haas C N (1983) Estimation of Risk Due to Low Doses of Microorganisms: Comparison of Alternative Methodologies. American journal ofF,pidemiology. 118 (4) :573-5820. 19. Haas C N, et al (1987) Assessing the Need for Wastewater Disinfection. ]WPCF. 59: 856-64. 20. Haas C N, Rose] B, Gerba C, Regli S (1993) Risk Assessment of Virus in Drinking Water. Risk Analysis. 13 (5)545-552. 2 1. Isaac RA, Morris] C (1990) Modelling the Impact of Nitrogenous Compounds on Wastewater Disinfection, p 29. In Jolley R L et al (e d), Water Chlorination: Chemistry, Environmental Impact and Health Effects, vol 6. Lewis Publishers, Chelsea, Michigan. 22. Jolley R L {1975) Chlorine-Containing Organic Constituents in Chlorinated Effluents. ]WPCF 47 (3) :601-618. 23. Koivusalo M, Jaakkola J J K, Tuomisto J (1994) Drinking Water Mutagenicity and Gastrointestinal and Urinary Tract Cancers: An Ecological Study in Finland. Americanjnl of Public Health 84 (8) : 1223. 24. Leong LY C (1983) Removal and Inactivation of Viruses by Treatment Processes for Potable Water and Wastewater - A Review. Water Science & Technology. 15: 91. 25. Mader B A, Regli S (1993) Use of Microbial Risk Assessment in Setting United States Drinking Water Standards. Intl jnl of Food Microbiology. 18 (4) :245-256. 26. National Academy of Sciences (1983) Risk

Assessment in the Federal Government: Managing the Process. National Academy Press, Washington, DC.

27. Ongerth J E (1989) Giardia Cyst Concentrations in River Water. jA WWA (September): 81-86. 28. Pontius F W {1993) D-DBP Rule to Set Tight Standards. jAWWA. 85 (11) :22-3 0. 29. Regli S, Rose] B, Haas C N, Gerba C {1991) Modelling Risk for Pathogens in Drinking Water.jAWWA. 83 (11) :76-84. 30. Rose J B {1990) Pathogens in Water: Overview of Methods, Application Limitations and Data Interpretation, pp. 223-234. In Craun G F {ed), Methods far the

investigation and prevention of waterborne disease outbreaks. Health Effects Research Laboratory, USEPA Cincinnati. 3 1. Rose] B, Gerba C P, Jakubowski W {199 1) Survey of Potable Water Suppli es for Cryptosporidium and Giardia. Environmental Science and Technology. 25: 1393-1400. 32. Rose J B, Haas C N, Regli S {199 1) Risk Assessment and the Control of Waterborne Giardiasis. Americanjnl of Public Health. 81: 709-13. 33. Rosebury A M, Burmaster D E {1992) Log-Normal Distributions for Water Intake by Children and Adults. Risk Analysis. 12: 99. 34. Ruffle B, Burmaster D E, Anderson P D, Gordon HD (1994) Lognormal Distributions for Fish Consumption by the General US Population. Risk Analysis. (4) :395-404. 35. Safe Drinking Water Committee. {1989)

Drinking Water and Health, vol 9: Selected Issues in Risk Assessment. National Research Council, Washington DC.s 36. Seyfried P L, Tobin R S, Brown N E, Ness P F {1985) A Prosprctive Study of SwimmingRelated Illness: I Swimming- Associated Health Risk. Amencanjnl ofPublic Health. 75: 1068-70. 37. Seyfried PL, Tobin RS, Brown N E, Ness P F (1985) A Prospective Study of Swimming-Related Illn ess: II. Morbidity and the Microbiological Quality of Water. American jnl of Public Health. 75: 1071-75. 38. Silbergeld EK (1993) Risk Assessment -The Perspective and Experience of. United States Environmentalists. Environmental Health Perspectives. 101 (2) :100-104. 39. Singer P C {1994) Control of Disinfection By-Products in Drinking Water. jnl of Environmental Engineering. 120 (4) :727-744. 40. Ward R L, Bernstein D L, Young E C, Sherwood J R, Knowlton D R, Schiff G M (1986) Human Rotavirus Studies in Volunteers: Determination of Infectious Dose and Serological Response to Infection. jnl ofInfectious Diseases. 154 (5) _:871. 41. Water Pollution Control Federation: Disinfection Committee (1984) Wastewater Disinfection: A State-of-the-Art Report. Water Pollution Control Federation, Alexandria, VA.

Author Charles (Chuck) Haas is Betz Professor of Environmental Engineering, Drexel University, Philadelphia, USA . He is an engineering graduate who in the early 1980s, as an Assistant Professor at the Illinois Institute of Technology, pioneered the concepts of risk assessment. He is responsible, along with Profs. Gerba and Rose, for the quantitative risk assessment methods used by the US-EPA in establishing drinking water guidelines. WATER NOVEMBER/DECEMBER 1995

WATER QUALITY

EPIDEMIOLOGY: THE SAFETY OF AUSTRALIAN WATER SUPPLIES R M Douglas*, L S Pilotto Abstract The new Cooperative Research Centre for Water Quality and Treatment brings epidemiology, risk assessment and water treatment technology into a partnership that will attempt to ~olve some of the difficult problems facmg water suppliers and regulators. The National Centre for Epidemiology and Population Health at the Australian National University is one partner in the CRC and is working on a selection of the problems in an effort to better specify the nature and potential avoidability of health risks to water consumers. !his ~aper outlines research designs to mvestigate the possible impact of water quality on gastro-intestinal disease, Alzheimer's disease, and cancer. The ~ealth consequences of cyanobactenal blooms and those associated with small rural supplies, particularly for isolated Aboriginal communities are also discussed .

Key Words Water quality, health risks, microorganisms, aluminium, Alzheimer's disease ' cancer, DBPs, cyanobacteria.

Introduction . E~idemiology is the study of the distnbut10n and determinants of disease in human populations. It involves a search f?r causal links between hypothesised nsk factors and disease entities, and uses a limited range of study designs, the strength and weaknesses of which are bri_efly . dis~ussed in this paper. Epidem10logists need to be both rigorous . and pragmatic, structuring their studies m ways which take advantage of an~ natural differences in exposures which may be available (Hennekens and Buring, 1987). Modem epidemiology began with John Snow, a physician in London in the mid 19th century who, by carefully structilring his observations of cholera deaths in relation to the origins of drinking water, was able to infer that cholera was being transmitt~d through the agency of sewage-contammated water. This was well before the germ theory of disease was propounded. Snow's stildies were simple and elegant. His brand of epidemiology is often referred to as "shoe leather epidemiology". He trudged around London counting 22

cases and deaths from cholera and relating them to the part of the Thames from which various houses and localities drew their water. Snow's analysis was a test of an association between an exposure (contarnmated water) and a disease outcome (cholera), (Richardson, 1936). Most modem day epidemiologists wo~k with four basic stildy designs in their efforts to reach conclusions about the causes of illness. Depending on circumstances, they choose between cross-

sectional surveys, retrospective case control studies, prospective cohort studies and prospective randomised control trials. This list of stildy designs is arranged in ascending order of informativeness, difficulty and cost, and examples of each are given below. Stronger inference is possible from randomised control trials, but there are often cost, logistical and ethical constraints on the conduct of such stildies, and the investigator must therefore oft~n choose fro:m one of the other srudy designs._ The ultimate objective of analytic ~pidem10logic stildies is to explore the link between a disease entity and an "exposure" or "risk factor" that has been hypothesised to be involved in its cause. Relations ips between exposures and disease are expressed in terms of "relative risks" and "odds ratios". The two measures are derived in different ways but are g_enerally interpreted similarly. A relative nsk or odds ratio of three means that individuals exposed to a hypothesised risk or exposure factor are estimated from the stildy to have three times higher likelihood of becoming ill, than those who are not exposed to that risk or exposure factor. . Sometimes epidemiologists find spurious associations between exposure and disease because of the effects of other factors which cause disease and happen t? be correlated with the hypothesised nsk factor. Such factors which are called " con1oun C d ers " may b e asymmetrically distributed in the exposed and unexposed populations, thus contributing to a false interpretation of the causal link. To avoid spurious inference, it is important that information on hypothesised and known confounders is collected and controlled for in the analysis. The surest way to ensure that confounders are neutralised is to randomly assign subjects to exposed or not exposed. This mcreases the likelihood that known and . unknown confounders will be symmetrically distributed. But of course ethically

and logistically, it is often not possible r~ndomly to ass~ people to have specific exposures-and that is usually true of drinking water sources.

Gastrointestinal Disease A Randomised Control Trial Minor gastrointestinal illness is common in Australia though serious infections such as typhoid and cholera are very rare. Infections caused by viruses, bacteria and parasites such as giardia and cryptosporidia are not uncommon, but we do not know what the contribution of water supplies is to their incidence. Many of them can be equally transmitted through food and most go untreated and undiagno se d because they are self limiting. A serious effort to discover how much gastrointestinal disease is attributable to water and whether it varies greatly in different water jurisdictions is a significant undertaking. Most Australians tolerate a few days of gastrointestinal upset without reporting it to the doctor or to health authorities. And when they do , it is unlikely that it will be investigated, or its causal origins identified unless there happens to be a major epidemic under way. We need Australian stildies which can estimate the contribution of water to the prevalence of minor endemic gastrointestinal infections. One of the best ways to elucidate this would be to fit real and sham reverse osmosis filters to a number of homes on a random basis. A recent study in Canada by Payment et al (1991) suggested that the addition of filtration to the domestic tap reduced gastrointestinal illness by as much as 30% in ~e fam~ly. This was in a water jurisdiction which had been considered to have "good" water! We need to do Australian studies of this issue and such studies are planned as part of the CRC's activity. Carrymg out such a study, however, is not a simple exercise. In this case, we would want to take a representative sample of households and randomly assign them to either receive a reverse osmosis filter on their tap, or an identically appearing piece of equipment that contained a sham filter. Ideally, neither the household members nor the study observers would know whether the kitchen tap was genuinely delivering filâ&#x20AC;˘ National Centre for Epidemiology and Population Health, ANU, Canberra ACT 0200.

WATER NOVEMBER/DECEMBER 1995

tered water or not. Household members would be asked to maintain diaries of their symptoms in the subsequent six or twelve months and some system would be instituted to ensure that they remained interested and involved in keeping the records accurately. Their diary records would provide the basis for estimating the incidence of disease in those who received filtered water and the incidence of disease in those who received unfiltered water. Assuming there was a statistically significant difference between the two groups, we would also presumably want to know which organisms were causing disease most frequently among the unfiltered water recipients. That would require a further level of complexity in which, when study participants got ill, they would automatically undergo a series of clinical investigations to assess the offending organism. It is apparent from this brief description of a possible randomised control trial that such studies are expensive and complex to run, though they provide data of high quality in making causal inference. The numbers of households required for such a study depends on the base line incidence of disease and the level of difference between filtered and unfiltered households which we would want to be able to detect if it was present.

Aluminium and Alzheimer's - A Prospective Cohort Study There is little objective evidence to support the proposition that aluminium in water might hasten the onset of Alzheimer's Disease in those predisposed to it, and there are many who believe it to be an implausible hypothesis. Nevertheless the possibility still causes widespread community concern and there are many who believe the association still needs to be rigorously explored. (Crapper et a~ 1991, Storey and Masters 1995, Doll, 1993). A strong epidemiologic study design is needed to test the proposition and the possibility of carrying out a cohort study comparing longterm residents of Brisbane (which has used alum as a flocculant for more than 60 years) with long-term residents of parts of Melbourne (where alum has never been used) is currently under consideration. The study would involve the use of identical methods of follow up over time of a group of older residents who had spent most of their lives in the one residential location. Because Alzheimer's Disease is difficult to diagnose and tends to appear most often in people over the age of 70, identical methods would be used in the two cities to recruit people at the age of 70 and follow them over 5 years to discover whether there was a greater rate of onset of Alzheimer's Disease in one group or the other. One of the difficulties with this design

WATER NOVEMBER/DECEMBER 1995

is that there may be other confounding factors that cause Alzheimer's Disease that are currently unknown, but asymmetrically clistributed in the two cities. For example, perhaps exposure to certain climatic conditions could be an unknown contributor to Alzheimer's Disease and of course that exposure could readily be correlated with residence in Brisbane or Melbourne. It has been suggested that the epidemiologic design of this cohort study could be strengthened by building into it a nested randomised control trial in which Brisbane residents who were recruited to the study would also be randomly assigned either to have a real filter to eliminate aluminium attached to their tap or a sham filter. Currently we estimate that a definitive cohort study of this issue would require several thousand participants followed carefully over five years. Such a study is only likely to be embarked upon by a large consortium with rigorous methodology and quality checks built in, to ensure that the important information that emerges from it can be generalised beyond Australia.

Disinfection Byproducts and Cancer - A Retrospective Case Control Study In case control studies the relationship between disease and exposure is explored retrospectively. A group of sequentially diagnosed patients with the disease of interest is asked an identical set of questions about past exposures to a group of randomly selected controls. On the face of it, this kind of study is more manageable and cheaper to conduc( but it is also highly susceptible not only to confounding but also to various kinds of bias. Bias is systematic error brought about by either the selection of cases or controls, or by the particular differences that may selectively affect cases or controls in terms of available information, or memory, or quality of information. Because of the difficulty surrounding case control studies, epidemiologists like to see several such studies arriving at the same conclusion before they trust that conclusion. A small number of case control studies have reported an association between chlorination of water supplies and the incidence of bladder cancer and possibly bowel cancer (Morris et al (1992). The evidence is not yet such as to satisfy the International Agency for Research on Cancer that chlorination byproducts are carcinogens and there have been methodological difficulties in most of the studies that have been carried out. Studies have not been done in Australia, though we have good cancer registries and good information about chlorination of water supplies and levels of trihalomethanes in the drinking water

of various parts of Australia. The difficult thing about plabning large case control studies relates to the way we estimate exposure to chlorinated water. We need to get patients with bladder cancer and appropriately selected community-based control patients, to tell us about their water consumption behaviour over their lifetime. At the same time, we would need to collect information both from patients and their controls about other potential confounders which will be controlled for in the analysis. We are proposing a major case control study on this topic in South Australia and New South Wales in the coming year. Obsessive attention to the methods of selection of patients and controls, and the methods of estimation of exposures, will ensure a study of high quality.

Recreational Exposure to Cyanobacteria - A Crosssectional Survey Approach Good surveys often provide valuable epidemiologic information about the relationship between an exposure and a health outcome. A problem facing water authorities around the nation is the safety or otherwise of various levels of infestation with cyanobacteria for people who swim in the water. During the summer of 1994/ 5 we undertook a multicentred survey in a number of sites in New South Wales, Victoria, and South Australia where recreational swimming, water skiing and boating was taking place and where we were able, at the same time, to standardise sampling and quantitation of cyanobacterial infestation. In each location, people who had water contact were compared with people who did not swim and a questionnaire was carried out on the day of the survey and repeated by telephone a few days later, to discover whether the subjects had experienced any symptoms. The results of this survey are still being analysed and will be reported on elsewhere. The cross-sectional survey method was particularly suitable for this problem, as it enables us to test the association between exposure and disease at a range of dose levels of exposure.

Rural Water Supplies and Epidemiology The studies described above are large in scale, expensive and complex. We also need to develop approaches to the assessment of risk that can be applied to small areas and water supplies which serve the needs of small population groups, such as outback Aboriginal populations. The poor health of Australian Aboriginal people is at last causing Australians real concern. Life expectancy is about 20 years less than that of white Australians, and disease patterns are different with serious problems of infections, heart disease, renal disease, diabetes, sub-

stance abuse, accidents and violence. Poverty, alienation and environmental deprivation appear to be major determinants, but the role of water in this sorry plight is currently unclear. Preliminary survey work has been done to identify issues relating to water quality (Hearn et al, 1993). It is not yet clear whether, or how, environmental interventions can be deployed in ways that can improve health outcomes. There are concerns that technical water quality "fixes" in small communities might have little genuine effect. The proper role of epidemiology in this situation is first to identify the questions and the opportunities that are available to answer them. That will involve broad consultations with Aboriginal communities and explo rations of key health and water parameters in communities which wish to advance this issue. A workshop is planned in the coming months to help to identify ways in which epidemiology can be used to improve Aboriginal health.

Conclusion Epidemiology is both an observational and an experimental science. It seeks to structure data collections in ways that can answer practical and public health policy questions. Further, it seeks to make use of natural experiments in order to provide information that can improve health. In that sense it is both pragmatic and opportunistic. But pragmatism is no excuse for sloppiness in design, measurement, or analysis. A good epidemiologic study must pass the test of rigour and scientific thoroughness. There are many modem challenges as we explore the safety and risk of exposure to water that is contaminated by a huge range of modem toxicants, albeit in low concentrations. The challenges are technical, scientific, technological, social and political. The multidisciplinary environment of the CRC is going to be a vital facilitator for that kind of research in the years that lie ahead.

References Crapper-McLachlan DR, Krock TP, Lukiw W, and Krishan S S.{1991) Would decreased aluminium ingestion reduce the incidence of Alzheimer's Disease? Canadian Medical Associationjournal 145 {7) : 793-801 Doll R {1993). Alzheimer's Disease and Environmental Aluminium. {1993) Age, Ageing, 22: 138-153. Hearn B, Henderson G, Houston S, Wade A, Walker B. {1993) Water supply and Aboriginal and Torres Strait Islander Health: an overview. AGSO Journal of Australian Geology & Geophysics 14 {2/3), 135-146, Hennekens CH, Buringj F (1987) 'Epidemiology in Medicine' Little Brown, Boston, USA. Morris RD, Ardel A, Angelillo IF, et al (1992). Chloramination, chlorination by-products and cancer: a meta analysis. Am]nl Public Health 82: 955-963. Payment P, Richardson L, Siemiakycki J, Dewar R, Edwards M, Franco E{l991). A randomised trial to evaluate the risk of gastrointestinal disease due to consumption of drinking water meeting current microbiological standards. Amjnl Public Health, 81 (6): 703-708. Richardson B W (1936) 'Snow on Cholera' ,

a reprint of two papers by John Snow MD together with a biographical memoir and an introduction by Wade Hampton Frost. Commonwealth Fund, New York. Storey E, Masters CL {1995) Amyloid, aluminium and the aetiology of Alzheimer's Disease. Med] Aust 163: 256-259.

Authors Professor Robert M (Bob) Douglas graduated from Adelaide in 1959 and worked far ten years in General Practice also undertaking research on hypertension in New Zealand and pneumonia in New Guinea. Later he gained a Masters degree in Population Studies from the University of Pennsylvania. .Returning to Adelaide his research interests concentrated mainly on respiratory infections, and the delivery of health care in developing countries, as a consultant far WHO. In 1988 he was el,ected Dean of the Faculty ofMedicine. In 1989 he was appointed Foundation Director of the National Centre for Epidemiology and Population Health at ANU, and is the convenor of the Australian Health and Water Research Consortium. Dr Louis Pilotto is the NCEPH's principal epidemiologist in relation to water and is a member of the Steering Committee of the Australian Health and Water Research Consortium. He has recently reviewed the association of disinfection by-products and cancer and is currently investigating the effects of exposure to blue-green algae. He holds a B Sc in Statistics, an Honours MBBS from Sydney and a Ph D from ANU

Guy Parker Award Pumping Sewage Downhill by Lindsay Mott of GHD Melbourne, has won the annual Guy Parker Award for the best paper published in Water. The paper was published in December 1994. This award commemorates the memory of Guy Parker, foundation chairman of the Journal Committee . He was responsible for the planning, management and development of Water from its inception in 197 4 until his death in 1981. Significance, originality, content and presentation are the criteria used to make the choice . It carries a prize of $400 . The selection of the best paper published in Water in any year (this year for the period June 1994 to April 1995) is made by a panel. This is comprised of the journal's editor and four committee members nominated by the chairman of the Journal Committee . For further information about submitting technical papers, please contact Features Editor, Bob Swinton, telephone/fax: (03) 9560 4752 or A WWA Federal Office, telephone (02) 413 1288 or fax (02) 413 1047.

,,-FILTRATION .MEDIA

SAND,GRAVEL,COAL,GARNET . MANGANESEGREENSAND Invest in accurately graded, durable media from your complete filter media professionals.

U'ea,iefdea4d,t(J.o/l«,tpUe: • Decades of mineral processing experience. • Media produced to the Al/I/WA 8100-89 Standard. • An extensive product range at competitive prices. • Manufacturing in accordance with Quality System AS 3902/ISO 9002 • Packaging alternatives to suit any requirement. • Proficient technical assistance and support. • Prompt delivery Australia-wide and Overseas.

So, tap into our extensive experience and helpful service when next you require filter media.

---~-RIVER SANDSt'.iti 683 Beenlelgh-Redland Bay Road, Carbrook QI~ 4130

INTERNATIONAL: +61 7 3287 6444

~ E E CALL: 008 on 744

FAX: (07) 3287 6445

WATER NOVEMBER/DECEMBER 1995

MANAGEMENT

ORGANISATIONAL RESTRUCTURING OF WATER AUTHORITIES BA Nosworthy* Abstract

Restructuring Options

This paper focuses on issues arising from moves to bring about structural reforms in the water industry. The governments of Australia are committed to achieving improvements in the equitable, efficient and effective delivery and management of water services in the country. In pursuing these objectives, they have sought to build on experience elsewhere to derive a feasible set of options capable of delivering the reforms envisaged. The reform process reflects a general trend to organisational reform of the public sector in Australia and elsewhere. The options represent a new decision environment for water resources management. Reforms to improve efficiency and effectiveness are welcome, but caution is needed. Accelerated change in any area of public service has the potential to compromise the very levels of service delivery towards which it is targeted. In any reform process, it is important to ensure that the core of technological and managerial expertise, which is encompassed in water resources, and valuable to the functioning of the water systems, is not undermined.

A number of options are available whereby the organisational structure of public water utilities may be modified to reflect a more commercial perspective. Commercialisation is only one term which can be used to identify a particular position along a continuum of arrangements for the provision of water services in a market-orientated way. In the same way, the term 'privatisation' is often used to denote. a polar position along this continuum, marked by significant diminution of government control, if not ownership, of water undertakings.Yet, used in a generic sense, privatisation can convey some or all of the following connotations (Musgrave, 1992): • Termination of public programs and disengagement of the state from particular activities • Privatisation of funding through usercharges • Market deregulation or liberalisation • Contracting out of services once performed by public agencies • Sale of publicly owned assets such as state;-owned enterprises, land, or capital items. Undoubtedly, real scope for competition does exist in the water sector: eg. in the market for water rights, in engineering and structural services, in the provision of core services spanning the borders of neighbouring authorities, and by franchising, outsourcing, or privatising particular activities, or indeed, whole services. Given the monopoly situation that exists, any degree of 'privatisation' will require an operating licence that specifies how charges are to be regulated, the need for customer liaison, and a prohibition on creating entry barriers to markets. These conditions, together with trade practices legislation, should create a framework which ensures that corporations are not in any position which will result in the abuse of monopoly powers.

Introduction Dramatic changes are taking place in the Australian water industry: changes with far-reaching implications for resource allocation, pricing and infrastructure management. A feature of this new organisational scene is a move towards commercialism and corporatisation, and perhaps, privatisation. The impetus for reform can be linked, in part, to a switch in emphasis from development to management of water resources. Advances in technology, too, accompanied by fluctuations in economic growth, changes in social values and priorities, and emerging environmental concerns, have prompted the search for more responsive organisational structures and institutional frameworks. In the new social, economic and political environment, efficiency and accountability are seen of prime importance. At the same time, important questions need to be addressed regarding the justification for change, the benefits and risks associated with private sector involvement in the provision of water supply and services, and the safeguards which may need to be. built in to any privatised arrangements. 26

Rationale Pressures on budgets and growing demands on resources encourage politicians to welcome proposals which promise to reduce such demands, yet can be defended to the electorate. Privatisation can be attractive because it offers a way of shedding loss-making government trading enterprises. Ace-

ording to Damberger and Piggott (1986) the benefits 9f the public provision of tradeable goods are offset by three considerations. These are: • The absence of clear and simple economic principles for public production policies • The absence of competitive market pressures with resulting inefficiencies of operation • The irresistible temptation to governments to use public production to serve non-commercial objectives. Damberger and Piggott (1986) point out, however, that private enterprise can also have its drawbacks in the pursuit of productive efficiency. Incentives can be inappropriate, and competitive pressures distorted or absent. Also, the original justification of public involvement should not be ignored. While privatisation must be accompanied by deregulation or liberalisation with respect to product and capital markets, if the move is to be successful, so too some degree of regulation of the privatised entity maybe necessary if the social costs of market failures are to be contained. Whether one form of organisation is to be preferred to the other is ultimately an empirical question.

Benefits of Privatisation Some of the gains of privatisation arguably flow from the creation of an environment for managers free from political interference, and with reduced bargaining power for unions. However, many of the benefits of privatisation may possibly be obtained by moving to a situation short of it, such as commercialisation.To the extent that this is true, careful consideration needs to be given to endorsing privatisation in any particular situation as the move is, typically, not costless and a satisfactory level of benefits may be obtained by the less extreme step of commercialisation. Thus, the same level of net social benefits of such a move might be obtained by corporatising all or 1part of a public water utility, or perhaps by merely commercialising certain parts of its operations.The following general principles are prerequisite to any reform leading to privatisation. • University of New England, Armidale, NSW 2351

WATER NOVEMBER/DECEMBER 1995

Clear Objectives Enterprises cannot be expected to operate efficiently when their objectives are either unclear or conflicting. Problems in this regard arise usually as a result of different traditional roles being somehow combined or blurred by governments thrortgh their Government Trading Enterprises (GTEs), i.e as owner; as regulator; and as a provider of services. Clarifying the objectives of public administrative units is an important goal of government. A major corporatisation or privatisation exercise would have the added benefit of separating those responsibilities which are essentially government or regulatory from commercial services, thus minimising the opportunity for eternal conflict of interest between the two. The provider of services would no longer exert regulatory control over its own service.

Managerial Autonomy Clear accountability for achieving objectives cannot be attained unless the Board of Directors and management of a GTE have authority and autonomy to implement the decisions required to obtain efficient outcomes. Increased managerial authority, autonomy and responsibility require a reduction in detailed direct external controls, and a commensurate increase in accountability for performance.

Enhanced Protection Government moves to tighten up legislation covering environmental issues reflect a concern that environmental standards be established to protect natural resources, water being significant among them. It is desirable that the links between environmental agencies, the planning process, and the management of natural resources be as direct and as clear as possible. To this end, it is preferable that natural resource management be concentrated in fewer, rather than more units, and those units have comprehensive coverage of management in their areas of responsibility. A robust water management role is required. There is concern that any break-up of public utilities will result in considerable losses in the integration, quality and range of environmental .services. Moreover, any concentration on commercial performance may result in failure to deliver services for which no market exists in monetary services.

Performance Monitoring Managerial autonomy and responsibility should not be increased without the development of rigorous, independent, external monitoring and assessment to ensure accountability for performance. GTEs would not be subject to takeovers or exposed to scrutiny by the share market and the have less risk of WATER NOVEMBER/DECEMBER 1995

bankruptcy. It is therefore necessary for government to develop performance monitoring instruments that enable replication of the scrutiny to which private enterprises are subject to by existing market mechanisms. The quality and timeliness of information flows are essential for monitoring agents to assess performance against agreed objectives and targets, and to ensure that a GTE properly addresses its accountability obligations. Monitoring should encompass accounting, economic and non-financial mechanisms.

Rewards and Sanctions A rigorous structure of independent monitoring requires a corresponding system of incentives which encourage and reward good performance, and a system of sanctions ranging from reprimands to dismissal in the event of less than acceptable performance. A system that replicates as far as possible market-based rewards and sanctions would apply to enterprise performance.

Competitive Neutrality The concept of competitive neutrality relates to the provision of a level regulatory playing field relative to competing enterprises in the private sector. By virtue of government ownership, GTEs are characterised by a combination of advantages and disadvantages relative to the private sector. When a GTE enjoys special advantages in either the cost of the resources, its uses or the price it can receive for services by virtue of its ownership, resources may be diverted from efficient competitors or from more highly-valued end users . Ideally the disadvantages and advantages of public ownership would be identified and effectively removed.

State Initiatives Clearly, the pace of reform in the Australian water industry has stepped up in the past decade, marked by a series of policy initiatives by both Federal and State Governments. The deliberations of the Working Group on Water Resource Policy, and the joint Water Policy Agreement endorsed by the Council of Australian Governments (COAG) in February 1994, represent important milestones towards establishing a strategic framework for ongoing reform (Pigram, 1994). Meeting in Hobart, the Council considered a range of measures designed to improve the efficiency of sectors of the Australian economy. Among these were proposals to sustainable water use and management, including institutional and organisational reforms in the water industry. The COAG Agreement demonstrates renewed interest by governments in water reform and has been matched by a number of initiatives at the State level. It has since been reinforced by COAG

endorsement in April 1995 of the Natural Competition Policy and the reforms embodied in the Hilmer Report. The states, too, have been active in the process of reform. In Western Australia, a Strategic Direction Plan has been drawn up, incorporating microeconomic reform of the Water Authority, and the endorsement of Total Quality Management to provide greater accountability and improved operational performance. Corporatisation of the Authority is under consideration and a number of policy options are also being pursued, including further tariff reforms for metropolitan and non-urban water users, and possible privatisation of the irrigation distribution system. Maintenance and design functions are being outsourced, among others. In South Australia, responsibility for water management is now split, with the water resources regulatory function transferred to the Department of Environment and Natural Resources (DENR) . The role of the DENR is to plan and regulate the use of the state's resources for all purposes. The management of utilities, services and construction activities will lie with the South Australian Water Corporation (the new name for the Engineering and Water Supply Department), which has been corporatised. Many of its functions in operations and maintenance will be outsourced. The increasing commercial orientation of the existing agency has generated profitable operations already. The concentration of the state's economic activity and population around Adelaide precludes decentralisation of management. However, the clear aim is to generate increased economic returns by the public sector in partnership with private components of the water industry. The South Australian Water Corporation is seen as being a success, taking into account the following: â&#x20AC;˘ Better services and better value for money â&#x20AC;˘ Economic activity in South Australia is being generated and entry has been made into new markets - local and overseas â&#x20AC;˘ It has become a profit making agency going from a loss to a profit in recent months. In Victoria, key components of a reform agenda are being pursued, aimed at introducing competition into the water industry (Department of the Treasury, 1993). The Government's policy for reforming the water industry was set out in the October 1993 document, 'Reforming Victoria's Water Industry A Competitive Future'. This statement identified two key changes - amalgamations and commercial management as necessary components of reform of the non-metropolitan water sector. The Water Act of 1989 brought together many of the recommendations of the Public Bodies Review Committee which initiated progressive restructuring 27

of the water industry. Further reforms followed the McDonald Review of the Rural Water Commission in 1992. As a result, the Rural Water Corporation has become four separate rural water authorities with additional responsibility for headworks operations. Other functions have been sold to private companies. In the urban water sector, Melbourne Water Corporation has been set up as a commercial organisation with a wholesale and three regional companies. The contracting-out program will continue. Responsibility for provision of public goods such as parks and waterways is a separate authority. Non-metropolitan urban water authorities have been amalgamated into a smaller number of commercially viable organisations, which are expected to generate operational savings through economies of scale. The objective has been to establish well-defined service companies with separate functions. The government role is to overview broad policies and public interest issues. The Office of Water Reform which had oversight of many of these developments was restructured into the Water Bureau fromJuly 1995. The Bureau will focus on recent and recurrent responsibilities emanating from legislation and restructuring, and continuation of the government's reform program. The Water Resource Management Branch will remain with Conservation and Land Management and continue to focus on sustainable water resource management issues including catchment and river management. New South Wales has also been engaged in similar reforms in recent years. Policy initiatives included pioneering work in Total Catchment Management, restructuring of urban and rural water authorities, nutrient and salinity management strategies, corporatisation of major urban water authorities, and moves towards a more flexible, market-driven water allocation system. In a major reform the previous State Government passed legislation to enable the transfer of management and ownership or rural irrigation schemes from the Department of Water Resources to irrigation interests. Under this legislation, government-owned corporations are being set up to manage and operate the schemes as an interim step towards full autonomy and irrigator ownership. lrrigators will be expected to establish a commercial business structure incorporating managerial arrangements for financial responsibility and accountability {Pigram et al 1994). Although Irrigation Management Boards expressed support for the transfer, the move has implications for ongoing responsibility for asset mainte28

nance. Negotiations will be necessary regarding continuing Government support for a capital refurbishment program to ensure that water supply delivery and drainage systems are upgraded to acceptable standards. Further initiatives culminated in the release in June 1994 of a White Paper which outlined sweeping reforms to the management of the state's rivers and waterbodies (New South Wales Government, 1994). Key points in the new structure were: â&#x20AC;˘ Establishment of a Catchment Assessment Commission to make recommendations to the Government on water quality objectives and uses of water for each catchment in the state â&#x20AC;˘ The establishment of a Council of Water supported by an Office of Water (in place of the existing Water Resources Council) with a broad advisory and stewardship role directed towards policy co-ordination between land and water management. These arrangements were seen as complementing the responsibilities in water management of agencies such as the Department of Water Resources and the Environmental Protection Authority. Although the proposals received a mixed reception from irrigators and environmental groups, the reforms were expected to contribute to the adoption of realistic, achievable water quality objectives and workable approaches to the sharing of water (including environmental allocations), specific to individual catchments and river systems. They were also expected to provide the necessary conditions for the establishment of secure ancl tradeable property rights in water (Pigram et al 1994). The White Paper aroused considerable discussion and expressions of concern over the imposed changes. On the positive side, the proposals did raise the level of Cabinet responsibility for setting environmental standards for water and provided logical planning targets in water resource management, environmental protection and land use management (Pigram et al 1994). Irrigators were a major group affected by the proposed changes and there was concern among them over the possible effect the new Catchment Assessment Commission would have on water availability, reliability and tradeability. There was also disquiet over the possible removal of responsibility for water management from the Department of Water Resources to the proposed Office of Water. Some saw this latter body as merely imposing another regulatory level of bureaucracy on the water industry with lines of responsibility and reporting becoming more confused and complex (Pigram et al 1994). .Although the Sydney Water Board {now corporatised as Sydney Water)

expressed support for the White Paper and the Catchment .Aissessment Commission, this contrasted with many other water using groups. Scepticism was expressed about the workability of the Commission and the magnitude of its task in drawing up specific Catchment Management Plans for the 46 separate catchments in the State (Pigram et al 1994). The proposed changes were quickly brought into focus by the election of a Labor government in New South Wales. In March 1995, Labor's Water Protection Plan issued prior to the election envisaged the abolition of the Catchment Assessment Commission and the Office of Water. This now has occurred and the Department of Water Resources has become part of the new Department of Land and Water Conservation. The responsibilities of the proposed Catchment Assessment Commission have largely been taken over by the Environmental Protection Authority, while the Office of Water has been subsumed under a newly created Office of Natural Resources. Despite these changes, it remains to be seen to what extent a new political perspective translates into a different resource management regime. Integration is now clearly the watchword: integration of water issues with other resource issues. Whereas integration makes a lot of sense, it could mean that water will lose its unique place and importance in the priorities of governments. It also means that water research and development must now compete for a share of. a common funding pool for resources research.

International Experience Forms of privatisation of water infrastructure and commercialisation of water services have operated for many years in several industrialised countries of the western world, notably France, the United Kingdom, New Zealand and the USA. These examples illustrate the general trend in the direction of privatisation on a global scale. Despite the differences between countries, international experience in water reform can provide a number of useful insights in developing the 'best' structural, competitive and regulatory arrangements for a more efficient water industry. The impetus for change in the United Kingdom and New Zealand is clearly important in the achievement of major efficiency gains over one or two years. The key ingredient to increased effectiveness and to ongoing improvements in efficiency has been the separation of the service delivery task and the government task. This is particularly the case in the United Kingdom. Other major organisational themes which are common in the United Kingdom, France and the United States are: â&#x20AC;˘ The separation of regulation policy from the service delivery task, strong external WATER NOVEMBER/DECEMBER 1995

scrutiny and regulation of formally contracted performance, providing a driving imperative for the organisation. • A clear separation within each organisation of corporate-wide responsibilities and local level responsibilities • The delegation of authority and responsibility in most organisations which is apparently consistent with the task required of the participants • A relatively lean workforce. The major gains from these organisational themes are that the task of water management has been simplified, properly regulated and feedback on performance is more immediate and clear. France. A world leader in research and development of water and waste water technology, France has attracted substantial investment into the industry by franchising both construction and operation of schemes to private companies. Although municipal governments ultimately 'own' most of the services, the private sector often manages all aspects of operations and customer service. The present French system involves the whole administrative machine: State, Regions, Departments, Communes; the users; the elected representatives at the basin level, the Basin Committees; and the mainstays of the organisation: the Water Boards (Agences de !'Eau). France offers the most striking example of widespread reliance on private water corporations. Municipal water departments retain, ownership of the infrastructure for storage and reticulation, and franchise their management to private companies. The degree of private sector involvement, and the balance of obligations between the municipality and the private company, depend upon the system which is in operation - concession agreement; leasing arrangement or Affermage; or management contract, eg. meter readings (Paterson, 1986). Recent developments include the passing of the 1992 Water Act which introduced compliance with EC wastewater directive, increased the power of enforcement of water police, and instituted a new polluter-pays tax to be levied on the farming community.The French system is unique in its flexibility and professionalism. Private f:rench corporations have expanded internationally, acquiring interests in water companies in England, other parts of Europe, Southeast Asia, the United States and Australia. The French experience with privatisation is clearly of great relevance to urban water authorities in Australia. United Kingdom. The most noteworthy example of the privatisation trend can be found in the share flotations of public enterprises undertaken by the Thatcher Government in Britain. WATER NOVEMBER/DECEMBER 1995

The costs of the corporatisation and The restructure and privatisation of privatisation program have been in water supply have now been in place since 1989. Despite considerable debate terms of their social, regional, and about its success the basic framework employment impact (Boston,1988). The appears sound. It has clearly separated process of divesting the Government of responsibility for commercial services responsibility for operations and mainfrom the government's role in environ- tenance of community irrigation mental oversight, water management and schemes in New Zealand began in 1987 consumer protection. A significant aim of with a review of existing schemes the British Government, which has been (Lewthwaite and Martin, 1987) . The achieved, was to attract sufficient invest- review concluded that Crown ownerment into the water industries' ageing ship should continue (Park, 1988), it infrastructure to meet the new and strinwould be practicable to transfer ownergent environmental requirements of the ship of community irrigation schemes to European community. the local irrigator groups, and that irriA major criticism of the program has gators would receive first ownership . been its increasing emphasis on denationThe New Zealand experience is alisation at the expense of the promotion instructive in that it details the process of competition, a stated objective of the required to facilitate the transfer of irrigaprogram (Gittins, 1988). According to Gittins, the two major natural monopo- tion schemes to the private sector, and illustrates the compromises which may be lies, British Telecom and British Gas, necessary between economic and politiwere denationalised with virtually no attempt to increase competition. A prima- cal consequences. USA. The responsibility for water ry reason for this was fear that the share management lies with the individual values of the floated companies would be diminished by deregulation and restruc- states. Federal funding for rural areas is turing prior to sale. mobilised through the US Army Corps The initial electoral enthusiasm for of Engineers. The USA has shown privatisation has diminished and a poll movement towards the use of economic conducted in 1989 revealed a 79 per cent instruments in the form of the auction of disapproval rating for the pi anned flota- permits in some states but does not, tion of the water authorities (Markus, as yet, have an integrated catchment Goodway and Renton, 1989). This was management approach to water resbased on concerns that privatisation ource management. would lead to immediate price rises for The water industry is a mixture of private water. This concern has proven valid. and public companies. The experience in Britain with priCompanies are mainly public monopovatisation of water is essentially urban- lies, being very similar to those operated orientated and would seem of most in Australia. Economic regulation is the relevance to urban water supply authori- responsibility of the State Commissions, which determine rate increases and tariff ties in Australia. Regional water supply structures of private utilities. Consumer authorities may also fit the model as interests are defended by the State regucould regional county councils for water distribution. The most important lesson, latory commissions, which exert pressure in keeping rates low. Federal however, is that while government may regulations governing water quality and divest itself of operational activities, it is necessary to retain control over regulato- the environment are set by the EPA Standards of service are regulated, with ry activities to ensure that operations act the force of law, by State regulatory in the public interest. New Zealand. A parallel devel- commissions. Australia does not seem to . have much to learn from the United opment has taken place in New States because of the nature of regulaZealand, where a new Natural Resource tion of water management. Act places all management of natural resources under a single framework, with a single set of development proce- Lessons From Restructuring dures. At the same time, mechanisms Experience overseas and in Australia are now available for local governments indicates that gains in efficiency and to commercialise or corporate their effectiveness can be realised in the water industry by embarking on well structured water supply service. The New Zealand Government has , reform processes. There is, however, no appeared to have learned from the mis- single recipe for success. When considertakes made by the British. This is evi- ing industry reforms that have taken place, the prevailing physical and finandenced by its firm commitment to cial situations, political objectives, and market deregulation and its decision not social doctrines with respect to water to privatise state-owned enterprises with need to be borne in mind. The way such monopoly and power until they have factors have impacted on reform processbeen restructured and · their statutory es has led to variations in the outcomes. privileges removed. 29

u.LC: E3

The University ; of New South 0 ~Wales School of Civil Engineering External Courses in Waste Management, and Water & Wastewater Treatment The School of Civil Engineering has been a leader in offering courses in waste management since 1986, and in water and wastewater treatment (public health engineering) since 1965. Both courses are now available to students in Australia and New Zealand, and in other selected countries, on an external studies fee paying basis. Subjects are available in the fields of environmental engineering science, solid waste, hazardous waste, wastewater treatment, water treatment, groundwater contamination and remediation, and management of engineering projects. Open learning style documentation is provided to guide students to a level of understanding that enables them to undertake 3 applied major assignments in each subject, and to participate in one week long residential courses at the UNSW campus each session. The formal Graduate Diploma course {8 subjects) and the Master of Engineering Science {7 subjects and a major project) can be completed in one year full time, or 2-3 years part time enrolment, and subjects may also be undertaken individually to meet continuing education needs. Applications for March 1996 commencement in these two courses are now being accepted, and should be submitted by 30 November. Late application may be considered, depending on the availability of places. The courses and subjects are fee paying, currently $1200 per subject, and do not incur HECS {graduate tax) liability. Fees may be tax deductible where the course is relevant to the student's employment. Intake into the course is limited to 20 in total. A detailed Information Package and Application Form is available from Ms Rosalie Pang, ph {02) 385 5072, fax {02) 385 6139, or postal address School of Civil Engineering, The University of NSW, Sydney 2052.

In general, however, there has been a trend towards separation of regulatory activities from service provision activities. The degree of service provision reforms has varied widely, from government authorities taking a more commercial approach to management, to full privatisation of non-regulatory activities. Where the private sector has been permitted to enter the industry, much attention has been paid to regulating, to capture potential efficiency benefits, and to protect the public from potential exploitation. This has meant that governments have had to exercise more rigor in rule setting and policing than may have been previously necessary. It has become fashionable to espouse the virtues of economic instruments as management tools and to highlight the deficiencies of direct regulation. There is no doubt direct regulation has its problems. Rigidities can be introduced into management systems which inhibit rapid, flexible responses to changing conditions; implementation and enforcement practices can restrict effectiveness; regulations do not necessarily produce the most costeffective solutions to specific problems; and direct regulation tends to be expensive, but with little income for the regulatory body. . It can best be concluded that a mixture of economic incentives and direct regulation needs to be employed. Regulation can facilitate the adoption of water saving measures by helping to ensure that water efficiency and effectiveness become key parameters in management of water resources.In the identification of management strategies for wat~ , four things must be ensured: • The sustainability of the water environment • Economic sustainability • Strong investment in supply enhancement • Social sustainability. Any reform of privatisation of the water industry chosen must initially satisfy these requirements and must continue to do so for the benefit of the public and the sustainability of the resource.

Author Bronwynn Nosworthy is a Lecturer in the Department of Accounting and Financial Management and also consultant to the Centre for Water Policy Research at the University of New Engl.and. Her research interests are public sector auditing, performance measurement and environmental indicators in water authorities.

Domberger S, Piggott] (1986) Privatisation policies and pul»lic enterprise: A survey. Economic Record.June, 62: 145-162. Gittins, R (1988) Privatisation: Thatcher's Contribution to Today's Political Economy. Sydruy Morning Herald, April 9: 38. Lewthwaite W, Martin G. (1987} Review of

Existing Community Irrigation Schemes in New Zealand with a View to the Future. Report for the Ministry of Works and Development, Wellington. Markus M, Goodway N, Renton] (1989} 'Water: Privatisation Crisis'. Observer. July 2: 51. ·Government (1994} White Paper: The Management

and Regulatio.n of Water in New South Wales. Office of the Premier, Sydney. Park J (1988} 'Irrigation - Legislation and Policy'. Proc.Conf.on Water and Society, Institution of Professional Engineers, New Zealand. University of Otago, Dunedin. Paterson J (1986) The Privatisation Issue: Water Utilities. In: Abelson, P. Privatisation: An Australian Perspective. Australian Professional Publications: Sydney, 181 -203. PigramJJ, Musgrave W, Hooper BP, Dudley N J, Bryant MJ (1994} Co-operative Federalism and Water Reform. Water 21, 4.

Further Reading Cheret I (1993), Managing water: the French model Proc. First Annual International Conference on Environmentally Sustainable Development, Washington, DC, USA, October. Hartley P, Trengove C (1986) Who benefits from public utilities ? Economic Record, June 62: 163-179. Kay J, Thopipson D (1986} Privatisation: A policy in search of a rationale, The Economic journal. 96, March: 18-32. Moore W, Arthur-Worsop M (1989} 'Privatising Water: An Analysis of Initiatives to Sell Community Irrigation Schemes and to Create Water Markets'. Conference of the New Zealand Branch of the Australian Agricultural Economics Society. Mulligan H K (1990) Private Sector Involvement in the Irrigation Industry. Centre for Water Policy Research, Occasional Paper No. 6, University of New England. Mulligan H K, Pigram J J (1989) Water

Administration in Australia. Agenda for Change. Centre for Water Policy Research, Occasional Paper No. 4. University of New England. Musgrave W (1992} 'Privatisation of Irrigation in New South Wales'. 36th Conference of the Australian Agricultural Economics Society, Canberra, February. Pigram J J, Delforce R, Coelli M, Norris V, Antony G, Anderson R, Musgrave W (1992}

Transferable Water Entitlements in Australia,

References Boston J (1988} 'From Corporatisation to Privatisation: Public Sector Reform in New Zealand'. Canberra Bulletin of Public Administration. December, 57: 71-86. Department of the Treasury, Victoria (1993},

Reforming Victoria s Water Industry.

Department of Water Resources, Victoria (1992}

A Scarce Resource.

Centre for Water Policy Research, University of New I;ngland. Pigram J J (1995) 'Water Reform- Rearranging The Deckchairs?' Water, 22, l. Porter M (1991} 'Agreeing on the level of regulation - post regulation'. Proc. Conf. Water Resources Reform 1991, Institute for International Research, Sydney.

WATER NOVEMBER/DECEMBER 1995

MANAGEMENT

CAN WATER TRADING ACHIEVE ENVIRONMENTAL GOALS? H Bjornlund,J McKay Abstract Recent federal attempts to push the issue of market forces in the water industry have put increased pressure on the State Governments to introduce various policies for tradeable water entitlements (TWE). Present knowledge of the actual impact of such policies is very limited. This paper analyses empirical data collected by mail questionnaire from all sellers and buyers of permanent water entitlements in the Goulburn-Murray Irrigation Area (GMIA) in Victoria. It finds that the policies have had some of the desired effects, but it also emphasises that the market alone cannot solve the environmental problems associated with irrigated farming, and that other instruments are also required.

Introduction The aim of tradeable water entitlements (TWE) is to enable market forces to re-allocate existing water resources and thereby solve the conflict between competing users and lead to a more sustainable and a financially more beneficial use for the community as a whole. It is also expected that this process will enable un-profitable and un-sustainable water users to exit the irrigation industry and the trade will provide these farmers with financial compensation. This process ideally would result in water moving out of the areas badly affected by environmental problems and into areas more suitable for irrigation. Stringer (1995) describes the conceptual framework of a water market, its potential for increased efficiency and profitability on a sustainable basis, as well as some future perspective of its use in Australia. The purpose of this paper is to highlight some of the important statements made by Stringer based on empirical data collected from the buyers and sellers of water in GMIA. This research concentrates on permanent transfers as these facilitate long-term structural change to the irrigation industry and represent a true re-allocation of Australia's water resources. Temporary transfers are important in their own right, but serve a different purpose. They are a tool to facilitate short-term fluctuations in demand and supply of water due to shortterm changes in the weather pattern or

WATER NOVEMBER/DECEMBER 1995

variation in market conditions for annually produced commodities. They do not facilitate long-term changes since most farm operators are unwilling to invest in long-term infrastructure changes without a long-term certainty of water supply.

Recent Federal Intervention The Council of Australian Governments {CoAG) Communique (1994) commits all Australian states to full cost recovery within the water industry by 2001 in the context of increased efficiency on a sustainable basis. The irrigation industry can only survive such conditions if the water is used to produce highest value crops in an efficient way. According to the Working Group on Water Resources Policy to the Council of Australian Governments (1994), the concept of tradeable water rights or entitlements is generally considered to be the best way to ensure that water resources are put to their most beneficial use. The Working Group stated that "there would appear to be scope to offset the impact of rising water prices by allowing farmers to sell, in part or in whole, their water entitlem nts and use the proceeds to re-structure their properties or leave the land altogether" (p. 18). The Second Report of the W orkir1g Group (1995) stressed the importance of establishing property right/ entitlement agreements at an early stage, since this holds the key to a range of reforms and changes in irrigated agriculture. Due to the long lead time before a return on investment is achieved in viticulture and horticulture, there needs to be a great deal of certainty and security about future water supplies (p.13). We should add that the size and long-term nature of investments in the dairy industry requires the same certainty (Pigram et a~ 1992). The seriousness of the issue is further highlighted in the CoAG Agreement (1995) (The Agreement) signed by all the States on April 11 1995. In this agreement the second payments of federal funds for the implementation of the National Competition Policy due in 1999-2000 are made dependent upon effective implementation of all CoAG agreements on the strategic framework for the effective and sustainable reform of the Australian water industry. The third payments, due in

2001 -2, will depend on the states having fully implemented, and continuing to fully observe, all CoAG agreements with regard to water.

Australian Research The Australian Bureau of Agricultural Research Economics (ABARE), in their study of irrigated farming in the Southern Murray-Darling Basin (Hall et al, 1993), used linear modelling to estimate the impact of two water price levels. One price level was proposed to recover operating costs and one to deliver a five percent return on invested capital. ABARE found that the demand for water was not likely to be greatly affected at any of these prices. However, farm gross margin is estimated to be reduced by 80/o and 260/o respectively. The study further estimated that although the rising prices will not significantly affect the demand for water, price increases are likely to result in substantial structural adjustments within the irrigation region, for example, consolidating farm properties into larger units. Pigram et al (1992) concentrated their study on dairy farmers. They expected that dairy farmers would be most likely to be active in the market for permanent water rights because the long-term nature of this production requires a high security of water supply. The dairy industry is especially exposed to fluctuations in available 'sales' water since the average dairy farmer uses 1600/o annually of their water right, and thus relies on the more uncertain 'sales' water. That this reliance on 'sales' water makes dairy farmers more exposed during droughts was confirmed during the 1982-83 drought. The study indicated that some financial benefiti would be realised as a result of permaÂˇ nent transfers, and stressed that the analyÂˇ sis underestimated the total benefit to thâ&#x201A;Ź region since it only looked at transfen between dairy farms and not from lesi valuable commodities than dairy farms.

GMIA Case Study The GMIA was selected for thf Victorian case study since it is the larges irrigation area in Australia and the area ir Faculty of Business and Management, University o South Australia, GPO Box 2471, Adelaide 5001

Victoria within which permanent transfers have been possible for the longest period of time. GMIA further represents the bulk of water trade in Victoria both in numbers and quantities. The aims of the research were to establish: • The reasons for selling and buying • How buyers and sellers perceive different farm problems (water table, water supply, water quality and soil degradation) • Water and land use of sellers and buyers • Changes to the geographical distribution of water.

Methods and Sample The research was based on a mail survey of all 299 permanent water transfers within GMIA (except the Murray Valley Irrigation Area) until June 1994, including both water right holders and private diverters. Useable responses were obtained from 188 buyers (63%) and 149 sellers (50%). In 117 transactions (39%), responses were received from both the seller and the buyer. This facilitates a comparison of the perception of both groups.

Results Why are people selling or buying water? Each seller and buyer was asked to rate the importance of a number of pre-specified reasons in their decision to sell or buy water. Each respondent rated each of the pre-specified reasons on a · scale of 1 to 7.

Further analysis of the sellers' and buyers' reasons for being active in the water market indicated that the TWE policies have some of the desired effects. It indicated that: • The sellers are releasing dormant 'sleeper' water into active production and offer added security for farmers with productions requiring long-term financial commitments. These are mainly the sheep and cattle farmers. • Farmers in financial distress are selling water. These are the farmers with the worst perceived farm problems and the producers of lower value commodities. The policy thus offers these farmers some financial compensation which helps them to re-establish an alternative life. • Irrigators are reducing their irrigated areas, stopping irrigating the worst affected parts of their farms. • There are very few dairy farmers among the sellers (Table 1). Those who do sell are to a large extent stopping irrigating altogether rather than reducing their irrigated area and they are greatly affected by financial problems. This indicates that transferable water entitlements have enabled these unprofitable dairy farmers to leave the industry and establish themselves in an alternative way. • The policies allow some farmers to sell their water and retire on the property. Looking at what proportion the farmers used of their annual allocation prior to

Figure 1. Reasons for Buying Water 70 60

50 40 30

20 10 0

I. Wanted to 2. Wanted to 3. Wanted to 4. Wanted to 5. Wanted to 6. Wanted to

increase application on existing irrigated area. increase the irrigated area of existing crop. start growing new irrigated crops. start their first irrigated farming enterprise. secure existing crops against future droughts. invest in water with the expectation of future increases in the value of water.

Figure 2. Reasons for Selling Water 50 45 40 35 30

25 20 15 10 5 0

2 I. Did not need the water. 2. Needed the money. 3. Wanted to stop irrigated farming. 4. Wanted to decrease the irrigated area 5. Wanted to retire.

transfer, it also becomes clear that the TWE policie~ are releasing sleeper water to more active use. The sellers used on average 35% of their annual water right whereas the buyers used on average 175%. This figure confirms the findings of Pigram et al (1992) of 160% water use among dairy farmers . It is evident that the buyers are relying heavily on annual sales water and have exposed themselves seriously in case of drought and cuts to annual 'sales' water. Farm problems. The farm problems specified in this survey were: water table level, poor waler quality, uncertain water supply and soil degradation.

Perception of farm problems by sellers. The results of the survey suggest that: • The sheep farmers are showing the most concern for all four categories of farm problems. This reflects that the sheep growers are farming the most marginal land. • The dairy farmers' main concerns are water table and water supply problems. This is hardly a surprise since the dairy farmers are in the most intensively irrigated areas and those with the highest water table and most supply congestion. • Among the three main groups, the cattle farmers are the growers who show the least concern for the four farm problems, except for water quality problems, where they come a very close number two to dairy farmers. • The horticultural farmers are only showing some concern for water quality and soil degradation problems. Perception offarm problems by buyers. As Table 1 shows, mainly dairy and cattle farmers are buying water. The results of the survey suggest that: • Of all the dairy farmers, 44% indicated some problems with the water table level, and of those, 28% said that the problems were severe. • Of all dairy farmers, 26% indicated that they had some problems with water supply, and of those, 18% said that the problems were severe. • Of all dairy farmers, 21 % said that they had some water quality problems, and of those, 19% said that the problems were severe. • Of all dairy farmers, 18% said that they had some problems with soil degradation, and of those, 29% said that the problems were severe. • Of all cattle farmers, 33% indicated some problems with the water table level, and of those, 18% said that the problems were severe. , The above findings among the buyers suggest reasons for some concern. As many as 44% of all dairy buyers are buying where they already have water table problems and 26% where they already have water supply problems. These purchases are likely to escaWATE R NOVEMBER/DECEMBER 1995

late the environmental problems. The water seems to aggregate in the already most intensively irrigated areas and is likely to cause an escalating increase in the water table level and increase the pressure on the water supply system in these areas.

Water and land use of sellers and buyers. An important issue was to examine the use to which the water was put before it was sold, compared to its proposed use after sale. To establish this fact the sellers were asked what they used the water for, and the buyers how they intended to use the water. Based on this, it was computed how big a percentage of the total irrigated area on each farm was used for each production and determined the predominant production for each seller and buyer. The results of this analysis are shown below in Table 1. Table 1 clearly indicates that the dairy farmers are the strongest buyers. The horticulturalists only come in as number three with 4.8% and are more strongly represented among the sellers. By comparing the transactions where responses have been obtained from both the sellers and the buyers, it is also clear that the main movement of water is out of sheep or cattle and into dairy. A total of 28% of all transfers are from sheep to dairy and 30% of all transfers are from cattle to dairy. The study area has a large number of irrigated properties which are used for what could be described as non-commercial farming purposes. These are hobby farms and properties used mainly for rural residential living. This research shows that 1,489 ML or 18% of the sold water is out of non-commercial farms while only 454 ML or 4% of the water purchased was bought into non -commercial farms. This indicates that TWE facilitates the movement of irrigation water from the increasing number of non-commercial farms to commercial farms .

Changes to the Geographical Distribution As mentioned by Stringer (1995) the water should move out of the badly affected areas such as the Kerang and Pyramid Hill areas and into the Rochester, Cohuna, Tongala and Table 1. Main production of buyers

and sellers Predominant Production Dairy Cattle Sheep Horticulture Other

Buyers

Sellers 0/o

71.5 19.9 1.1 4.8 2.7

9.2 46.6 29.0 7.6 7.6

WATER NOVEMBER/DECEMBER 1995

Table 2. Water sold and purchased

Irrigation Area

ML Sold

Shepparton Rodney Tong-ala Rochester Campaspe Cohuna Kerang Swan Hill Pyramid Hill Boort

2,365 1,417 613 493 0 592 1,627 2,166 4,817 718

Sold

ML Bought

%' Bought

ML Intra2 Regional

ML Net Import

ML Net Export

0.90 0.78 0.30 0.28 0 0.42 1.08 2.77 2.79 1.16

1,511 2,333 2,416 2,622 0 3,233 752 401 1,541 0

0.58 1.28 1.23 1.47 0 2.29 0.50 0.51 0.89 0

1,164 919 563 196 0 592 441 401 1,541 0

347 1,414 1,853 2,426 0 2,641 311 0 0 0

1,201 498 50 297 0 0 1,186 1,765 3,276 718

1. Percentage of the area's total allocation. . . . . . 2. Water moving within the irrigation area ie both the seller and the buyer are located withm that 1rngation area.

Shepparton Areas. The immediate impression is that this has happened. Further analysis is required to fully determine what goes on in the Shepparton area. Table 2 lists the transfers in terms of ML and as percentages of the area's total allocation. Among the sellers, the four areas selling most water are Pyramid Hill, Shepparton, Swan Hill and Kerang but since Shepparton has a very large allocation, in percentages the order becomes Pyramid Hill, Swan Hill, Boort and Kerang. Areas buying the most water measured in ML and in percentage of total allocation are Cohuna, Rochester, Tongala and Rodney. Allowing for transfers within each area (column 5, Table 2) the net exporters are Pyramid Hill, Swan Hill, Kerang and Shepparton., and the major importers are Cohuna, Rochester, Tongala and Rodney. The quantity and direction of water transfers are further explained in Figure 3. In this diagram both number of transfers and number of ML transferred are indicated. No transfers take place from the Cohuna, Kerang and Swan Hill areas and into the Shepparton, Rodney, Rochester and Tongala areas. This is because of the limitations specified in Schedule 6 of the Water (Permanent Transfers of Water Rights) (Amendment) Regulations 1994. Transfer in the other direction is possible but has obviously not been desirable, which further emphasises that these areas are those most affected by farm problems. During this period, transfers have also not been possible between private diverters and water right holders. This has been changed in 1995 so that now transfers are possible between water right holders and diversion licence holders. In total, 1.03% of all the allocated water within these areas was traded during the three year period from 1991 -1994 or an

average of 0.33% per year, as stated by Stringer (1995). Stringer also said that the low percentage could be because of the relÂˇ atively wet seasons and seemingly endlesi supply during these years. The experiencei in the first half of the 1994-95 irrigation yea.i clearly indicate this. During the first sb. months of that period, applications wen made to transfer 10,347 ML or 0.72% o total allocation. We can only guess what wil happen during the next irrigation year wher possible limitations on sales water will be introduced as' a result of the drought.

Conclusions

TQ.ese results confirm that the policie! are working in the desired direction Water seems to begin to move away frorr the farms with problems either with wate1 table level, water supply, water quality 01 soil degradation. It also seems to allo.1 farmers with low value commodities tc opt out of irrigation farming or retire. Tc a large extent it has released previous!) unproductive dormant 'sleeper' wate into active production. Water is moving out of the higl saline mixed farming areas in Keran1 and Pyramid Hill and into the higl value-producing areas, mainly daiq farming. It is probably here that nega tive effects can occur. This water move, into the areas which traditionally havt the major problems of high water tablt as well as water supply, thus forthe escalating these problems and congest ing the supply system. It was already apparent in the 1991-9, period that 65% of all purchasers of perma nent water indicated that it was crucial fo them to secure their existing crops ag~ future droughts. This indicates uncertaint in the irrigation community about access to future 'sales' water. If this uncertainty mate rialises next irrigation season, more irriga tors are likely to want to secure their crop 3:

cation policies need to provide environmental criteria ,ro direct the operation of the market tools.

Figure 3. Water Movements 1991-94 Goulburn-Murray Irrigation Area 1/ IO

References

IOCW Boort

0'<1 40 1/11 so,'176

0'2 Sheppa rlon

28 / 919

Roche al er

1,U

1'17/11

IV l 'ifl

41 / 1164 25/0

4/196 17/0

1052,t)

21\'t)

l?R<,tll

Pyra mid Hill

12 / 1541

'""

Priva le Div.

3 1/ 1643

Kerang

1185.K>

6/592

6/441

Noles : I. rigures .ihove lines or to 1he lefl of line~ a,e number or 1r;:insfe1s. 2. fi gu,es below lines .ind to the rig h1 of lines are Ml lransferred. 3. f igu,es in b<utes indiule inlra region movemet1h. Number of transfers -1nd ML.

4. Auows i11dic,,te the clirecti0'1 nf lfansfer. e .g. Rodne y: 28 internal ti.1nsfe,s moving 919 Ml , ]4 11.insfers moving !OSI Ml f10m Sheppar1on to Rodney ;md 8 u.insfers moving 322 Ml from Rodney to SheppMon.

with water rights instead of relying on uncertain 'sales' water. Policies related to permanent transferable water rights are very important tools in a future re-allocation of Australia's water resources. However, transfer policies have been shown to achieve some permanent re-allocations which might be economically justifiable but not necessarily appropriate from an

Swa n Hill

5/401

environmental perspective. As expressed in the CoAG Communique (1994) "that water be used to maximise its contribution to national income and welfare, within the social, physical and ecological constraints of catchments" (p 22) is the really hard issue to be dealt with. It is necessary to blend all of these and not just focus on structural adjustments in the irrigation industry. Existing transfer/ allo-

Council of Australian Governments (1995) Agreement to Implement the National Competition Policy and Related Reforms, Canberra, April 11. Council of Australian Governments (1994) Communique, Hobart, 25 February. Hall N, Poulter D and Curttoti R (1993) ABARE Model of Irrigation Fa:rtning in the Southern Murray-Darling Basin, ABARE Research Report No. 93n. Pigram J P, Delforce R J, Coelli ML, Norris V, Antony G, Anderson RLand Musgrave F (1992) Transferable Water Entitlements in Australia, Report to the Land and Water Resources Research and Development Corporation. Stringer D (1995) Water Markets and Trading Developments in Victoria, Water, 22, 1. pp 11 -14. Working Group on Water Resources Policy to the Council of Australian Governm ents (1994) Report, February. Working Group on Water Resources Policy to the Council of Australian Governments (1995): Second Report, February.

Authors Both authors are at the University of South Australia. Henning Bjornlund is a Bachelor of Business, and a research student (PhD) in the Faculty ofBusiness and Management. Dr Jennifer McKay is a Senior Lecturer in the School of Law in that Faculty. She is also a Ministerial Appointee to the South Australian Water Resources Counci(.

MONO PUMPS SLUDGE HANDLING The MonoÂŽ Wide Throat Pump assembly can pump dewatered sludges to 30%. Designed to handle highly viscous materials such as sludges, slurries, thick non-flowing pastes and de-watered raggy type products with viscosities up to 1,000,000 cP. The range is based on two and four stage pumps which can develop pressures up to 24 bar with capacities to 50 m3/h. Dry solids approaching 40% can be handled by the pumps when fitted with the integral bridge breakers.For further information contact your nearest MonoÂŽ office.

Branches: Melbourne : Sydney: Brisbane: Adelaide: Perth : Darwin : Townsville:

(03) 9580 5211 (02) 521 5611 (07) 350 4582 (08) 346 4242 (09) 479 4999 (089) 84 3099 (077) 216 102

Mono WATER NOVEMBER/DECEMBER 1995

ASSET MANAGEMENT

CHICHESTER TRUNK GRAVITY MAIN AN ASSET MANAGEMENT APPLICATION G Pyke, G Hales* Summary This paper outlines some of the main components involved in the preparation of the Asset Management Plan for the Chichester Trunk Gravity Main. Gutteridge Haskins & Davey Pty Ltd (GHD) incorporated condition assessments, development of decay curves and probability of failure curves in the preparation of an asset management plan. Analysis of treatment options considered the life extension of the pipeline, associated risk and failure costs and calculated resulting NPV savings. Future stages in the process will include the assessment of the optimal timing of the renewal strategy.

Keywords Asset management, pipeline life, pipeline rehabilitation.

Introduction GHD was commissioned by the Hunter Water Corporation (HWC) in September 1994 to prepare an asset management plan for the Chichester Trunk Gravity Main (CTGM). The key objectives were to determine: â&#x20AC;˘ The optimum strategy for managing Chichester Trunk Gravity Main assets from stakeholder driven, technical, financial and risk perspectives â&#x20AC;˘ An asset management process that will enable managers and custodians to manage their assets in perpetuity, meeting changing expectations of stakeholders and new technology. The Chichester Trunk Gravity Main is approximately 100 km long, running from the Chichester Dam in Dungog to supply water to the lower Hunter Valley and Newcastle. The main was constructed in 1923 using 900 mm diameter wood stave and lead-jointed (LJ) pipe steel locking bar pipe. Shortly after World War II all the wood stave pipe was replaced with 1350 mm diameter welded steel cement lined pipe. The pipeline now comprises 47 km of900 mm diameter (LJ) locking bar pipe and 48 km of 900 mm to 1350 mm diameter welded steel (WS) pipe. WATER NOVEMBER/DECEMBER 1995

The project team comprised representatives from HWC, GHD and C W Pope & Associates with the Water Authority of W estem Australia (WAWA) providing specialist review input. The project was implemented in stages involving an anecdotal survey, data collection and condition assessment, system analyses, analysis of options and development of an implementation plan and strategy guidelines.

Anecdotal and Assessment The majority of the pipeline is above ground with three sections running through tunnels and numerous minor sections being buried. In all, 959 segments of pipe, each approximately 100 m long, were inspected and condition assessed through an anecdotal survey together with their impact of failure and construction difficulty. Field testing by cement lining sampling, wall thickness measurements, water quality analysis and internal video inspection was undertaken at various locations to enhance the condition assessment. The physical attributes and condition of the pipeline were recorded in GHD's Water Supply asset management system WAT AR which enabled the identification of "Parent" groups (ie sections of pipeline with common characteristics). In addition to the pipeline, 57 major valves, 13 bridges, three tunnels and 100 km of access tracks were inspected and assessed.

Pipeline Deterioration Prediction and Criticality Deterioration/ decay curves indicate the way in which an asset both has already decayed and is expected to continue to decay over time. Development of such curves is a significant part of managing assets and being able to estimate the effective life and deterioration. The effective life is the life that an asset is likely to achieve based on its current condition, life to date and likely maintenance programme. Associated with the decay of an asset is the probability of failure. Probability of

failure curves relate to both condition and effective life expired, and enable the development of an assessment of future maintenance repair costs and indirect failure costs. Deterioration and Probability of Failure curves were produced for each "Parent" asset of the CTGM based on the condition assessments undertaken, historical data provided by HWC, and GHD's experience with similar type assets in other locations. It was recognised that these initial curves were only theoretical or "our best guess" and may vary considerably depending on function, maintenance, location etc . of each asset. What was important to realise was that the true shape of the deterioration curves can only be obtained by regular condition assessments. The itl.dividual pipeline segment condition scores were plotted against the decay curve for each Parent Asset to enable derivation of information relating to each Parent Asset (such as effective life, average condition score, impact of failure , etc.) to be estimated for use in the options analysis. An example of the output from "WATAR" is provided in Figure 1.

Risk Assessment Knowing the rate of decay and probability of failure, the issue that will separate similar assets is their different consequence of failure which in tum will enable the relative business risk cost to be identified for Parent assets. In general, some assets have catastrophic failure modes, requiring treatment before failure occurs. Other assets have less catastrophic failure modes and multiple failures are likely to occur before renewal can be economically justified. Most sections of water supply pipelines are usually associated with the latter. I It was important that the consequences of failure were determined for each parent or individual asset. By multiplying the consequences of failure with â&#x20AC;˘ GHD Pty Ltd PO Box 408, Wyong 2259

the probability of failure, the relative business risk cost exposure level represented by each individual asset was estimated. This in turn can be aggregated to indicate the entire business risk for all assets or asset categories. During the anecdotal survey, the ancillary impact of a pipeline failure for each pipeline segment was assessed and a preliminary estimate was made of the cost of a pipeline failure for each hazard rating. This cost consisted of: • Direct costs due to - Repair to pipeline - Loss of water - Pipeline shutdown - Supply water from alternative source (ie Tomago). • Indirect consequential cost due to: - Personal injury - Damage to private property - Effect on business customers - Contribution to HWC licence fine - Effect on public image/ relations.

Treatment Strategy Options and Evaluation The identification of the modes of failure and assessment of the causes of failure along the pipeline were required for formulation of treatment strategies. For the purposes of this analysis, a pipe failure was defined as: "Decay in a location along the pipe sufficient to cause a large loss of water that requires the pipeline to be shut down for the repair." The major modes of failure identified were as follows: • Decay of lead joint • Decay of locking bar joint • Decay of pipe barrel. The treatment strategy types, options and techniques identified for the evaluation of the pipeline are outlined in Table 1. The most appropriate treatment techniques for each treatment option were selected using a coarse filter approach which considered the advantages and disadvantages of the various techniques for repair or rehabilitation. Then the Optimised Renewal Decision Making (ORDM) process developed by GHD was used to determine the optimum treatment strategy.

Optimised Renewal Decision Making The main functions of optimised renewal decision making involve: • Identifying the failure modes • Identifying the consequence of failure • Identifying the options which are available to overcome pipe failures • Determining the relative benefits and costs associated with these options • Carrying out a benefits-costs analysis of all options • Adopting the most cost effective option in terms of the total business picture.

In comparing treatment options, ORDM considers the following: • Timing • Treatment cost • Reduced operations and maintenance costs • Improved level of service • Improved production • Risk exposure cost reduction. Inputs into the renewal decision making process are shown in Figure 2.

clearing around pipeline, maintenance of valves, treatmen of external corrosion and reinstatement of embankments and access roads. The implementation of the rehabilitation of vertical joint strategy would result in effective treatment of the CTGM and it has been estimated to provide the HWC with a potential monetary saving of over $37m under current practices, discounted at fours percent over 30 years. The main reason for this high savings i~ that the majority of current maintenance practices on the CTGM is related to poor condition of the vertical joints on the LJ pipeline sections. Once these joints are rehabilitated in accordance with the proposed treatment strategy, the yearly maintenance costs are estimated to be reduced by approximately 73 percent (equivalent to approximately $1.Sm) . With regard to the buried sections of pipeline, the number of pipe failures is currently very low. Replacement would not be required until a significant increase in the number of failures occurs. The estimated replacement date varied for each buried section, depending on the assessed condition, risk cost exposure, and opportunity cost of capital adopted. By the time replacement will be required, internal lining technology would have further

Optimum Treatment Strategy Based on the evaluation of treatment strategies the main recommendations for immediate (ie in 1995-96) implementation were as follows : • Rehabilitate vertical joints for aboveground sections of LJ pipeline sections. The technique involves the encasing of the existing lead joints with a new collar, welded to the pipe. This technique is shown in Figure 3, and has been successfully used by WAWA for similar pipelines. As the joints would now be rigid, additional pipe restraints are required to be installed. • Continue status quo maintenance practices for WS pipeline sections. This practice mainly involves repairs to pipe,

Table 1 Identified treatment strategy types, options and techniques ,Technlque1

Strategy Type

Deflnltlon

Treatrnant Option•

Do nothing

No maintenance except for repa ir of failures

The Asset Is allowed to fail

Maintenance

Maintenance ~ncluding repairs) which do not extend the life of an osset

Consideration of: • Current maintenance • (Status Quo) moditi e'd maintenance practices

• Reduction of activities ie.lifting & straightening

Rehabilitation

Work on an asset that extend s its original design life

Consideration of: • Rehabilitation of joints on ly • Lining renewal only • Combination of joints/ lining • External coating renewal

• • • • •

Replacement

Replace an asset with its modern engineeri ng equ ivale nt

Full replacement with welded steel pipe s

• Buried • Above ground

Non Asset

Improvement to operating conditions of the asset, but not to the asset

Includes management techniques for pipeline operation that will reduce or eliminate the cause of failure or reduce the impact of failure

• Pressure reduction • Water hammer reduction • Water quality

Condiion

External collar over joint Weld up joint Cut out joint Renew lining Renew external coating

P.o.F.

Oet111ioration Cwve

- - ; - - - - ~ - - - ~ - - - -:- - - - ; - - - ; - - - - ; - - - ; - - - - :- - - - ; - - - ; - - - - ~ - - - ; - - I

' I

800

700

600

500

300

200 80

100 100

100

110 120

130 140

:t Effective life

Figure 1 Distribution of condition scores for pipeline segments on decay and probability

offailure curves. Parent Asset: LBT (Tarro to Beresfiel,d) WATER NOVEMBER/DECEMBER 1995

improved and would most likely be a preferable alternative to replacement. The evaluation also showed that: • The general renewal of the external coating was both not viable for the LJ pipeline sections (mainly because the existing coating contained various levels nNANCIAL

TEC HNICAL

M AIN'lr.NANCE

MAINTENANCE

COSTS

HIST ORY

O PERATION AL

COSTS MANAGEMENT

COSTS

Asset Management Implementation Plan

LIFE EXIBNS ION

LEVEL OF SERVICE

The overall plan prepared included the previous mentioned strategies and further recommended work such as: • Further investigation of re-lining strategies • Further testing and assessment of cement lining condition • On-going pipeline monitoring • Further development of pipeline decay prediction and risk assessment • Further analysis and re-evaluation of strategies • Optimised maintenance practices

POTENTIAL BENEFITS

RIVETS SPIGOT

BARREL OF PIPE

of lead) and not required for the WS pipeline sections (as the existing coating was in gqod condition). • Renewal of cement lining is generally only viable when the existing cement lining had reached the end of its effective life or had fallen off in large amounts. Parent Assets possibly requiring this treatment in the near future were identified. • The development of internal liners, such as polythene, is progressing and may be a viable alternative to replacement of the pipe.

I tf---'------'----r-L+r --r-f½W/,___, wir

COLLAR

y-I

LEADED JOINT

U ~. .

BARRE L OF

TYPICAL EXISTING JOINT

t COLD-LEAD JOINT

GROUT FILLED CAVITY

PROPOSED REHABILITATION TECHNIQUE FOR VERTICAL JOINTS NEW COLLAR OVER EXISTING JOINT

Figure 3 j oint rehabilitation options ·

WATER NOVEMBER/DECEMBER 1995

Conclusion

The Hunter Water Corporation w.u considered well poised, and certainly wel committed, to further implementation o: asset management systems and practice! over the coming years. Advanced asset management offer: great potential to organisations such a: HWC in many areas of management especially in the day-to-day maintenanc1 and operation activities (to increase effi ciencies) and in longer term plannini functions (to increase effectiveness). However, implementation of the sys terns need to take place in a structure< and orderly manner and will require ful commitment and backing of all involved Throughout this proj ect, the followin. system procedures were established fo continued utilisation by HWC: • Strategy plan • Asset numbering system (incorporatin HWC's existing numbering system) • Condition assessment procedures • Deterioration prediction and probabilit of failure modelling procedures • Optimised renewal decision makin process.

Acknowledgements The authors thank Hunter Wat€ Corporation, C W Pope & Associates an Western Australia Water Authority fc their contribution to this project. Specific acknowledgment is made < both Brad Foot (HWC) for his reviev and provision of background informatio and also Roger Byrne's (GH Melbourne) important role with speciali asset management input into this proj e as project director.

Reports

WELDED DRY JOINT

- - NEW COLLAR

• Utilisation of asset management information syst~s.

Institute of Municipal Engineering Austral, (1994), National Asset Management Manu. Gutteridge Haskins & Davey Pty Ltd, (199 Chichester Trunk Gravity Main Asset Management Plan - Volumes 1-5 and Proji Summary and Overview .

Authors George Pyke, currently the Manager GHD's Central Coast Office at Wyong a Manager - Asset Management for NS'W, is civil engineer with 25 years experience Australia, Fiji and Indonesia. His pres1 interests are state and local government iss1 and advanced asset management systems a implementation. Geoff Hal.es ,is a senior civil engineer a asset management engineer at GHD's Cent Coast Office at Ji.Jong. He has had 12 ye, experience in NSW and the North, Territory. His current interests are in il.'. management of infrastructure, water suP, and sewerage engineering.

ENVIRONMENT

WESTERN AUSTRALIAN SEAGRASS TRANSPLANT EXPERIMENTS Cf Walker* Abstract The decline of seagrass meadows has drawn attention to the factors that influence their survival and development and the possibility of their restoration . A study of the effects of transplant size and type (diameter) on the survival and growth of Amphibolis and Posidonia transplants was conducted near Carnac Island, Western Australia, to assess their suitability for revegetation programs. Transplant units (plugs) consisted of the whole plant, including leaf blades, roots, rhizome and the surrounding intact sediment. Transplants were monitored between January 1993, to mid-April 1994, at three sites within a seagrass meadow and at two sites at an unvegetated area adjacent to the meadow. Different sized plugs of Amphibolis and Posidonia were collected and the factors influencing survival were assessed. While transplants experienced an immediate decline in their survival when they were disturbed, Amphibolis plugs were able to anchor at the transplant sites more successfully than Posidonia. Survival and growth was highest in transplants of 10 cm and 15 cm plugs than in 5 cm plugs. The effect of sediment stability on transplant survival was assessed by adding mesh to the two unvegetated sites. Amphibolis plugs fared well at those sites while Posidonia plugs were buried by the accreting sediments. The successful survival of plugs was influenced by the uprooting of the seagrass and the unit size of the transplant.

Key Words Seagrass transplants, seagrass meadows.

Introduction Seagrass systems have been the centre of much debate and research in recent years. There is concern that the area covered by seagrasses has decreased globally and these reductions have stimulated interest in their ecological importance. In Australia, it has been estimated that over 45,000 ha of seagrass meadows had been lost by 1992 (Walker and McComb, 1992). It is therefore necessary to determine methods for revegetation in an effort to restore and minimise further meadow loss and to maintain species diversity.

As seagrass beds generally have a coastal distribution, they are particularly prone to disturbance and stress which may be naturally occurring or human induced. Seagrass loss has been caused by disease, eutrophication and coastal development activities. Seagrass transplantation and restoration trials have been carried out since the 1920's (reviewed by Thorhaug, 1986; Kirkman, 1989; Walker, 1994) at sites in Australia, the United States, Canada and Europe. One fundamental problem which exists with seagrass transplantation trials is that many projects do not relate directly to Australian seagrass species. Various planting techniques and large scale restoration have been attempted around Australia (Larkum, 1976; West et a~ 1990), but have largely been unsuccessful and expensive (Walker and McComb, 1992). Restoration efforts have had some success in the United States (Thorhaug, 1986) and although the success of planting techniques varies among species, sites and geographical range, they provide a valuable source of information to Australian researchers. Unti recently, planting attempts in Western Australia had centred on determining environmental influences (Cambridge et a~ 1986) rather than restoration and have met with limited success. Hancock's (1992) study was among the first attempts in Western Australia to evaluate the dynamics of seagrass transplants in a particular habitat. He reported some success with Amphibolis transplants in Cockburn Sound, Western Australia and concluded that the overall survival of Amphibolis was due to the anchoring of its rhizome root system. Various anchoring methods have been used in an attempt to reduce the loss of planted seagrass units from animal disturbance (Merkel, 1990), sediment movement and current action (Lewis, 1987). While both the role played by seagrasses in the stability of sediments and the effects of sediment movement in determining seagrass survival have been documented (Patriquin, 1975; Birch and Birch, 1984), few attempts have been made to develop techniques for the stabilisation of sediment to promote seagrass persistence and growth. Posidonia and Amphibolis are endem. ic to Western Australia and are the

major meadow-forming species growing in semi-sheltered marine embayments between 25°S and 35°S (Cambridge, 1979). Previous studies conducted by Hancock (1992) and Nelson (1992) examined environmental parameters affecting planting trials in Cockburn Sound. As a continuation of their work, the present study examined the survival of three different sized plugs (5 cm, 10 cm and 15 cm diameter) of Posidonia sinuosa and Amphibolis griffithii to determine whether the species in question could be successfully transplanted and to assess the effect of sediment stabilisation on transplant persistence.

Experiment Study area. Carnac Island is located approximately eight kilometres south west of Fremantle on the western coast of Australia (Lat, 32°07' S, Long. 11 5° 40' E) and to the north of Cockburn Sound. Although the region is subject to strong currents and surges, the waters where t~ study sites were located are relatively sheltered. Five study sites were chosen east of Carnac Island (Figure 1). Three were within a seagrass meadow (Sites 1, 3 and 4) and two were in an unvegetated adjacent area (Sites 2 and 5). The five sites were selected to match the same water depth (4 m) and physical conditions (sediment type and shelter). At the control sites (Sites 1 and 4), seagrasses were extracted and replanted in the same place within the seagrass bed. At the unvegetated sites, transplants (at sub-sites 2A and 5A) were either surrounded with plastic mesh (Bora] Cyclone) or left unprotected (Sub-sites 2B and 5B). Transplants surrounded with mesh are referred to as 'meshed', while those in the field without any protection are referred to as 'non-meshed'. Material collection and planting. Plugs were extracted using 200 mm lengths of PVC piping with three different diameters; 5 cm, 10 cm and 15 cm. After extraction, piping bases were capped with plastic sewer lids to minimise rhizome and root disturbance. They were then carried to their transplant • Environmental Science Murdoch University

WA 6150

WATER NOVEMBER/DECEMBER 1995

sites, where holes were excavated to receive them. The PVC piping and the caps were removed when the plugs were · planted. Amphibolis griffithii and Posidonia sinuosa were collected and planted in January, 1993. Twenty, 5 and 10 cm plugs, and fourteen 15 cm plugs were planted at each of the sites. At Site 3, ten 10 cm Posidonia sinuosa plugs were extracted and transplanted into a recipient Heterozostera bed in June, 1993. All material collection and monitoring was performed using SCUBA. Plug transplants were monitored bimonthly for survival over a period of fifteen months from January 1993 to March 1994. Survival was measured in terms of how many transplant units remained at a site. Further growth parameters; underground biomass, leaf base number and leaf and cluster production and an alternative planting technique consisting of root-rhizome groups, were examined and described by Walker (1994) to assess successful transplants. Sediment movement and nutrient analysis were also monitored, however for the purpose of this paper, sediment and growth monitoring have not been included. The similarities within and between plug sizes were determined using an analysis of variance with a one factor, multi-comparison significance level of 95%, on the rates of growth for Amphibolis and Posidonia.

Results Posidonia plugs. The 15 cm plugs had the highest survival rates for 200 days after planting. Although losses had occurred at Site 2, there was no decline at Site 1 in plant numbers during this period (Figure 2). In the 15 cm plugs (Site 1), there was a reduction in survival rate of 0.45% d·' during the following 157 day period, giving a survival rate after one year of 70%. At the time of transplanting, 10 cm Posidonia (>lugs covered total areas of 7.85 m2 (Site 3), and 44.4 m 2 (Sites 1, 2A and 2B). By March 1994, the planted area cover of sites had fallen to 50% (Sites 1 and 3), while at Site 2, 220 days after planting, no 10 cm Posidonia plugs remained. Further, at Site 2, the highest declines had occurred at rates of 60 to 80% after 75 days (meshed) and 100 days (non-meshed). Although the 5 cm plugs at Site 1 had the lowest survival rates {<30%, 250 days after planting), the numbers at the site remained constant for a further 150 days. A similar trend was observed in the 10 cm plugs where survival declined over the first 250 days to between 50 and 70% and then remained unchanged over the succeeding time period {Figure 2) . Initially the reduction in survival rate of 10 cm Posidonia plugs at Site 3, mirrored WATER NOVE MBER/DECEMBER 1995

Owen Anchorage Camac Island * Site 2 . *.

: Site

: Sii:e4.

Seagrass meadow Figure 1 Site locations for the study ofPosidonia and Amphibolis transplants in Owen Anchorage, Western Australia. Sites 7, 3 and 4 were within a seagrass meadow and Sites 2 and 5 were in bare sediment adjacent to them.

the rates of 10 cm plugs at Site 1 (a rate of approximately 0.15% d·') (Figure 2b). However 200 days after planting, a further reduction in survival rate was observed both in this size (Site 3) and in 5 cm plugs after 325 days (a further reduction of 15%). Meshed transplants (Sub-site SA) tended to have the lowest survival rates, especially in the 5 cm and 10 cm plugs. Sediment built up over the mesh and consequently smothered Posidonia. The apparent rise in survival rates with time for 5 and 10 cm plugs at Site 1 (Figures 2 and 3), is due to the burial by sedim nt at some earlier samplings. Amphibolis plugs. As the plug size increased, the general trend was for survival rates to be higher. Meshed transplants had the highest overall survival (> 80%) while an increase in plug size from 10 to 15 cm, resulted in survival rates becoming progressively more similar at Site 4 and Sub-site SB. Survival rates for Amphibolis plugs also tended to be higher than those of Posidonia. When planted, individual plugs covered total areas of 47. 7 m2 (Sub-site SB) and 44.4 m2 (Site 4 and Sub-site SA). By January 1994, those which had not been planted correctly had either floated away or had died back. Those remaining at the sites were in a healthy state and had 86% (Sub-site SA) and >SO% (Site 1 and Subsite SB) of their original cover. The percent survival of Amphibolis plugs decreased over time (Figure 3). During the experimental period, meshed 5 cm plugs experienced reductions of up to 45%. Survival continued to decline in 5 cm plugs until 150 days, where rates tended to stabilise with only marginal losses {of no greater than 25%). Although survival rates of 5 cm plugs appeared to stabilise at Site 5, they tended to fare

poorly and showed substantially lower survival rates than plugs at Site 4. Mesh addition. By April 1993, sediment covered the mesh by depths between 15 cm and 19 cm at Site 2, and between 3 cm and 13 cm at Site 5.

Discussion In both J?osidonia and Amphibolis transplants, the decline observed at the control sites (with the exception of Posidonia 15 cm plugs at Site 1), suggested that•these species were immediately under stress when they were uprooted. The extraction of plugs resulted in damage to the seagrass root-rhizome system thus diminishing the amount of rhizome available for attachment to the substrate. Regrowth was not observed in the planted control areas. This suggests that once Amphibolis and Posidonia have been physically disturbed and uprooted, subsequent growth is impeded within the same meadows and may be an indication that the natural recovery of meadows by these species is slow. Few reports have been documented on the natural recovery of donor beds and the extent to which the damage of extraction of planting units has caused (Walker, 1994). Furthermore, in the context of the present study it indicates that there is a significant transplanting shock for transplants regardless of the size of planting unit, or the site at which it is planted. An increase in plug size resulted in increased survival rates of transplants. The exception was observed at Site 2 where an increase, in water movement and fluctuating sediment levels had resulted in the loss and burial of Posidonia transplants at this site. Persistence may have been due to the amount of root-rhizome contained within a plug unit, hence the ability of the plug

Figure 2 Percent survival of total Posidonia sinuosa plugs of differing diameters at Sites 1, 2 and 3.

5cm

10cm

15cm 100 1"'"""_ _ _ __

100

120

100

1a !

80 60

60 60

.I:

100

150

200

250

300

Time elapsed since plan1ing(day1)

350

400

Site!

-·-r·· Sub--si1c2B ------

-050

100

150

200

250

300

350

Sub-site 2A SitcJ

400

Timeclapscd sinccplanting(days)

to attach itself to the substrate was pro- The unvegetated regions (Sites 2 and 5) at Carnac Island were high energy regimes portional to the amount of intact rhizome (Nelson, 1992), with greater water movecontained within the plug. For example, ment and high levels of sediment accreit was assumed that as the area of 15 cm plugs was nine times that of 5 cm plugs, a tion and erosion compared with the nearby seagrass meadows (Walker, 1994). larger area of rhizome (of up to nine Consequently, transplants at Sites 2 and 5 times) would be collected and contained were either uprooted or buried. within 15 cm plugs than 5 cm plugs (and The survival of planted plugs was not similarly, four times the amount of rhiinfluenced by seasonal variations such as zome would be collected in 10 cm plugs the severe storm activity reported by the compared with 5 cm plugs). Bureau of Meteorology (Perth, WA), in The percent survival of Amphibolis May and September ·1993, Rather, sur10 cm and 15 cm plugs was similar vival was infiuenced by the up-rooting of between Site 4 and Sub-site 5B after a the seagrass and the effects of transplantation. Tliis was further evident in the simiperiod of 150 days (Figure 3) showing lar pattern of survival rates of the 10 cm that although survival was affected when Amphibolis was uprooted, the effects of Postdonia plugs which were planted at different times within an estaolished, stable disturbance are reduced as the size of the Heterozostera bed (Site 3) and within the transplant increases. Consequently, the seagrass meadow (Site 1). Although larger the transplant size, the greater its Posidonia plugs were unable to survive at potential for survival as it has more root the unstable, unvegetated Site 2, their and rhizome contained within the plug. persistence at Sites 1 and 3 indicated that Posidonia plugs can be transplanted to The addition of mesh at Sub-site 5A areas of moderate sediment stability. resulted in better survival of Amphibolis 10 Whereas the number of Amphibolis plugs cm and 15 cm plugs than at other sites and aided transplant persistence at Sub- at Site 4 had begun to stabilise indicating that tra splants had already adapted to site SA by stabilising the sediment, but the mobile sediments and turbulent water did not influence the rate of accretion or movement during these periods. erosion at that site (Walker, 1994). The ability of transplants to cope with The better overall survival of sediment movement and water regimes Amphibolis plugs compared with determined their survival at a site. This Posidonia plugs was due to the differ- was demonstrated by the loss or burial of ences in morphology of their root-rhi- Posidonia transplants from Site 2. zome systems (Tomlinson, 1974). Kirkman (1978) and Hancock (1992) both Posidonia has approximately 80% of its reported that large amounts of accreting biomass below ground; consisting of sediments caused plants to be buried. Therefore the planting of Posidonia in soft, thick rhizome with few root hairs, high energy regimes or areas prone to compared with only 20% below ground erosion (such as in depression areas that biomass of Amphibolis (Paling, pers. have been left by the extraction of plugs, comm.), which has a branched rhizome system (Kuo and McComb, 1989), capa- or boat moorings) is not a good manageble of more successfully attaching itself ment option. As Posidonia species have only been to the substrate (Kirkman, 1989; reported to inhabit sheltered areas where Hancock, 1992). As Posidonia were there is reduced water movement unable to suitably anchor at the unvege(Kirkman and Walker, 1989) and few tated site, transplants were either smothrecords of successful transplants for the ered by sediment or floated away. same species have been documented Kirkman (1989) has reported that (Cambridge, 1979), it is believed that Amphibolis seedlings are capable of Posidonia cannot be transplanted. anchoring themselves with grappling However, the comparison between Sites anchors, whereas there have been few, if 1 and 3 above, indicates otherwise. any, reports of Posidonia sinuosa successPosidonia would be best planted in stafully attaching itself to the substrate. ble environments, such as in areas that The energy regime at both seagrass have been recolonised by pioneer meadows and unvegetated sites had a species, such as Hetero,:,ostera tasmanica. considerable effect on transplant survival. · The reason behind this is that the area

I 00

I SO

200

250

300

350

400

Timcclapscdsinccplan, ing(days)

now recolonised by the pioneer species would have created a stable environment enhancing the persistence of Posidonia. Such was the case in this study. Alternative methods of stabilisation should be examined such as the use of small groins close to transplant sites, or differences in the size of mesh squares used around transplants to determine if they influence sedimentation processes. An alternative planting technique was examined anchoring Amphibolis multiple rhizome groups, washed free of sediment, 15 cm below the sediment. These proved to be unsuccessful as the rhizome groups worked themselves free and washed away within 100 days of planting (Walker, 1994).

Conclusion From this study, the following conclusions were drawn: • Seagrass can be transplanted but its survival is qependent upon the species, the ecological characteristics of the recipient site and the size of the planting unit • The greater the plug size, the higher the plant survival rate : Amphibolis plugs had higher survival rates than Posidonia plugs, and both species had higher growth rates at stable sites than at unstable, sandy areas. • Sediment stability was important in determining the persistence and survival of Posidonia at the unvegetated site. By artificially stabilising sediments with mesh, the survival and growth of Amphibolis can be enhanced at high energy sites. The major factors that have influenced the survival of seagrass transplants in Owen Anchorage were the species planted, the disturbance of natural seagrass meadows, their removal from that habitat, the suitability of the recipient site and the transplant's ability to adapt to the conditions of that site, and the type (Walker, 1994) and size of the planting unit utilised. The physical disturbance of habitat and the stability of sediments are also important in determining survival immediately following transplantation, with the performance of smaller units falling well below the levels attained by the larger units after planting.

Acknowledgements The author thanks Dr Eric Paling and Dr Richard Bell for critically reviewing WATER NOVEMBER/DECEMBER 1995

Figure 3 Percent survival of total Amphibolis Griffithii plugs of differing diameters at Sites 4 and 5

5cm

10cm 100

100

15cm 100

~ 60

4()

100

150

200

250

300

350

400

Time elapsed 1mcc plantmg!dll)S)

the manuscript. The author also acknowledges assistance provided by Cockburn Cement Limited, Wes tern Australia.

References Birch W R, Birch M, {1984). Succession and pattern of tropical intertidal seagrasses in Cockle Bay, Queensland, Australia: a decade of observations. Aquatic Botany 19: 343-367. Cambridge M L, {1979). Cockburn Sound Environmental Study: Technical Report on Seagrasses. Department of Conservation and Environm ent. Report No. 7. Cambridge M L, Chiffings A W, Brittan C, Moore Land McComb AJ , {1986). The loss of seagrasses in Cockburn Sound, WA II. Possible causes of decline. Aquatic Botany 24: 269-285. Han cock C, (1992). Seagrass Transplantation Trials and Factors Affecting Their Success. Hons. Thesis. Murdoch University, Perth, Western Australia: 1-123. Kirkman H, {1978). Decline of seagrass in northern areas of Moreton Bay, Queensland. Aquatic Botany 5: 63-76. Kirkman H , {1989). Restoration and creation of seagrass meadows with special emphasis on Western Australia. Environmental Protection Authority, Perth, Western Australia, Technical Series 30. Kirkman H , Walker D I, (1989) . Regional Studies-Western Australian Seagrasses, in AWD , Larkum, AJ, McComb, and S, Shepherd , {Eds), Biolo[fj of Seagrasses. Elsevier, Amsterdam. ppl 57- 181. KuoJ, McComb AJ, {1989). Seagrass Taxonomy, Structure and Development, in A WD Larkum, A J McComb, and S Shepherd, (Ed s), Biolo[fj of Seagrasses. Elseiver, Amsterdam . pp 6- 73. Larkum AW D, (1976). Ecology of Botany Bay. I. Growth of Posidonia australis (Brown) Hook. f. in Botany Bay and other Bays of the Sydney Basin. Australian journal of Marine and Freshwater Research 27: 11 7-127. Lewis RR, (1987). The restoration and creation of seagrass meadows in the Southeast United States. Pp 153-173 in M J Durako, R C Phillips, and R R Lewis, (Eds). Proceedings of

the Symposium on Subtropical-Tropical Seagrasses of the South &stem United States. Florida Marine Florida Bureau Florida

Research Publications, No 42. Department of Natural Resources of Marine Research St Petersburg, 209 pp.

WATER NOVEMBER/DECEMBER 1995

•

o+--,---,---,-,--,---~-~..,......,~

.......... / •• 50

100

150

200

250

)00

4()

--o--

,!!

0 +-,-,--.----.,--,-,--,---~-,-,--,---,-,

350

400

Timcdapsed1inccplan1in1,{days)

Merkel KW, (1990a). Eelgrass transplanting in South San Diego Bay, California. pp 28-42 in SW Echeverria R S H offman , and K W Merkel, (Eds). Proceedings of the California Eelgrass Symposium, Chula Vista, California, May 1988, 78 pp. Nelson M, {1992). Environmental Characteristics Affecting Revege tati on of Seagrasses in Owen Anchorage, Western Australia. Hons Thesis. Murdoch University, Western Australia Patriquin D G, {1975). "Migration" of blowouts in seagrass beds at Barbados and Carriacou, West Indies and its ecological and geological implications. Aquatic Botany l: 163-189. Thorhaug A, (1986). Review of seagrass restoration efforts. Ambio 15 (2): 11 0-1 I 7. Tomlinson PB, (1974). Vegetative morphology and meristem dependence - th e found ation of productivity in seagrasses. Aquaculture 4: 107-130.

Sub-sue 58 Sub-snc5A

o-1-.-~-~-~-~-~'- ------:--~ 0

100

150

200

250

300

]50

400

Time elapsed smcc plonung (days)

Walker C J , (1994). The effect of size and sediment stabilisation on the survival of seagrass transplants in Owen Anchorage, Western Australia. H ons Thesis. Murdoch University, Perth, Western Australia: 1-104. Walker D I, McComb A J , (1992). Seagrass Degradation in Australian Coastal Waters. Ma rine Pollution Bulletin 25 : 191-195. West R J , J acobs N E, Roberts D E, (1990). Experimental transplanting of seagrasses in Botany Bay, Australia. Marine Pollution Bulletin 21 : 198-203.

Author Catherine Walker compl,eted this research project for her Honours Degree in Environmental Science at Murdoch University. She continued there as & search Assistant and is now working for the Department ofAgriculture. Her paper won the Undergraduate Water Prey, awarded by the WA branch.

Specialist Consultants & Analysts • • • • • • • • • • •

S1tc4

Analysis of Water & Wastewater Analysis of Sludge Process Design Treatment Plant Audits & Operation Treatment Plant Operating Manuals Operator Training Microbiological Analysis of Water, Wastewater, Swimming Pools & Spas Environmental Impact Investigations Corrosion Investigations Analysis of Concrete, Soils, Sands , Gravel

Simmonds ~, & Bristow Pty. Ltd. '

!IJ

30 Shottery Street, Yeronga, Queensland , 4104 Telephone (07) 848 7699. Fax (07) 892 3345

ENVIRONMENT

ROLE OF SEDIMENTS IN PHOSPHORUS CYCLING IN AQUATIC SYSTEMS Report by W Maher*, I Lawrence, T Donnelly Introduction A joint workshop of the CRC for Freshwater Ecology and the CSIRO Division of Water Resources in 1994 brought together researchers investigating the role of bottom and suspended sediments in phosphorus cycling in aquatic systems. The workshop examined the following: • The hypotheses underlying the research presently being undertaken to elucidate the processes involved in phosphorus cycling • State of the art techniques available for undertaking experimental studies of phosphorus cycling in aquatic sediments, and • Use of models to translate scientific knowledge to managers.

Processes What emerged from the individual presentations is that many processes must be considered (hydrological, sedimentological, chemical, physical, microbiological) to understand the role of sediments in phosphorus cycling in aquatic systems. Sediments are multistate, dynamic and complex systems and the various processes are occurring on different time scales. However, at present there is a lack of a conceptual framework for organising and integrating research. Four important questions emerged: • How do we identify the critical system processes (ie the processes that largely determine the release or mobilisation of phosphorus in sediments?). • How do we integrate the emerging knowledge on individual processes into an ecosystem perspective? • What is the relevance of temperate region concepts and models to Australian ecosystems since Northem Hemisphere approaches (eg. Vollenweider's Model) are often not applicable in Australia? • Do Australian aquatic ecosystems differ significantly from other ecosystems . around the world? 42

A number of research specific questions emerged: broadly categorised as information required to increase our understanding of how nutrients are derived from catchments and cycled in aquatic environments.

Catchment Exports Discussion of the sources of P have tended to accept that the predominant sources are some combination of sewage, fertiliser, feed-lot effluent, detergent, etc., and that what is causing our algal bloom problems can be overcome by reducing one or more of these inputs. This axiom may be true in the more heavily populated and industrialised countries such as North America or Europe. In Australia, which is characterised by extensive (rather than intensive) agriculture, extreme climatic changes and generally turbid water systems, northern hemisphere approaches to our water problems may not be pplicable. However, an acceptance of this axiom has lead to catchment remedial work (and model development) which has focused on the fertilised parts of the landscape and assumed that fertilised topsoil moving into stream systems is causing problems of nutrient enrichment in our rivers and storages. Recent studies have provided examples where sources of P in a catchment other than fertilised topsoil caused eutrophication problems in waterways. As well, these studies tend to suggest that in rural Australia, natural climate-driven processes (affected by European land clearing) are generally the means of introducing the major loads of nutrients into our waterways. That is, the introduction of nutrients into our waterways, sufficient to cause algal blooms under the right environmental conditions, is a historical process. What has clearly changed in these water systems in more recent times has been the control on water flow (e.g. storages, weirs) and the potential for much more interaction between the sediments and the overlying water column.

Land use practices cause P-exports to rivers, but there is some evidence that the sediment erosion processes in Australia (e.g. the effect of gullies) are different from those in the northern hemisphere catchments. A review of available data shows that the mechanisms, magnitude and processes of P transport in catchments are poorly understood and at present there is a need for experimental studies leading to the development of a general catchment model of P sources and transport. A conceptual framework needs to be developed to investigate the mobilisation and transport of sediment and phosphorus from catchments. Information is required on where sediments come from and how they move and what are the effects of geomorphology (geo-chemistry, rainfall ~lope) and modification by land use and vegetation changes. The main question is how much of the P sources are natural or derived from human activities. It was acknowledged that general experience indicates that: • P-adsorption on particulate material can dominate P-transport • Internal recycling via the sediment/ water column interface is a major component of nutrient loading to water bodies • Microbial processes can be the principal mechanism of P-transformation in sediments and subsequent release • As sediments become anoxic, release of iron-bound P will occur • Soil and sediment transport processes are poorly understood. For example, little information is available on transport and sedimentation/remobilisation processes • Rainfall and runoff energy control mobilisation and transport of sediments from upland areas.

P-cycling in W,ater Systems It was generally agreed that the rate of phosphorus release from sediments determines algal growth in the absence • University of Canberra, PO Box l, Belconnen,

2616

WATER NOVEMBER/DECEMBER 1995

Evapo-transpiratlon Fertilizer

Rainfall

' '

' ' ,:

Grazing ............

>-i,;,i

'' Catchment export load

----------------------··z Erosion & desorption Figure 1 Catchment mobilisation, interception & export of sediment associated materials

of sources of bioavailable P from sewage or animal wastes. Over long time scales in Australian waters, most of the sedimentary phosphorus is bioavailable and the mass of phosphorus already accumulated in sediments is sufficient to maintain algal growth for years. It was argued that it was the loading of biologically available forms of carbon stimulating microbial processes in association with mixing (oxygen transfer) and pH regimes in water bodies that determined phosphorus release rates from sediments. If this hypothesis is correct, the critical nutrient pathway becomes the bioavailable carbon. Information is required on sedimentwater column exchange processes and kinetics, the major questions being: • How is material reworked through sediments ? • How is P-bioavailability modified? • What are the time scales of P-bioavailability? • What are the major forms of P? • What is the nature of transformation between phosphorus forms? • What is the relationship of the P-species to algal growth? • What is the role of microbial processes? • What is the role of other cycles (N, C, Fe and S)? • Does deoxygenation initiate the release of sediment-bound nutrients? • Is sulfate reduction in sediments a major cause of P-remobilisation from sediments? • What is the role of sediments in the wider ecosystem? Turbid shallow mesotrophic waters appear to have a high sediment/water column interdependence. • What factors limit algal growth in turbid waters (light, nutrients, flow regime)? • What are the temporal and spatial scales of processes? • What is the role of euphotic depth and mixing? WATER NOVEMBER/DECEMBER 1995

Techniques State of the art techniques are available for examining sediment transport processes and nutrient cycling in Australia but are not in widespread use. Tracing the sources of fine sediments (and P) and the movement of this fine sediment through landscapes can be achieved using various radionuclides, natural magnetic minerals and geochemistry. Surface soil tracers, 137Cs and 210Pb, compared to 226 Ra can allow conclusions to be drawn about the main erosion mechanisms which supply the sediments delivered to natural waters. For example, these techniques used in Chaffey Dam in NSW, indicated that 90% of sediment to the dam was derived from unfertilised soils from the tertiary basalt. Where accumulations of fine sedimlnt (and P) occur, it is important to know their age, and to establish a chronology for sequences of fine sediments. In this way we can relate the deposited sediments to an event in the catchment, or changes in a sequence of sediments to events which may include pre-European compared to postEuropean changes. Dating techniques now available within the CSIRO Division of Water Resources in Canberra include the use of 137Cs and 210Pb, "C and various luminescence dating methods. The use of phosphate-oxygen isotopes allow the tracing of very small amounts of phosphate in transported sediments. In situ automatic water quality monitoring instrumentation (eg QUATEL) allows information at almost any temporal scale to be collected. Event monitoring and information on episodic events can then be obtained. Physical fractionation and concentration of suspended particulate matter from natural water can be achieved using sieving, continuous flow centrifugation and/ or tangential flow filtration techniques. Gram

Groundwater discharge

quantity of materials for physical and gf chemical characterisation can be separat over the entire particulate (> 1µm) and c, loidal (<lµm) size range as well as a d solved fraction (<0.003µm). Some measurement problems st exist. Standard methods currently us for measuring phosphorus concentratio in natural waters provide little infom tion on bioavailability and estimating rate of supply to phytoplankton. The u of iron impregnated strips suspended water samples for a defined period pr vide a better estimate of bioavailat phosphorus. Many soil science bas, approaches to defining bioavailable P a still being used in many instances. these approaches the form of phosphor in sediments is inferred from sequent extraction techniques, in which phosph rus is lumped into operationally defin, pools based on its solubility in vario media eg; MgCl2 , HCl, NaOJ citrate/ dithionite/ bicarbonate. Modern chromatographic and sp€ troscopic tools (e.g. NMR) allow us define phosphorus pools based on stru tural information. The use of NMR h shown that in some systems, eg bill bongs, up to 80% of phosphorus appea to be associated with organic matter.

Use of Models by Management

Research in this area may be driv( by techniques, knowledge or by the qUE tions asked by management. Even if v have the knowledge, the questic remains: what part of the system can I effectively managed ? It was agreedI that there is a role t decision support systems but they ne( to be based on the dominant process{ They must also be built on hydrodynam models as flow/ discharge is a maj, component that drives ecosystems.

Discharge of carbon & P absorbed onto suspended solids

Transfer of oxygen to water column & sediment

Figure 2 Phosphorus rycling in aquatic ecosystems

Forming models is difficult because of the complexity of systems, ie: • Chemical, biological, hydrological interactions • Heterogenous nature of systems {air, water, soil, biota) • Processes operating at different times and spatial scales • Lack of data to develop and test models. It is probably not possible to construct generic environmental models which include all processes in all regions. Constructing models which are transferable within a region or context where there are common system features, seems a reasonable alternative. Models must be developed from the managers' perspective rather than from the modellers'. Because of the scarcity of data in catchments (especially of episodic events) accurate predictions may not be an important model requirement for management but showing the links between management actions and possible outcomes may be. There is a danger that model predictions may be accepted as accurate when they are not, nor ever can be. During the workshop, two conceptual models were developed that focused on the export of nutrients from catchments (Figure 1) and the cycling of phosphorus in the water column (Figure 2). The export of nutrients from catchments involved identifying major sources of phosphorus {natural, fertiliser use, grazing), pathways of phosphorus movement {runoff, erosion, infiltration, groundwater), cycling within the catchment {sedimentation, uptake by vegetation) and processes causing P to move on and off particles (adsorption, desorption). P-cycling within the water column requires the identification of the physical, chemical and biological processes and the role of bioavailable-C and 02 transfer in Prelease from sediments.

Conclusions From the presentations in this workshop it became apparent that the control of sediment and hence non-point source addition of phosphorus to a water body, requires an understanding of catchment sources and transport, so the history of sediments will probably help our understanding of systems and their management. If stimulation of microbial processes by bioavailable carbon and oxygen trans-

BIOTECHNOLOGY RESEARCH CENTRE

fer (mixing) become accepted as the major controls on phosphorus release from sediments, management needs to focus on the reduction of bioavailable carbon and management of flow/ mixing and associated oxygen transfer. Consequently, knowledge (and research) needs to be focussed on these processes. A summary of the individual presentations can be made available from the CRC Freshwater Ecology, University of Canberra.

~ LATROBE -.

UNIVERSITY, BENDIGO

A RESIDENTIAL COURSE IN MICROSCOPIC ~ IDENTIFICATION OF FILAMENTOUS BACTERIA IN ACTIVATED SLUDGE Monday, February 5th to Thursday, February 8th, 1996. The Staff of the Biotechnology Research Centre present the popular filament identification course for those wishing to develop skills in monitoring and identification of the filamentous bacteria which cause bulking and foaming in activated sludge systems. The course consists of lectures, seminars, workshops, an excursion and intensive laboratory sessions using filament identification procedures. The workshops cover the latest research in the control of activated sludge bulking and foaming. Operators, scientists, engineers and managers involved in wastewater treatment will benefit by attending from this internationally recognised course.

Reeistratjons close January 29th. 1996 Cost ( including meals and extensive notes, but not accommodation) $1500

Please contact Mrs Beth Seviour at: Biotechnology Research Centre La Trobe University, Bendigo, P.O. Box 199, Bendigo, 3550, Victoria, Australia, Tel. +61 54 447222 or Fax +61 54 447 777

WATER NOVEMBER/DECEMBER 1995

BOOKS

INDUSTRY NEWS Murrumbidgee On Line Junee Shire NSW will upgrade sewage treatment facilities through 'beneficial reuse' irrigation before the end of the year under a $2m proposal recently adopted by council. Junee Shire is one of ten councils throughout the Murrumbidgee catchment which have banded together to form the Murrumbidgee Water Action Plan, a program aimed at improving the quality of waterways throughout the region. Features of the plan are: • The proposed Junee 'off-stream' solution would see more than 500 megalitres of treated water a year used to irrigate agricultural pastures adjacent to the sewage treatment plant, recreational land and council parks • W agga W agga is irrigating more than half a dozen parks and reserves with effluent reuse and is well on its way to meeting a pre-determined sewage treatment objective of being completely 'offstream' by the year 2000. Wagga Wagga is also in the process of establishing a trial residential treated water reuse project, which would see the lawns and gardens of residential areas irrigated with recycled water • Wastewater from sewage treatment plants makes up only half of the nutrients urban communities contribute to the river system each year. To combat these other nutrient sources, the Murrumbidgee Water Action Plan partners are actively involved in a community education program which includes advertising, promotional activities and a NSW Department of School Educationapproved schools program. The six key messages being promoted to local communities are: • Do use phosphorus free detergents • Do wash the car on the lawn • Do bury pet droppings • Do not over fertilise lawns and gardens • Do not sweep waste and organic matter into drains • Do not wash part loads. The ten member councils involved in the Murrumbidgee Water Action Plan are Cootamundra, Griffith, Gundagai, Hay, Junee, Leeton, Murrumbidgee, Narrandera, Tumut and Wagga Wagga.

lnnogaz Electrofusion The Innogaz electrofusion system for jointing medium and high density polyethylene pipe is now available from Milnes Pty Ltd. For potable water supply, WATER NOVEMBER/DECEMBER 1995

the system allows jointing in pressure pipes up to 315 mm in diameter, as well as for domestic supply systems. It is toxicologically harmless. The Innogaz system has been specifically designed to eliminate the possibility of human error at operator level. Because it is simple to use, even in remote areas, the fully automated system saves operation time and requires only a limited training period for operators. A range of control boxes makes the Innogaz system suitable for a variety of applications, from simple manual jointing up to the most sophisticated automated systems. The system has a full range of electrofusion couplings, tapping tees, repair saddles, stop bag saddles, 45/ 90 degree elbows, tees, reducers and caps. All fittings are available in sizes from 20 mm to 315 mm, and are designed for maximum working pressures of 4 bar for gas, and 12.5 bar for water, at 20 degrees C.

Teemburra Dam Award CMPS&F Environmental were highly commended in the annual Engineering Excellence Awards for their impact assessment and environmental management plan for the Teemburra Dam in North Queensland. The awards are presented by the Institution of Engineers Australia. Pat O 'Dwyer (CMPS&F regional manager) said the project was the first da and irrigation project in Queensland to be designed and built to the principles of ecologically sustainable development. The dam will release stored water principally for irrigation, but also for industrial development and town water. Water may be released directly down Teemburra Creek, a tributary of the Pioneer River, as well as through a tunnel into adjacent creeks to enhance distribution effectiveness. The environmental management plan covers design, construction and operation of the dam and irrigation scheme including water management strategies, waste minimisation, best practice farm and irrigation management and environmental flows . The plan proposes revegetation of fauna corridors along creeks using selected rainforest species and possibly tropical fruit trees. The water management strategy for , the project includes re-use of Mackay sewage effluent and industrial wastewaters and increased irrigation efficiency which may alleviate the need for a further dam in the future.

Pesticides in Drinking Wate DI Gustafson 1993 ISBN O442 011 3 241pp $119.95 Van Nostran Reinholt

This book provides a general sumrr ry of the state of useful knowledge on p, ticides in the soil-water environme1 Comprehensive lists of references, dra ing on experience between 1982 a1 1991 are provided. It is noted that in the author's vie pesticides are seldom a "contaminant" levels high enough to render the wal "unsafe to drink". The scope of the book covers: • Case studies of contamination, mainly groundwater in the USA • Assessment of surveys for specific pe~ cides in the USA, UK and Euw groundwaters and wells. Some surfa water surveys are also covered. • Causes of contamination coveri1 actual cases of intentional and accide ta! spills. The potential point sourc associated with water contaminati, are summarised. • Theory and results of modelling of p, ticide movement through soil-wal transport. {,Jseful rules of thumb are p1 vided which permit assessment of t behaviour of a particular pesticide in range of soils. Pesticide selection can nc include assessment of risk of water conmination using empirical and simulati, modelling tools. • Regulatory control, how it works a1 how limits are developed for drinki1 water. The author covers how field stu ies are conducted and the resu applied. Only USA, Canada and t EEC are covered. • Ways to minimise or avoid rii General details are provided on ways optimise use , alternative microbiologic pesticides and careful selection of cro to avoid pests. The author provides a very useful 1 of relevant physical properties of ov 300 pesticides, with trade names. There is a lack of detail on diffic1 areas such as how to mount a suital monitoring program in terms of detecti, limits, numbers of samples required to · 95% confident, but much of this is cc ered in reference material. It is easy to read and would be a m ful reference , for the practisi: engineer/scientist who occasionally has do a project in this area. The comm, sense rules of thumb and discussion past events and comprehensive ref1 ences are the book's strong points.

MChapman