The current emergence of a long dry period caught Australia, and more particularly those responsible for managing our rivers, by surprise.

River systems, such as the Murray–Darling System, have had ‘long drys’ before, notably in the first half of last century, but our planning and allocation systems were not ready for them.

In the case of the Murray River, many people think that lack of rainfall and runoff in recent years is the main problem. The reality is that, even before the emergence of the current dry, there clearly was an over-allocation problem. That is, the sharing regime that all states had agreed to was making too much water available to water users – largely irrigators – and neither setting aside enough water for environmental needs, nor for the river itself. The river was in strife before the current long dry started.

It is now more than a decade since the Murray River experienced an over-bank flow. Time is not on the side of the many ecosystems that sit next to the river. Arguably, the clearest indicator of the severity of the Murray River’s problem was the decision to put a dredge in the mouth of the river in October 2002.

What went wrong? In retrospect and with the benefit of hindsight, the reasons we got into trouble are simple to see.

First, we decided to tack water trading onto a suite of entitlement, allocation and water- sharing arrangements that were never designed to cope with the nature of management challenges that water trading reveals. Trading drives innovation and investment, and encourages increased water-use efficiency. The result was a dramatic decrease in the amount of water returning to the river system.

Second, we also failed to get the accounting arrangements right in a number of other aspects of water management. In water accounting, the most fundamental principle is that that whenever one person or one process is allowed to increase the amount of water being used, someone or some process is going to get less water.

For example the extent of connectivity between ground and surface-water systems was not fully recognised. The reality is that in many areas increased groundwater use means decreased river flow and a reduction of the volume of river water that is available to downstream users, and to the environment.

The impact on water availability of increased plantation forestry and farm dams in the catchments above the public storages was also not adequately addressed.

The really sad part of the story is that all the governments involved – Commonwealth and state – were aware of what was happening, but were unable to act in a positive and coordinated manner.

And finally, we set a volumetric cap on surface water extractions in a manner that lacked sufficient flexibility to cope with a shift to a drier regime.

The good news is that the new Murray–Darling Basin Agreement, with the referral of powers to the Commonwealth and providing for the establishment of a new independent Murray–Darling Basin Authority, is written in a manner that should make it possible to rectify these accounting and water sharing problems. While the new agreement is not perfect, it is hoped that it it will prove to be strong enough to enable progress.

When it gets drier, the amount of water that runs off tends to drop exponentially. The result is a dramatic reduction in the amount of water available. But this is not the end of the story.

Unfortunately, a significant amount of water is needed to maintain a river system – enough to cover evaporation, critical human needs and conveyance of water for use. The result, if it gets permanently drier, is a dramatic reduction in the amount of water available for consumptive and environmental purposes.

Once enough water has been put aside for system maintenance and conveyance purposes, the remaining water has to be shared. The emerging solution is one that defines the environment’s interest in exactly the same way as all other entitlements.

Currently this is mainly being achieved by purchasing entitlements from irrigators and transferring them to the Commonwealth environment water holder. Some water savings are also being pursued through investment in infrastructure upgrades. Once this process is completed, environmental managers will have the opportunity to determine when and how to use ‘their’ share.

How much water does the river need? Immediately before the start of this current long dry, scientists estimated that that the environment needed about 20 per cent of all the entitlements on offer. In cap-equivalent terms, this amounted to about 1500 gigalitres (1,500,000 megalitres) of water. Note that this approach only secures a share: if it stays drier we will still have to accept considerable environmental loss.

The sober reality is that at current market prices, one megalitre (ML) of cap equivalent costs about $3000. That is, to purchase 20 per cent of all water entitlements will cost at least (1,500,000 x $3000) or $4.5 billion in current dollars.

The current value of the money on offer for purchasing water is less than $2 billion. Even with the estimated water savings from investment in infrastructure upgrades that are to be shared 50:50 with irrigators added, it will clearly not be enough.

After adjusting for inflation and the fact that the money on offer is to be made available over 10 years, it is clear that something is going to have to give. In recognition of this reality, many analysts are suggesting that all the money for infrastructure and entitlement purchases be put into a single bucket. After due diligence, this should ensure that the most water for the environment is secured for each dollar invested according to opportunities available.

We also need to remember that if this strategy was followed there would be no money left for the Darling System. Unfortunately, that system has a suite of problems that are similar, but not identical, to those in the Murray System.

Unlike the south, the northern system has few large storages at the top and flows tend to be episodic, can be quite large in volume and spread out over large areas. In such an environment, it is necessary to first decide where water is needed for the river system, control harvesting upstream and then find a way to shepherd the resultant water past thirsty pumps to the place where it is needed.

As the late Peter Cullen once commented, “praying for rain is not a sustainable strategy”.

Unfortunately, unless there is a dramatic return to the very wet times experienced in the 1950s and 1970s, considerable triage both of irrigation and environmental assets will be necessary.

The reality is that if it remains dry we will not be able to keep all the environmental assets, all the irrigation assets and all the river going. Working out what to let go and what to keep is never enjoyable, but is an essential discussion to start

Professor Mike Young is a Fellow of the Academy of Social Sciences and holds the Research Chair in Water Economics and Management at the University of Adelaide. He is a member of the Wentworth Group of Concerned Scientists and was awarded a Centenary medal for contributions to Environmental Economics and the Land and Water Australia Eureka Prize for Water Research. He established CSIRO’s Policy and Economic Research Unit and is a member of the SA Government’s Sustainability Roundtable.