Home > Oz Politics > Marohasy vs the Murray

Marohasy vs the Murray

April 24th, 2004

Following up on my previous posts, I thought it might be useful to summarise my objections to the claims put forward by Jennifer Marohasy of the IPA, supporting the conclusion that scientists have “greatly exaggerated their claims that the Murray River’s health was declining.” You can read her full paper here (1.5 Mb PDF).

As I read it, Marohasy has three main claims. The first and most important relates to salinity trends. There appears to be agreement between Marohasy and the scientists working for the Murray-Darling Basin Commission (among other scientists working in the field) that salinity was generally getting worse until the 1980s, and that the Salinity and Drainage Strategy adopted by the MDBC at least partially reversed this increase. The disagreement comes with respect to projections for the future. As shown below the MDBC regards the Salinity and Drainage Strategy as providing breathing room rather than a permanent fix.
Murray
Marohasy says that the scientists have got all this wrong. She doesn’t appear to offer any real scientific basis for this claim. Rather she asserts (in comments on this blog) that the scientists working for MDBC, CSIRO, ABS and other key research institutions are “environmental activists masquerading as scientists” and (in her report) that they are committed to “promoting the myth of an ecological disaster” or are lying in order to improve their chances of getting research funding (p5). Similar points have been made, even more strongly, in The Land for which Marohasy writes. Reader David Nash has sent me this reply to one of Marohasy’s pieces attacking scientists.

Land.20040422p18

Unless you’re prepared to redo thirty years of scientific research yourself, the debate on this point comes down to a pure question of comparative credibility. I don’t think scientists or government agencies are perfect by any means, but the kind of large-scale conspiracy theory required to explain the absence of any significant scientific support for the IPA position is implausible. On the other side of the ledger, I’ve already stated my reasons not to give any credence to anything said by the IPA.

Marohasy’s second point relates to the use of median rather than mean measures of river flows. As she correctly observes, the use of means rather than medians would make things look better in many cases. As she does not inform her readers, any elementary statistical text will tell you that, where means and medians differ, the median is generally more representative. I think this is clearly true in relation to river flows.

The final set of points made by Marohasy are a set of isolated pieces of information supportive of a favorable assessment. There is no serious attempt at a general assessment of trends, This kind of pointscoring is essentially useless, except as an answer the kind of caricature opponent who believes that everything in the system is getting uniformly worse. No serious scientist or policymaker working on the Murray-Darling believes this, though some popular discussion tends that way.

As I’ve mentioned, I’ve done a good deal of work on this topic myself. You can read a fairly recent (2001) survey article here (PDF). It’s mainly concerned with economic theory and policy, but it naturally draws on the scientific evidence. I think anyone who reads it will search in vain for the kind of doomsday mentality that Marohasy suggests is driving the policy debate. Nevertheless, there’s pretty good evidence that the system, like other irrigation systems in the mature phase, is overstressed and needs, among other things a reduction in irrigation diversions.

Categories: Oz Politics Tags:
  1. Tony Healy
    April 24th, 2004 at 14:24 | #1

    I was astonished when I first read Mahoney’s claims about the Murray River, because they conflict so strongly with my personal experience of having grown up near the river.

    In the 1970s it was possible to see fish swimming under the surface of the river, and it was possible to see parts of submerged rocks. That’s not possible now. The turbidity that’s always present in the water now used to occur only when the river was in flood, about once a year.

    Second, the magnificent Hattah Lakes in north west Victoria spent most of the 90′s without any water at all, which is historically unusual. The lakes now have reasonably sized trees growing across their beds. Those lakes used to fill each year when the river flooded, but have been denied that water since about 1992.

  2. Tony Healy
    April 24th, 2004 at 16:44 | #2

    That should be Marohasy.

  3. Harry Clarke
    April 24th, 2004 at 20:36 | #3

    Australia has low (and irregular) rainfall and high evaporation. Its rivers contain low and variable amounts of water and since we don’t have high mountain ranges like the Rockies or the Alps, not much water gets naturally stored as snow that can melt and be used in sumnmer. Hence the need for non-natural storage,

    The salinity problems of the MDB probably existed prior to white settlement and it is likely that, while aboriginal burning kept the countryside in western NSW open, it raised salinity issues long before Europeans arrived.

    Salinity in the MDB have improved since 1982. Setting salinity quotas and/or transferable quotas subject to taxes reflecting cumulative downstream salinity costs can deal with the remaining salinity issues.

    I think there is some ‘scientific interest group hysteria’ over the MDB issue. Of course the stakes are high given the role the MDB plays in Australian agricultural production. But the environmental costs that arise in the MDB are manageable if water users pay the full costs of the environmental damage imposed. Achieving this, though difficult politically and perhaps technically (given non-point issues) will get done. I am optimistic.

  4. John Quiggin
    April 24th, 2004 at 22:14 | #4

    I’m optimistic too, and agree in broad terms about the necessary steps.

    But I think it’s clear that the sustainable level of quotas is below the current allocation – the IPA is trying to argue that no reduction is needed.

  5. Harry Clarke
    April 25th, 2004 at 22:36 | #5

    John, what’s meant by ‘clear’ here in relation to the quotas issue? Do you mean: In a wide enough ball of information we should be well-enough informed to see the self-evident truth of the proposition you advance i.e. that quotas should be lower?

    I don’t see it though maybe my personal ball is insufficiently wide. You don’t provide evidence that bears on it (nor does Gary Jones) and Jennifer Marohasy advances the counter hypothesis that (i) the data on upstream salinity is at best inconclusive and (ii) the evidence on rising watertables suggests things are not as bad as is claimed. It should be possible, in principle, to show these claims are wrong or irrelevant without relying on ‘consensus disagrees’ arguments.

    And the main thrust of your counter-argument is that many well-informed scientists agree things are bad.

    I could counter that most university leaders in Australia believe that the competitive ‘dumbing-down’ model of tertiary education is the way we should see the future of Australian universities (a proposition you disagree with) and, what is more, they have had the backing of successive Labor and Liberal governments as well as the Mr Nortons of this world who seek to push things further.

    Maybe Jennifer is wrong but the ‘consensus’ argument you put is unconvincing. Her arguments need to be answered.

  6. John Quiggin
    April 26th, 2004 at 22:56 | #6

    Harry, I honestly don’t see how public debate can function unless, at some point, you’re willing to accept expert judgements and to reject the claims of non-experts with an obvious axe to grind. I’d suggest you look at the piece on the ozone layer by Baliunas that I linked to here. It’s every bit as plausible as Marohasy on the Murray and uses many of the same techniques, but I assume you would agree that it’s completely wrong.

    I’ve pointed to some obviously misleading features of Marohasy’s analysis, such as her discussion of means and medians and her cherrypicking of evidence. I also have a number of reasons for thinking the consensus position to be plausible (for example, irrigation systems worldwide have been overallocated more often than not, and merely halting growth in resource extraction at a high level rarely prevents decline) but without detailed understanding of the hydrology (which neither you, nor I nor Marohasy possesses) it’s impossible to make a final judgement.

    How then, do you suggest we proceed?

  7. Harry Clarke
    April 27th, 2004 at 16:00 | #7

    Well I suggested scrutinising i) data on upstream salinity and (ii) evidence on rising watertables. Do you need to be a hydrologist to do this? (Serious question, not being ironic).

    You have been exploring this line several times on your blog since I began reading it. Will follow the general idea with interest.

  8. John Quiggin
    April 28th, 2004 at 12:19 | #8

    ‘Well I suggested scrutinising i) data on upstream salinity and (ii) evidence on rising watertables. Do you need to be a hydrologist to do this? (Serious question, not being ironic)”

    Harry, you can see the figure I posted, showing the MDBC projection. Marohasy prints the underlying data on an annual basis, which is all over the place as you can imagine and (implicitly) fits a horizontal trend through it.

    As far as I can see, you do need to be a hydrologist (or accept the judgement of hydrologists) to determine which is right. The crucial question, as I said, is whether the Salinity and Drainage strategy represents a permanent cure (Marohasy’s implicit position) or a temporary fix (the view of the people who actually implemented it). I don’t think eyeballing the data, or fitting an ARMA model to it for that matter, will help resolve this.

  9. Harry Clarke
    April 28th, 2004 at 14:13 | #9

    The ‘recorded salinity’ line and the ‘S&G’ strategy are partly projections not ‘records’ in this graph so it is a bit confusing.

    Yes, mechanical extrapolation tells you almost nothing (does it ever?) so I guess its whether land use becomes more intensive and/or whether long-term lagged effects mean that recent improvements will be undone. I can well believe that knowledge of hydrology helps answer the latter question.

Comments are closed.