Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
Murray-Darling Basin: ecologically sustainable irrigation?


Major Issues

Glossary of Terms


The Water Resource

Surface Water


Principles of Irrigation

Reducing Groundwater Recharge

Irrigation Waste Water Disposal

Irrigation Water Use

Managing the Water Resource

Managing Water Supply

Managing Water Quality

Managing the River Environment

The Irrigation Industry Dilemma

The Government Dilemma

Government Initiatives

Determining Irrigation Policy

The Present Approach

Possible Future Approaches


Major Issues

The Murray- Darling Basin (MDB) is a major contributor to Australian agricultural production, with an annual output of $10- 12 billion annually. Of this, the irrigation industry contributes about $4.5 billion.

About one third of the 1.6 million ha irrigated land has groundwater tables within 2m of the soil surface and 150,000 ha is salinised. These figures could well double within the next 30 to 40 years unless effective solutions are implemented immediately.

Off- site effects of the irrigation industry include saline groundwater discharge into adjacent lowlands and surface streams with the result that a number of terrestrial and aquatic systems are being damaged or destroyed. Waste water containing agricultural chemicals, such as nitrogen and phosphorus, are contributing to algal blooms in waterways and pesticide and herbicide damage may also be implicated in the future. Groundwater discharge will, in the future, affect practically all irrigation areas in the MDB because, unlike coastal catchments, groundwater cannot drain freely to the sea.

The dilemma for the irrigation industry is that it is virtually impossible not to use more water than is actually required by the crops. The excess water adds to, or recharges, the underlying groundwater systems which then mobilise and transport soluble salts back into the surface environment. It is unlikely that improved management practices will be sufficient to overcome a continuing problem of groundwater recharge and subsequent discharge.

The dilemma for the Commonwealth and State governments is that, having embraced the principles of Ecologically Sustainable Development (ESD), they are now committed to putting those principles into practice. To date the approach with respect to the irrigation industry has been:

. to provide information and training in better land management practices; and

. to devise economic incentives to encourage a more efficient use of water.

The information and training exercise is proving to be of limited success as not all farmers can afford the suggested measures. The economic measures have yet to be fully implemented and are mainly based on the premise that users should pay a realistic price for water and that they should pay for any pollution generated. The only 'carrot' offered has been the introduction of tradeable water licences, but even this measure is only likely to shift the problems to new areas.

The uncertainty generated by a new environmental awareness on the part of governments is not helping to produce a sensible solution. Irrigators are denying evidence of their involvement in environmental damage and are in no mood to negotiate on increased water charges. Government departments are conveniently forgetting that they have in the past encouraged and contributed to practices resulting in irretrievable environmental changes in the major rivers of the basin, and to continuing groundwater/ salinity problems. Policy makers seem to be committed solely to economic instruments, and they persist in expressing their views in a language that is unlikely to result in any effective communication with irrigators.

One possible solution is relatively simple, although it may be costly in the first instance. The States and State governments have, through the MDB Agreement, already demonstrated that the heat can be taken out of an argument simply by agreeing to forget the past and to get on with the future. The future in this instance is to admit that no amount of research and development is going to result in universal implementation of 'best practices'. Even if the 'best practices' were largely successful, governments would then be faced with a problem of similar magnitude resulting from dryland farming.

A realistic approach is to provide a comprehensive drainage and wastewater disposal system to service both irrigated and dryland groundwater recharge problems. At an initial cost of $6 billion and an annual service charge of $66 million this does not seem to be outside the realms of reality - particularly when the annual servicing would cost less than 1 per cent of annual agricultural production.

In addition there is a large urban population that also has a stake in having access to a healthy rural and riverine environment for recreational purposes. It would be reasonable to expect that part of their tax money should be used towards establishing a permanent system of environment protection.

There would, of course, also need to be a contractual agreement with irrigators concerning water usage and cost and with dryland farmers concerning practices resulting in excessive groundwater recharge. The development of such contracts would simply mirror the process already established between the Commonwealth and States

Until governments and farmers sit down together and discuss realistic solutions more and more productive land will be lost and more and more ecological systems will be destroyed.

Glossary of Terms

aquifer Any part of the soil profile or underlying rock material that will allow water to drain when opened to the atmosphere. For example, if a hole is drilled into the soil and water drains into the hole then an aquifer has been penetrated. There are many types of aquifer, depending on their position and structural characteristics.

capillary water In some ways the water that is held in soil pores behaves in the same way as if it were held in a bundle of glass tubes of very small diameter. Water moves upwards from the saturated zone by capillary forces alone in the same manner as water rises up through a piece of blotting paper dipped in a glass of water. The finer the pore size the greater the capillary rise.

EC unitsThe electrical conductivity (EC) of water provides a measure of the amount of salt dissolved in the water. The International Unit for electrical conductivity is deci- Siemens per metre (dS/m) but frequently in the literature we find the use of milli- Siemens per centimetre (mS/cm) or micro- Siemens per centimeter (



S/cm = 1 mS/cm = 1 dS/m

To avoid completely confusing the farming community it has become the practice to refer to

S/cm as 'EC units'.

free water evaporation At meteorological stations the daily evaporation of water is measured from a standard sized pan of water. It is reported as Pan Evaporation in mm/day. Evaporation from open bodies of water (lakes, rivers) is taken as about 0.7 x Pan Evaporation and is referred to as the potential free water evaporation.

Gigalitre (Gl or GL) 1,000 megalitres.

groundwaterWater contained within the pore space of soil and rock materials.

groundwater discharge Groundwater that escapes into a stream bed, lake or ocean, or through the land surface. Sometimes it is referred to as return flow.

groundwater mound A groundwater table is usually more- or- less parallel with the soil surface. In sites where groundwater recharge rates are high, a mound may develop in the groundwater table. When recharge is reduced or stopped the mound slowly disperses back to the same level as the surrounding groundwater table.

groundwater rechargeWater that has drained below the root zone of any local vegetation and which is then able to drain downwards to add to the underlying layer of saturated soil. In an irrigation system this water may be referred to as the leaching fraction.

groundwater table Refers to the upper surface of a layer of soil or rock material that is saturated with water.

levee bank When streams overflow they dump suspended soil particles on the surrounding flood plains. The largest particles drop out of suspension first so it is quite common for a stream to have sandy banks. If the stream then changes course the old sandy levee banks are left stranded in the landscape in the form of low dune- like structures.

light textured soil Soils are described as light textured if they contain a large proportion of sand particles. Heavy textured soils, on the other hand, contain a high proportion of clay.

Megalitre (Ml or ML) 1 million litres.

pore space Soil is a porous material made up of solids, water and air. The pore space in a clay is often about half of the total soil volume. i.e., 50 cubic centimetre (cc) pore space per 100 cc soil. Usually about 45 cc of that pore space would be permanently filled with water, with the remaining 5 cc occupied by air. Thus the capacity for the clay to absorb more water is 5cc per 100cc soil. This is referred to as the effective or available porosity.

riparian zone The land immediately next to a creek or river. For soil conservation purposes the width of the riparian zone is usually taken as 20m on each side of the creek and the natural vegetation in the zone is protected by law.

salinisation The process of accumulating soluble salts at or near the soil surface. This usually occurs by evaporation of groundwater that discharges through the soil surface.

salt scald When salt concentrates at the soil surface it kills the vegetation, thus leaving the soil surface exposed to erosion by wind and water. The bare soil surface is referred to as a 'salt scald'.


Water is a relatively scarce resource in the 1.06 million square kilometre Murray- Darling Basin (MDB). Annual rainfall across the basin varies from less than 300mm per year in the west to a maximum of about 3,000mm per year in the eastern ranges. This is offset to some degree by an annual potential free water evaporation rate of 3,000 mm in the west to 1,200 mm in the east. Only two to three per cent of the annual input of 466,000 Gigalitres of rainfall actually enters the stream system (1 Gigalitre (Gl) = 1000 million litres).

The geographic range of the MDB also means that frequently one or more parts of it can be experiencing drought conditions whilst the remainder is having normal rainfall. For example, in recent times the Darling River ceased to flow in 1991 and again in 1994 due to drought in southern Queensland and northern New South Wales.

Historically the Murray and Darling Rivers have had, and still do have, an important influence on the development of agriculture in this country. Annual agricultural production from the basin is in the range of $10 to 12 billion and it contains 75% of Australia's irrigated lands.

The role of the Murray has changed markedly since early European settlement. Initially, it was a major transport link between the coast and inland grazing areas but, because of its erratic flow patterns, action was taken to regulate the flow with a series of locks and weirs. Only 16 of the 35 planned structures were ever completed because railways took over as the preferred form of transport. However sufficient regulation was achieved to ensure that the Murray would never be the same again in terms of what is now called its 'natural environment'.

Added to river regulation there has been a series of land settlement schemes instituted by governments with a vision of converting the MDB into one of the major food bowls of the world. Unfortunately the knowledge of how to manage a diverse environment was almost completely lacking. Hence the hydrological regime of the MDB has been irreversibly altered and with it many biological systems have been modified or destroyed.

Less than 15 years ago government departments were still very reluctant to admit that their water management policies and actions were responsible for high groundwater tables, soil and stream salinisation, dying redgum forests and the disappearance of many bird, animal, and aquatic species. The 1980s saw a massive turnaround in the environmental consciousness of governments and they are now attempting to persuade landholders to help reverse the damage of past actions. The new objective of governments is 'Ecologically Sustainable Development' (ESD). This derives from the World Commission on Environment and Development (WCED, 1987) definition of sustainable development, viz., a policy that 'meets the needs of the present without compromising the ability of future generations to meet their own needs.' More specifically, the Commonwealth Government (1990) enunciated the following principles of ESD:

. improvement in material and non- material well- being;

. intergenerational equity;

. maintenance of ecological systems and protection of biodiversity;

. global ramifications, including international spillovers, international trade and international cooperation; and

. dealing cautiously with risk, uncertainty and irreversibility.

In spite of the legalistic type language used, there would be few that disagree with the notion of ESD since it appeals to the environmental conscience. However it remains to be seen how successful such an appeal will be since, unlike governments, landholders still have to make a living from their land. They may not be able to respond rapidly enough to prevent degradation of the soil, water and biological resources of the MDB from continuing well into the future. The real questions that need to be answered are whether governments:

. can define realistic objectives within ESD principles; and

. what incentives will be required to achieve those objectives.

At the moment government objectives appear to be generic rather than specific. For example, it is easy to formulate a policy saying that salinity and nutrient levels in the river system need to be lowered but real progress will not be made until specific objectives and specific incentives/disincentives are agreed to with individual landholders.

Since the irrigation industry is by far the largest user of water diverted from the Murray- Darling river system, this paper examines the question of whether irrigation is an ecologically sustainable form of land use within the MDB.

The Water Resource

The greatest proportion of irrigation water in the MDB is obtained from surface streams and reservoirs, with only about 3 per cent being extracted from groundwater. Surface water supplies, although of good quality, are subject to the vagaries of climate whilst groundwater use is restricted to areas having high yielding aquifers with low salinity water. The basin is comprised of 26 sub- catchments which provide discrete geographical units for management purposes (Figure 1). The Murray- Darling Basin Commission is responsible for managing water flow in the River Murray and as far north as Menindee Lakes in the Darling River. This will be discussed in more detail in a later section.

Surface Water

Stream flow results from two sources:

. surface runoff during and after rainfall. This flow is usually of quite short duration (hours), and

. groundwater returning to the surface in situations where creek and river beds are incised deeply enough to intercept the groundwater system. Return flow of groundwater may continue unabated for weeks to many months. It also carries with it various amounts of soluble salts which are discharged into the stream system.

About 60 per cent of water flow in the Murray River originates from the wetter south- eastern catchments. The Darling and Murrumbidgee catchments contribute about 12% and 11% to total flow, respectively.

The average discharge of water at the mouth of the Murray is now about 4,500 Gigalitres/ annum as compared with the 12,600 Gl that would be expected if there were no upstream diversions of river water. In extreme drought years the Murray barely manages a surface discharge into the sea because of the high evaporation losses that occur in Lake Alexandria. In fact about half of the water delivered to South Australia each year is lost in this way.

The highly variable climatic conditions experienced in the MDB are reflected in the very erratic stream flow record shown in Figure 2. Also shown in Figure 2 is the salt concentration of the Murray River water at Morgan (South Australia) over the same time period. Since flow and salt concentration are more or less inversely related, salinity has become an important indicator of both the quality of the natural resource and management of the flow regime.

Diversions from all of the rivers in the MDB, for human purposes, now account for 10,000 to 11,000 Gl/ year (Close, 1990), or more than twice the amount that flows out of the River Murray into the sea. Diversions have continued to increase over time (Figure 3) and, even with an increased concern about water usage by governments, the upward trend shows little sign of abating.


Very large volumes of groundwater are present in the MDB but the quality and ease with which it can be pumped is very variable. The most useful supplies, for irrigation purposes, are generally obtained from relatively shallow sand and gravel beds in the vicinity of existing streams. The proximity of these aquifers to surface streams also improves their recharge characteristics due to downward and lateral leakage from the surface water.

In 1984/85 8,320 Gl of water was used within the Murray- Darling Basin for irrigation purposes (MDBMC, 1987). Of this 263 Gl, or approximately 3 per cent of the total, was groundwater. The ratio and quantities for the remainder of the MDB may be somewhat different and changes may have occurred in the intervening decade but detailed, up- to- date statistics are not in a readily available form.

Principles of Irrigation

It is a basic premise of any irrigation system that an excess of water, above plant requirements, must be applied to the crop. This excess water is called the leaching fraction and as a rule of thumb should not exceed about 10 per cent of the applied irrigation and rain water.

A leaching fraction is necessary because plants, when taking up and transpiring water from the soil, leave soluble saltsbehind. Taken to the extreme, plants could actually kill themselves by concentrating salt in the root zone. Hence it is necessary to have either an excess of irrigation or rain water to flush the soluble salts below the depth of the root zone.

The excess water (and salts) continue to drain downwards and eventually add to the underlying groundwater table (Figure 4a). If the additions exceed the capacity of the groundwater system to drain out into a stream, or the sea, a groundwater mound will develop beneath the irrigation area. If the mound rises to within 1 to 2m of the soil surface then groundwater can evaporate directly to the atmosphere by moving upwards through the soil pores in much the same way as water rises up through a piece of blotting paper when it is suspended in a glass of water. As the capillary water evaporates at the soil surface, salts are concentrated in and around the root zone.

Discharge of groundwater at the soil surface creates either waterlogging and/or salt scalds (Figure 4b), depending on the rate of discharge. Even good quality groundwater contains some soluble salt and if it discharges slowly at the soil surface, salt concentrations in the root zone will gradually increase, simply due to evaporation.

It is not always appreciated by irrigators that quite small amounts of water escaping below the root zone can result in a relatively large rise in the water table (Figure 5). Although 40 per cent or more of the subsoil volume consists of pore space, it is never completely dry. Hence the 'effective porosity' may be as low as 1 to 5 per cent of the total soil volume. Thus the drainage of 1mm of water below the root zone can result in an increase in the height of the saturated soil zone of 20 to 100mm. That is, the height of the groundwater table increases by 20 to 100mm by the addition of 1mm of water. Increases in watertable heights, beneath irrigated areas, are frequently in the order of 200 to 500mm/ year. For example in the Berriquin Irrigation Area (NSW) groundwater tables were 20m or more below the surface in 1950 but rose to within 2m of the surface in about 30 years of irrigation.

The mound is not only a hazard to the irrigated area but, because it creates a pressure on surrounding groundwater systems, it has offsite effects as well. These effects may be manifested as increased flow of saline water into neighbouring streams or as discharge through the soil surface of lowlying neighbouring areas. Given that there is about 1.6 million ha of irrigated land in the MDB then groundwater mounds, due to irrigation, have the potential to influence a considerable area within the basin.

Unless the problems of groundwater recharge associated with irrigation practices can be solved on a basin- wide basis, the irrigation industry cannot possibly be regarded as being ecologically sustainable.

Reducing Groundwater Recharge

Reference to Figure 4a shows that reduction in groundwater recharge is related to a number of factors:

The physical characteristics of the soil are a major determinant factor. Excessively permeable soils allow rainfall and irrigation water to drain below the root zone before the crop has a chance to use it. Given the spatial variability of many soils of alluvial origin this frequently presents a problem in the physical layout of irrigation bays.

The water use characteristics of the crop grown can influence the amount of groundwater recharge. Such factors as the depth of the rooting system, the duration of the crop, the amount of leaf area produced, and physiological factors related to the amount of water used per unit of dry matter production all affect the water use efficiency of the crop.

Irrigation practices that need to be considered include the method of applying water (flood, furrow, sprinkler, trickle), the speed with which excess surface water can be drained (this also involves paddock design and land surface levelling), the frequency of irrigation, and whether any soil moisture monitoring techniques are employed.

The longer water is ponded on the soil surface the more opportunity there is for drainage down through the profile and for groundwater tables to rise. Thus the rice industry, which depends on maintaining a flooded soil surface for periods of 100 days or more, represents a major hazard with respect to groundwater recharge. On the other hand, with a deep- rooted fodder crop such as lucerne, it is technically possible to achieve quite low leaching fractions.

Electromagnetic induction methods are now available for mapping soils in terms of whether they will be 'leaky' or not when subjected to irrigation (Beecher, 1994). However State departments have been reluctant to apply the techniques on an industry- wide basis. Various infiltration tests or evaluations based on the thickness of underlying clay layers, as used in the past, have simply not been good enough for determining the suitability of riverine plains soils for irrigation purposes.

All of these factors are under the direct control of the irrigator. Some options may be relatively expensive but they are not impossible to achieve. The decision to implement one or other practice is usually made on financial grounds but undoubtedly the availability of cheap irrigation water has not encouraged the adoption of environmentally responsible practices.

Irrigation Waste Water Disposal

Groundwater discharge will, in the future, affect practically all irrigation areas in the MDB because, unlike coastal catchments, groundwater cannot drain freely to the sea. The Murray Basin is a saucer- shaped structure filled with largely water- saturated sediments. The deepest part of the basin (approximately 600m) is in the Renmark - Wentworth area and groundwaters naturally gravitate in that direction. The River Murray also flows in that direction and has cut an outlet to the sea through the south- western rim of the basin. For any groundwater to escape it must pass through this relatively narrow, sediment- filled gorge or discharge into the river itself. The restriction to flow means that watertables are close to the surface throughout the basin and that rainfall and irrigation water entering the land surface can only escape by being transpired by vegetation or by discharging through the soil surface. Unfortunately the type and distribution of vegetation systems that have been established since European settlement cannot handle the rainfall inputs, let alone irrigation inputs.

Even with the management tools described above it is likely that a large proportion of irrigated farms will continue to contribute excessive amounts of water to the underlying groundwater systems. In parts of northern Victoria and in the Murrumbidgee Irrigation Area, waterlogging and salinisation have reached such dimensions that some land is simply being written off as 'sacrificial discharge areas'. Acceptance of sacrificial areas really means an admission of failure to manage irrigation systems in an environmentally responsible manner.

A part of that failure is due to past and present land managers, but inappropriate water distribution systems designed by departmental irrigation engineers have also contributed to leakage into groundwater systems over many decades. For example in Victoria water supply channels to the Wimmera District are now having to be replaced with a pipeline due to the excessive channel leakage. Also in the Jemalong- Wyldes Plains Irrigation Areas of NSW a main supply channel was built on top of an old levee bank of the Lachlan River simply because the levee was the highest part of the landscape and so facilitated water distribution to farms. However the light textured levee bank soils allow considerable leakage into a local groundwater mound.

It is possible that a vegetation strategy could be devised that would restore the water balance to a state where groundwater discharge did not represent a threat to the environment. However this is only being approached in a piecemeal way at the moment through such Landcare- type projects as tree planting (Williams, 1995). In order to buy time whilst an effective biological solution is implemented, the most sensible approach would be to install a comprehensive drainage system and dispose of the waste water in an environmentally acceptable manner.

Extraction of groundwater can be achieved by installing horizontal drains below the depth of the crop root zone or by pumping water ( vertical drain) from the groundwater zone. The cost- effectiveness of either depends on the hydraulic properties of the soil. For example, in some situations one pump (tube well) may draw down the water table over an area of one to a few square kilometres whilst in less permeable soils pumping may not be an option at all. In the latter case horizontal drains are used and these drain into a common sump from which the waste water is then pumped.

Depending on the salt concentration of the waste water, it is sometimes possible to mix it with excess irrigation water that has been drained from the soil surface and then add it to good quality supply water for the next irrigation (Figure 6). There is of course a limit to how many times such water can be recycled as the salt concentration increases with each cycle.

A large proportion of drainage water is not re- used at all. In the past it was often discharged into a natural drainway (and eventually into the Murray River system). Now it is commonly pumped to an evaporation basin. Evaporation basins may be self- enclosed systems (in which case they have a finite life) or the concentrated brine may be pumped back into the river at times of high river flow. The evaporation basin/dilution flow technique is presently the favoured option although the longterm environmental consequence of creating a pool of concentrated salt brine beneath an evaporation basin is still poorly understood. If that brine should unexpectedly be mobilised into surrounding groundwater systems then there would be a potential for considerable environmental damage.

About 2 million tonnes of salt is discharged into the sea from the Murray River each year. Figure 7 shows that on average, salinity in the River Murray increases from 100 EC units at the headwaters to about 800 EC units at Lake Alexandrina. Major increases in concentration occur at the junction of Barr Creek with the Murray and again downstream from Lock 5. The latter input is due to groundwater discharge directly into the Murray, as it comes under the influence of the bottleneck to subsurface flow, described earlier.

The health of land within the Murray basin depends to a great extent on the ease with which groundwaters can drain naturally, or be pumped, into that drain. In other words we are trying to manage the Murray both as a supply line of good quality water and as a drain for highly saline ground and evaporation basin water. It is not a happy combination. Other countries faced with similar problems have separated the supply and waste disposal problems by constructing dedicated drainways.

For example Pakistan is building a canal stretching more than a thousand kilometres through the Indus Valley to carry waste groundwater to the sea. In California, where irrigation salinity problems commenced in the 1880s, there are numerous concrete- lined drains leading into wetlands. Agro- politics and environmental concerns are still holding up the construction of a main disposal drain to the sea - 20 years after its first proposal. Egypt also faces a massive drainage problem now that the Aswam Dam has prevented a regular flushing of salt out of the Nile floodplain.

Irrigation Water Use

The irrigation areas of the MDB are shown in Figure 8, and surface and groundwater irrigation usage for 1984/85 is provided in Table 1. It will be noted that groundwater extraction by irrigators was highest in the Condamine- Culgoa (Qld), Namoi and Murrumbidgee (NSW), Broken River (VIC), and Lower Murray (SA) catchments. Diversions for irrigation increased by 33 per cent between 1984/85 and 1992/93 (Table 1) with an extraordinarily large increase in the Upper Murray/Kiewa/Ovens region of Victoria. There is, however, some doubt concerning some of the comparisons as the MDBC has now changed its stream flow accounting system. Such changes make it difficult to extract valid trends for particular irrigation areas.

Some 75 per cent of Australia's irrigation occurs in the MDB and it presently produces an annual gross income of about $4.5 billion from the production of livestock, fodder crops, rice, cotton, fruit and vines.

The total area irrigated annually is approximately 1.6 million ha (Table 2). One difficulty in obtaining accurate figures for irrigated areas is due to the nature of the agricultural enterprise. For example data for irrigated grain and hay crops is relatively straightforward, but where farmers simply apply water once or twice a year to grazed pasture land, they do not necessarily keep accurate records of the area irrigated. An 'autumn flush', to stimulate pre- winter growth of semi- improved pastures is a common practice in many irrigation districts. Whether this qualifies the land as being 'irrigated' is a moot point.

Water application rates are commonly in the range of 5 to 15 Ml/ha, i.e., the equivalent of an extra 500- 1500mm of rainfall per year. High water use crops such as rice may have application rates well in excess of 20Ml/ha, depending on the soil characteristics.

Water is supplied under licence either by State diversions from the rivers, private diversions or private groundwater pumping. The cost of supplied water varies from state to state but ranges from as low as $2.78/Ml in the Namoi Valley in NSW to $62/Ml in the Wimmera District in Victoria. The large price variation is due almost entirely to the cost of distributing water. The Namoi irrigators provide their own distribution system whereas in the Wimmera District water has to be piped to the area.

Various formulae for water use are applied by the State water authorities to cover basic and excess allocations and many historical anomalies still persist. For example, until recently in the Jemalong- Wyldes Plains Irrigation Districts, water used in excess of the entitlement cost considerably less per unit that for the entitlement itself. Similarly, some areas are charged for their full allocation whether they use it or not. Such pricing structures encourage wasteful use of the water resource.

Managing the Water Resource

Management of River Murray flow is the responsibility of the Murray- Darling Basin Commission (MDBC) which is the executive arm of the Murray- Darling Basin Ministerial Council (MDBMC), representing Queensland, New South Wales, Victoria, South Australia and the Commonwealth. It is the primary responsibility of the MDBC to regulate river flow on behalf of the member States. In effect this means that South Australia, which contributes virtually nothing to surface flow, receives at least 1850 Gl each year. New South Wales and Victoria share flow above Albury and retain control of their respective tributaries below Albury.

The initial River Murray Waters Agreement 1914 between the Commonwealth, New South Wales, Victoria and South Australia concerned water flow only. In 1976, on the recommendation of the four governments, the River Murray Commission (RMC) commenced monitoring water quality and a number of environmental parameters in the Murray Basin.

A new River Murray Waters Agreement 1982 became a Schedule of the River Murray Waters Act, 1983. The latter was amended in 1987 to become the Murray- Darling Basin Act 1983 and formally established a Murray- Darling Basin Ministerial Council (MDBMC) and the Murray- Darling Basin Commission (MDBC). Queensland joined the Murray- Darling Basin Agreement in 1992 and now participates in the Commission's Natural Resource Management Strategies (NRMS). The Murray- Darling Basin Act 1983 was repealed to become the Murray- Darling Basin Act 1993. and included the new Murray- Darling Basin Agreement 1992. Schedule C of that Agreement contains the details of a Salinity and Drainage Strategy. It is expected that other Strategies (Algal Management, Irrigation Management and Natural Resource Management) will be written into the Agreement as they become finalised.

Managing Water Supply

Traditionally State water resources departments have been responsible for the supply and distribution of irrigation water. However, in NSW, Victoria and South Australia, there is a move towards privately run distribution systems. This has been brought about largely by the irrigation industry's long proclaimed conviction that it could handle the task more efficiently than the 'government'. Also, it appears that governments are not showing too much reluctance at passing on the responsibility of managing a resource that is both costly to operate and which has major adverse impacts on the environment.

The Irrigation Corporation Act 1994 of the NSW Parliament formalised this arrangement for Coleambally, Jemalong- Wyldes Plains, Lower Murray, Murray and Murrumbidgee areas. The Act made specific reference to the environment and to the efficient use of water as matters that should be considered before granting a licence to an irrigation corporation. It also mentions 'land and water management plans' on a number of occasions but nowhere does it define just what they might be.

Regardless of these provisions Irrigation Management Boards, consisting of community members, are already assuming responsibility for water deliveries whilst Land and Water Management Plans are still being developed. It remains to be seen whether the Boards (or Corporations) pay any more than lip service to removing the groundwater problems that have developed under decades of government department control.

Managing Water Quality

A number of resources management strategies, including a 'Salinity and Drainage Strategy', have been developed (MDBMC, 1989). For the first time past arguments about who was to blame for polluting the river system were put aside and the States, together with the Commonwealth, agreed on a course of action to restore water quality and other natural resources within the Murray- Darling Basin.

Under the Murray- Darling Basin Agreement 1992 each State makes a financial contribution towards the management of the river and remedial works designed to reduce the inflow of salt into the Murray River (Table 3).

The States then receive salt credits that allow them to dispose of some of their own saline drainage water back into the river. Only drainage actions approved by the MDBMC can be implemented. The net result should be:

. a steady decrease in average salinity at Morgan (SA) which, because it is downstream from the major irrigation areas, is the reference site with respect to monitoring river salinity.

. triumph for a commonsense approach to solving a natural resource problem.

The performance of any salt interception scheme designed to reduce inflow of salt into the river can be assessed quite accurately at the extraction point in terms of tonnes of salt removed. However the variable nature of water and salt flows (Figure 1) means that little credence can be given to their statistical trend lines (Close, 1990). Thus the effect on river water quality due to landuse changes, such as more efficient irrigation techniques or reafforestation, cannot be assessed quantitatively at this stage. Given the large temporal variability in river salinity it is curious that the Murray- Darling Basin Agreement 1992 defines a 'significant effect' as one that alters the average salinity at Morgan by the extraordinarily small amount of 0.1EC unit. i.e., about 60 milligrams salt per 1000 litres.

Managing the River Environment

Salinity is not the only aspect of water quality in need of attention. Pollution of the river systems with urban and agricultural wastes has seen a marked increase in the occurrence of toxic algal blooms, particularly during periods of low river flow. Hence considerable effort is now being directed towards protecting the river system from being used as a waste disposal unit, and for the reuse of waste water for other productive purposes. On average, irrigation drainage contributes about 10% of the phosphorus and nitrogen loads measured in the Murray- Darling river system each year. (GHD, 1992)

Urban and sewerage waters account for greater than 30% of phosphorus and nitrogen loads in the river system in an average year. Treatment and use of this water for irrigation of tree lots has been trialled successfully in a number of places and this may well become the norm for many growth centres along the MDB waterways. Similarly, revegetation of the riparian zone may soon be seen as the most effective means for preventing diffuse sources of surface pollutants from entering waterways.

The MDBMC is now reserving water for flushing rivers at times of very low flow as a means of preventing algal blooms. Such 'environmental' water use does, of course, create some conflict with irrigators, particularly during periods of drought. Another 'environmental' water management initiative in 1994 was the agreement by the MDBMC to allocate an average of 100 Gl/yr towards the periodic flooding of the Barmah- Millewa redgum forest area. Regulation of river flow over many decades had seen a marked deterioration of this forest due to lack of regular flooding. In a similar vein New South Wales reserves water in the Lachlan and Macquarie valleys for wetland and waterbird habitat management.

It seems that at last we are recognising that the surface water resources of the MDB are not just for agriculture and human consumption, but are also a very necessary part of the wider biological environment.

The Irrigation Industry Dilemma

The dilemma facing the irrigation industry in the MDB is that, on one hand, it is regarded as being highly efficient and productive; on the other hand, it is clearly responsible for a considerable amount of environmental damage. The MDBMC (1989 background papers) estimated that there are more than 500,000 ha of irrigated land with a water table at less than 2m from the soil surface. Of that area, about 150,000 ha has surface soil salinity problems. If attempts to reverse this process are not successful then over the next 30- 40 years the area with shallow watertables is likely to double and the proportion with salinised surface soils will increase markedly.

Surrounding dryland agricultural areas are also suffering from rising groundwater tables and secondary salinisation of surface soils caused by excessive clearing of deep- rooted, permanent vegetation. The cause is the same as in irrigated areas, viz, recharge of groundwater systems by rainwater infiltrating below the rooting depth of existing crops and pastures. Although the rate of groundwater rise is probably only one fifth to one tenth of that in irrigated areas, it never- the- less is responsible for more than 1 million ha of salinised land in the MDB.

The problem is not just one of degrading irrigated land as adjacent dryland areas are also affected. However the fact that some of the dryland problems originate from their own landuse practices makes it difficult to apportion blame. This situation is both a challenge to the environmental conscience of irrigators and an escape from accepting responsibility.

The Government Dilemma

There has been a massive increase in information provided to irrigators and dryland farmers through the National Landcare Program (Williams, 1995) but there is still a strong sense of denial, or at least a considerable amount of selective acceptance of scientific information, in the irrigation industry. This is well demonstrated by Graham Blight (1992), (then) President of the National Farmers' Federation and himself a rice farmer, who in presenting a paper on 'Sustainable water use: an agricultural perspective', said:

In the Murray- Darling Basin, salinisation is largely due to the slow change over the last 500,000 years to a more arid climate, and not solely due to the impact of settlement. The region's salt problem is more a result of the geological structure which acts as a salt trap than of the impact of farming practices.

Of course it is not only the irrigation community that has difficulty in coming to terms with environmental realities. For example, Ashley Cooper, Vice- President of the Queensland Cattlemen's Union, speaking about clearing trees in areas close to the Dividing Range, was recently quoted (AAP, 31- 3- 95) as saying,

It is a fact of life that land degradation in those areas is the result of too many trees that do not allow grass to grow.

Both landusers reflect a very common attitude amongst irrigation and dryland farmers, viz., accepting information supportive of their own industry and rejecting less favourable evidence.

However one might well be inclined to sympathise with both landholders when faced with academic jargon from governments and emotive statements by some environmentalists. For example, the following erudite statement on salinity would not be likely to arouse any feelings of environmental responsibility amongst irrigators ( Powell, 1994):

The grim spectacle of salinisation could be interpreted as a warning of primeval natural purifications, a violent purging of presumptuous intruders.

Whilst these statements may cause wry comment, or righteous indignation in some areas, they are really symptomatic of a major problem facing governments committed to implementing ecologically sustainable development policies. Environmental advisors and governments, for whatever reason, are simply not creating a sense of urgency amongst irrigation farmers to review their options before large amounts of land are irretrievably damaged.

Water diversions, presumably dominated by irrigation demand, continue to rise regardless of environmental considerations (Figure 2). Much of that increase, at least since the early 1980s is due to a fuller utilisation of the original water entitlements and to trading in licences, rather than to governments issuing new licences. Nevertheless it indicates scant regard for the very land resources upon which the industry depends for its survival.

The irrigation industry would probably argue that it actively supports research through various government Corporations and participates in the development of Land and Water Management Plans. However most of the research is directed towards maximising production rather than devising sustainable systems. It is possible that the regional Land and Water Management Plans, now nearing completion in a number of irrigation areas, will provide a vehicle for addressing this problem in the future.

Of all the irrigation practices contributing to environmental damage, using permeable soil is the most intractable. From the author's personal experience there is still considerable resistance within the industry to any thought of retiring land from irrigation, even when that land is demonstrably unsuitable with respect to ground water recharge. It really should be the first point of negotiation between irrigators and governments but to date little effort has been directed towards this problem by either party.

Government Initiatives

Much has been written about government involvement in developing irrigation areas, the supply of water storage and distribution systems, irrigation agronomy research, and research into the subsequent environmental problems associated with irrigation. It is not the purpose of this paper to review a topic that has been, and still is being, reviewed regularly. The overriding fact to be faced is that practically all irrigation areas in the MDB are creating environmental problems.

There is now 10 to 15 years experience of government programs designed to assist in technology transfer in areas suffering from water logging and salinity. There is also the realisation that although information transfer schemes are generally regarded by the farming community as a 'good thing', there is still a major blockage in the system with respect to the adoption of ecologically sustainable farming techniques.

Even in the longer running Victorian experience the process is painfully slow. For example, Stephen Coats (1994) in recounting his experience with irrigators in the Shepparton area, wrote:

Denial is the first reaction, followed by frustration and perhaps anger and then rationalisation and discussion. This eventually leads to acceptance and then action. ... Perhaps the Goulburn Valley is up to the rationalisation and discussion phase...

That is a description of an area that has been subjected to intense community education and training for more than a decade!

It is difficult to believe that a majority of irrigators has not heard of groundwater problems either through Landcare groups or via various agricultural media outlets. If they are aware and are choosing to ignore the messages then there will need to be a radical rethink of the approach that should be adopted by governments. A clue to this situation may lie in a statement by Cape et al (1994):

Most people will not change unless the pain of change is perceived to be less than the pain of staying the same.

This could be interpreted as saying:

. change will not take place until the landholder is on the verge of going broke, or

. change will not take place unless governments impose compelling financial incentives.

In the first case the process of 'going broke' would almost certainly entail considerable environmental damage to the particular irrigation farm and also to offsite areas. Hence it is not sensible for governments to sit back and wait for the inevitable to take place.

In the second case considerable thought has already been given to the measures governments might employ. Recommendations (ESD Working Groups: Agriculture, 1991; Council of Australian Governments, 1995) seem to favour:

. encouraging use of water in more profitable enterprises by making irrigation licences a tradeable item;

. increasing the cost of irrigation water to achieve complete cost recovery;

. removal of cross- subsidies used to reduce water costs; and

. imposing a tax on waste products, i.e., the polluter pays.

Simmons et al (1991) point out that subsidisation costs are in the order of $300 million/year for the basin and that efficiency gains of about $40 million/year could be achieved through the transfer of water entitlements.

Undoubtedly this 'carrot and stick' approach would have some beneficial effects but like many strictly economic approaches there could well be some unintended consequences.

One obvious consequence is that the sale of irrigation quotas simply shifts the groundwater recharge problem to another area, where the degradation process commences all over again. A second difficulty would be in devising and administering a sensible monitoring system on individual properties to assess a pollution tax.

The then Commonwealth Minister for Primary Industries and Energy, Simon Crean (1992), expressed the opinion:

You are not going to address, in my view, the solution by saying you go out and charge farmers more for water, because that marginalises the issues in a way that polarises the debate. I think the real problem with the ESD debate is that it has been polarised.

Determining Irrigation Policy

The Present Approach

The Australian and New Zealand Environment and Conservation Council (ANZECC) and the Agricultural and Resource Management Council of Australia and New Zealand (ARMCANZ) (1994) released an outline of policies directed at a National Water Quality Management Strategy (NWQMS) as part of the National Strategy for Ecologically Sustainable Development (1992). The stated policy objective of the NWQMS is:

to achieve sustainable use of the nation's water resources by protecting and enhancing their quality while maintaining economic and social development.

This all- embracing objective is a clear statement of where the nation wants to go but, if taken literally, it might well be regarded as impractical and idealistic. Even if irrigation in the MDB ceased today it could still take decades before groundwater mounds dispersed and damaged land could be rehabilitated. Also groundwater recharge in dryland agricultural areas would have to be controlled before there was any chance of preventing discharge of highly saline groundwater back into the River Murray.

The objective, enunciated above, requires some clarification. In the 'Policies and Principles - A Reference Document' (ANZECC/ARMCANZ, 1994b) the statement on objectives commences with the words 'to pursue sustainable use ...' rather than 'to achieve sustainable use ...' However in the main text it reverts to 'achieve'. The irrigation industry, for the reasons described previously, would find it virtually impossible to fulfil the first objective but might well be prepared to 'pursue' sustainable water use.

The 'Policies and Principles' document is directed principally at water quality in relation to point and diffuse sources of pollution. However in assessing irrigation, or indeed agriculture in general, an over- riding factor is water quantity. Excess water (rainfall or irrigation) is itself a waste product of those industries because it mobilises soluble salt and frequently discharges that salt into distant parts of the environment. The absence of an easily monitored cause- and- effect linkage appears to have been recognised by ANZECC/ARMCANZ since it included a 'best management practice' philosophy for diffuse source pollution management. But this begs the question of whether the authors simply chose the soft option or whether they were implying that it really falls into the 'too hard' basket.

Waste management options were clearly defined in a series of actions of decreasing desirability:

. waste avoidance

. recycling or waste reclamation

. waste re- use

. waste treatment to reduce potential degrading impacts

. waste disposal

In applying these options to the irrigation industry it is clear that there are a number of land management options designed to avoid the wastage of water. Recycling and re- use of drainage water have been discussed and their limitations recognised. There is no economically viable potential for treating drainage water to remove soluble salts and the presently preferred evaporation basin/dilution flow disposal system has been described in less than enthusiastic terms. Together they add up to 'best management practices' and together they are unlikely to solve the problem of groundwater recharge to a sufficient degree to prevent continuing environmental damage in large parts of the Murray- Darling Basin.

Hence discussion of 'regulatory' and 'market' approaches are more academic in nature than realistic at this stage. In effect the micro- economic reform approach, adopted by the Council of Australian Governments, says 'we will keep on increasing the cost of irrigation until either the inefficient users drop out or the problem goes away'. Of course this approach is tempered with statements concerning social equity, conservation of biodiversity, dealing cautiously with risk, economic diversity/resilience and global issues. However these issues have barely surfaced with respect to the irrigation industry. It would seem that econocrats, endorsed by the major political parties, have little appreciation of the biological system with which they are dealing.

It also seems surprising that government technical advisers are apparently unaware that the groundwater problem simply will not go away in the foreseeable future by more efficient water use alone. As explained earlier, the necessity of a leaching fraction and the spatial variability of soils will almost certainly ensure that the process of excessive groundwater recharge will continue, even under so- called 'best management practices'.

Possible Future Approaches

Clearly, the first policy statement governments need to make with respect to a system of irrigated agriculture in the MDB, is one that lets irrigators know whether they have a long term future in that industry. So far this does not appear to have happened since the only statements released to irrigators concern options for increased future controls. If governments, in the knowledge of certain continuing environmental damage, do support the notion of a $4.5 billion/year industry then they must:

. state the level of environmental damage that society is prepared to condone; or

. be prepared to install a groundwater disposal system to protect the environment.

The first option would mean providing very specific, area by area, definitions of the amount of groundwater recharge that would be permissible, and providing monitoring systems to ensure compliance. The second would mean a comprehensive drainage system with disposal of waste water external to the basin. Surface and subsurface drainage schemes already exist in most irrigation areas. For example in the Murray Basin about 700,000 hectares are serviced by surface drains and 115,000 hectares by subsurface drains (MDBC, 1990). However a further 900,000 ha and 350,000 ha would benefit by surface and subsurface drainage, respectively. Moreover it is questionable whether the existing disposal systems can be regarded as being environmentally sound.

It is technically feasible to install a drainage system that would remove all waste groundwater from irrigated areas and dispose of it in an environmentally acceptable manner. A 1990 review of the 'pipeline to the sea' proposal (MDBC, 1990) showed that full watertable control would cost about $6 billion, with an annual operations and maintenance cost of $66 million. The latter represents less than 1 per cent of the annual $10- 12 billion of agriculture production from the MDB. All options considered showed a negative present value. However such a calculation takes no account of the value that the community in general places on being able to visit an environmentally healthy countryside or river system. Neither does it take into account the fact that past governments have actively supported land use systems that have created the present problems and that there may be a moral obligation to rectify that situation.

Two major advantages of installing an effective drainage system would be:

. for a moderate extra cost it would service not only irrigated lands but dryland farming areas as well. In fact, as mentioned previously, there is little point in remedying irrigation recharge in isolation because dryland recharge is of at least the same magnitude and is a far more intractable problem to solve.

. it would enable the implementation of a drainage tax on individual irrigated holdings and hence act as an incentive for more efficient water use. Drainage volumes per unit area would provide a very good monitoring test for assessing compliance and performance.

One argument against the provision of a comprehensive drainage system is that it could encourage the continued wasteful use of irrigation water. However, as suggested above, it would be quite feasible, in consultation with the irrigation industry, to arrive at 'reasonable' water application rates for particular soils and for particular crops. The technology exists for mapping the relative risk of different soil types (Beecher, 1994) and the technology also exists for monitoring and mapping shallow groundwater systems. The fact that these tools have not received widespread use in the past is probably more a reflection on the willingness of departments involved in the irrigation industry to produce this type of information rather than a lack of resources. In fact one Land and Water Management Plan, at Berri in South Australia, has implemented a system based on performance. If a property is unable to meet an agreed set of water- use conditions the water entitlement is bought back by the water authority and the farm is converted to a dryland system (Conroy, 1995).

Although not specifically advocating a drainage system, a more radical funding system has been suggested by Meyer (In Conroy, 1995). He supports an 'environmental levy' on all food consumed to ensure that food- producing resources are protected. Such a proposal for the $10 billion/year food industry is certainly consistent with present government policy in other areas; e.g., funding road infrastructure through a multi- billion dollar per year fuel tax.

Regardless of the final agreed formula, it is essential that the irrigation industry be consulted in a meaningful way as a 'bunker' reaction has developed in response to being singled out for criticism. It is relatively easy for the community to blame irrigation farming systems for environmental damage and to ignore other landuse practices. Likewise, State Departments are equally sensitive to the issue of blame.

The Commonwealth and States Agreement on the management of the Murray- Darling river system provides a good example of what can be achieved in such a divisive and threatening situation. The governments simply agreed to wipe the slate clean, as far as past management practices were concerned, and to start again with an agreed set of rules. All partners to the agreement contribute financially and receive trade- offs in return.

There is no reason why similar contractual agreements could not be worked out with irrigators and dryland farmers. The trade- off between providing an effective drainage system, and hence increased productivity, would be the rationalisation of the irrigation water system to meet the micro- economic objectives of COAG. Also there would need to be trade- offs concerning irrigation practices and environmental performance with realistic compensation mechanisms for converting to an approved dryland system.

It is unlikely that any one formula would apply to all irrigation areas due to their wide geographic spread and their relative impact on the surrounding environment. However it should not be beyond the wit of irrigators and governments to come to a mutually profitable approach.


Australian and New Zealand Environment and Conservation Council (ANZECC) and Agriculture and Resource Management Council of Australia and New Zealand (ARMCANZ) (1994a) Water quality management - an outline of policies.

Australian and New Zealand Environment and Conservation Council (ANZECC) and Agriculture and Resource Management Council of Australia and New Zealand (ARMCANZ) (1994b) Policies and principles for water quality management - a reference document.

Australian Water Resources Council (AWRC) (1992) Water quality management in the rural environment: A reference document. AWRC, Australia.

Beecher, H. G. (1994) Identifying groundwater recharge sites in irrigated landscapes using EM induction techniques. NSW Remote Sensing Newsletter, 5, No. 2.

Blight, G. (1992) Sustainable water use: an agricultural perspective. in, Proceedings National Conference on 'Water quality management and ecologically sustainable development: Delivering opportunities. AWRC ANZECC, Adelaide.

Close, A. (1990) The impact of man on the natural flow regime. in, The Murray (eds N. Mackay D. Eastburn). Murray Darling Basin Commission, Canberra.

Cape, J., Chamala, S. and Syme, G. (1994) National Program for Irrigation RD: Technology transfer and adoption in irrigation. Land Water Resources Research Development Corporation (LWRRDC), Canberra.

Coats, S.(1994) Keys to successful industry development. in, National Program for Irrigation RD: Technology transfer and adoption in irrigation (LWRRDC), Canberra.

Commonwealth of Australia (1990) Ecologically sustainable development. A Commonwealth discussion paper. Dept Prime Minister Cabinet, Canberra.

Conroy, F., (1995) Sustainability: irrigation in the balance. Rural Research, No 166, 15- 17.

Council of Australian Governments (COAG) (1995) Report of the expert group on asset valuation methods and cost- recovery definitions for the Australian water industry. Feb. 1995

Crean, S. (1992) Water quality management and ecologically sustainable development. in, Proceedings National Conference on 'Water quality management and ecologically sustainable development: Delivering opportunities. AWRC ANZECC, Adelaide.

Ecologically Sustainable Development Working Groups (1991) Final Report. AGPS, Canberra.

Fleming, P. F. (1982) Irrigation and drainage in the Murray- Darling Basin. In Murray- Darling Basin Project Development Study: Working Papers. CSIRO, Division of Water and Land Resources, Canberra.

Gutteridge, Haskins Davey (GHD) (1992) An investigation of nutrient pollution in the Murray- Darling River system. Report to MDBC, Canberra.

Murray- Darling Basin Ministerial Council (MDBMC). (1987) Murray- Darling Basin environmental resources study. MDBMC, Canberra.

Murray- Darling Basin Ministerial Council (MDBMC). (1989) Murray- Darling Basin natural resources management strategy: background papers. MDBMC, Canberra.

Murray- Darling Basin Ministerial Council (MDBMC). (1989) Salinity and Drainage Strategy. MDBMC, Canberra.

Murray- Darling Basin Commission (MDBC) (1990) A pipeline to the sea. Report prepared by Gutteridge, Haskins Davey; ACIL Australia; Australian Groundwater Consultants.

Murray- Darling Basin Commission (MDBC) (1990). Report to the Murray- Darling Basin Ministerial Council on a Drainage Program for the Murray- Darling Basin.

Murray- Darling Basin Commission (MDBC) (1993) Algal Management Strategy for the Murray- Darling Basin. Draft. MDBC, Canberra.

New South Wales Government (1994) Irrigation Corporations Act 1994 No.41, New South Wales.

New South Wales Water Resources Council (1991) Water Facts.

Powell, J. M. (1994) Environmental degradation, 'sustainability' and the Murray- Darling Basin. People and Place, 2 6- 13.

Simmons, P., Poulter, D., and Hall, N.H., 1991 Management of irrigation water in the Murray- Darling Basin. ABARE Discussion Paper 91.6. Commonwealth Govt. Printer, Canberra.

Williams, B.G. (1995) Landcare and the Mythical Money Tree. Parliamentary Research Service, Research paper No. 22. Dept Parliamentary Library, Canberra.

World Commission on Environment and Development (WCED) (1987) Our Common Future, (the Brundtland Report). Oxford University Press, Oxford.