Australasian Science: Australia's authority on science since 1938

An Irrigation Channel Too Far

The confluence of the Murray and Darling rivers at Wentworth, NSW.

The confluence of the Murray and Darling rivers at Wentworth, NSW.

By Ian Lowe

How much will reduced water allocations in the Murray–Darling Basin really hurt regional communities?

The media has been full of confected outrage about the Murray–Darling Basin Plan even though we have known for at least 15 years that the Basin was in deep trouble as a direct result of over-allocation of water for irrigation. The 1996 State of the Environment report noted that the approved extraction was 80% of the median annual flow. Clearly, then, water allocations have to be reduced.

During the recent drought, much less water was made available and the towns along the rivers survived, although some were doing it tough. Like energy-intensive industry, which has known since the Kyoto conference in 1997 that Australia has to curb its greenhouse gas emissions, the irrigators have known for at least a decade that the water allocations need to be reduced.

Some industrialists are now saying that we should not rush into any scheme to price carbon dioxide, lest it curb the growth of polluting industry. So it was not surprising that the announcement of the Murray–Darling Basin Plan, proposing reductions that are probably not enough to restore the health of the Basin, led to a flurry of media reports about the alleged economic damage, with dramatic claims that whole towns would die. The clear implication was that using less water must cause a proportionate drop in production and hence employment, with flow-on impacts for the whole community.

Illustrating the television reports were video clips of water flowing along channels or being sprayed into the air, and I recalled travelling down the Murrumbidgee basin earlier this year. In an arid zone where the daytime temperatures were above 40°C, I saw the precious river water flowing in shallow open channels, ensuring huge evaporation losses. I also saw gantries spraying the water into the air, with some hitting the ploughed land but some flowing onto the road, lots evaporating, and relatively small amounts soaking into the land and reaching the roots of the crops.

By contrast, downstream in the riverland, I saw many of the irrigators using advanced technologies like pipe and drip irrigation. Here they were getting much better value from the water being used.

So before we accept that saving the river must cause economic mayhem, I would like to see serious discussion of whether more effective use could be made of the water. The Romans developed the pipe more than 2000 years ago, so it has had a fair field trial.

We might also be discussing whether it makes sense to grow crops like rice that require flooding of land in the arid zone – but that’s another story.


The Australian university system is gearing up for research assessment. While quantitative indicators can be used to disprove common furphies, they are almost inevitably an over-simplification.

You can count papers published, but nobody believes that every paper is equal. Sometimes a paper can revolutionise a whole field but, at the other extreme, about half of all papers are never cited by any other researcher, and they sink without trace.

So you should make some effort to measure impact. Citations can be counted, but the accuracy of the result depends on the coverage of the search engine. And there are systematic problems, such as the tendency of new techniques to be extensively cited and therefore receive an inflated impact assessment. Still, at least the approach is measuring outputs, however imperfectly.

I recall a silly proposal to rank researchers according to their grant income. This would be like giving the Melbourne Cup to the horse that ate the most oats! The logic was that grant income was some sort of measure of peer esteem, but it clearly discriminated in favour of researchers whose projects required large equipment items or many person-hours of paid help. Theoreticians whose work needed only a pad and time to think would not be ranked highly by that system.

The bottom line is that quantitative indicators are helpful within a defined sub-field, but need to be used with caution. They certainly don’t help make judgements about the relative strength within an institution of, say, theoretical physics, microbiology and chemical engineering.

Ian Lowe is Emeritus Professor of science, technology and society at Griffith University.