This post, written in the immediate aftermath of the floods (and the subject of some controversy) is looking pretty good in the light of the recent decision to release water from Wivenhoe Dam, with a target of 75 per cent.
Having had some time for reflection, the obvious modification to my initial position is that we shouldn’t have a fixed target, but rather should take account of the seasonal pattern of rainfall and, to the extent that this is possible, of the El Nino/La Nina/SO cycle. One way to do this would be to set targets for the beginning and end of the wet season, designed to be consistent with expected rainfall and usage for the wet and dry seasons. As I mentioned in my previous post, the flexibility associated with desal and recycling plants and the Water Grid would make this kind of management much easier than in the past.
Looking back at the controversy this post aroused, it’s clear that it was due in part to the involvement of The Australian and the anti-science lobby on climate change, which, for reasons that remain obscure to me, decided to run with the line that early release of water from Wivenhoe Dam would have greatly reduced the severity of the floods. It’s interesting to find that being in partial agreement with the Oz is even worse than being attacked by them. The Oz presentation of news on the dam management policy, as on all issues where it decides to push a line, was so selective and skewed that it even relatively simple issues, like the proportion of floodwater that came from the Bremer river, became hopelessly confused.
So, to clarify, both my original post and this one refer to policy options for the future, which might be adopted in the light of the floods, and of the likelihood of more extreme climate events in the future. With perfect hindsight, and discretion unfettered by a rulebook, managers would certainly have made different decisions in the days leading up to the flood. But what is really needed is a long-term change to the management procedures that set 100 per cent of water supply capacity as a fixed target. IIRC, the only calls for such a change in the leadup to this wet season were from those objecting to the release of water, and implicitly calling for a higher target.
John Quiggin 
I wouldn’t take the Qld State Labor Govt’s agreeing with one as a sign that one was right; quite the contrary in fact. This is a bad, knee-jerk reaction if the decision is for a firm 75% limit. (There might be a case for the alternative, namely varying target capacity in a band, say 85% to 115%, depending on rainfall projections.)
Should people who are wise enough, to build or purchase well above any possible flood zone, pay higher insurance AND higher water costs to subsidise those who choose to live beside rivers and in other known flood-risk areas? Should bad town planning by councils and developers be rewarded by further changing the goalposts in this manner?
Should we commit ourselves to higher energy use and greater greenhouse emissions (from coal fired power stations) to desalinate water when we could be storing that water from natural run-off? These higher green house emissions hold even greater future dangers for us.
This is a clear case of moral hazard. This is a poor outcome showing short-sighted stupidity and lack of civic and governmental will to tackle the real issues of poor land use decision making by consumers and developers.
I might add, this decision also means there is a higher probability of S.E. Qld running out of water entirely during the next el Nino event which is likely to be as severe and protracted as this la Nina is severely unstable and wet. Short-sighted fixes of the moral hazard variety always create greater problems for the future.
CSIRO and BOM have a joint project called the Seasonal StreamFlow Service
Obviously BOM run the service etc, while the CSIRO contribution was in the mathematical/statistical research necessary for this kind of service. I’m not at all certain what I can and can’t quote these days so rather than risk confidentiality (as if I have access to anything confidential in this area 🙂 ), I ‘ll just point to an article and anyone interested may google them and/or read what BOM have to say at the above link.
These guys have published more recently but this is the easiest reference I can find on topic (paywalls and such):
A Bayesian joint probability modeling approach for seasonal forecasting of streamflows at multiple sites,
Wang, Q. J., D. E. Robertson, and F. H. S. Chiew (2009),
Water Resour. Res., 45, W05407, doi:10.1029/2008WR007355
So there you have it, a shiny-bright new service of the sort that may help dam monitoring and release-to-forward-target scenarios. Of course in the current “Natural World” anything could happen to blow the technical skill into tiny little shards of Boltisms, for which that protagonist is paid staggeringly large amounts of cash. Unlike any researcher who actually puts their reputation on the line every time they publish such applied work. Or the BOM staff. Or the dam operators, more recently.
Just sayin’, you know?
Surely the reality to accept is that major dams in Queensland cannot be very effectively run as both water supply and flood mitigation devices at the same time. As many have noted these are incompatible aims and any compromise position reached (as was theoretically in place for Wivenhoe before the recent event) will be subject to revisionist criticism after the fact as either aim fails in the event. With our monsoon & cyclone driven rainfall situation flow volumes are immense in Qld east coast rivers. Dams would have to huge beyond belief e.g. much bigger then the 1.8 million ML of the Burdekin Falls Dam, the biggest in Qld, for any real flood mitigation downstream. Running our current dams at say 75% capacity will do little to mitigate floods in major events.
It would be good to see some modelling of rainfall anomalies – positive and negative, short and long term – tested against different dam management strategies so that the cost/benefit/risk propositions are on the table. I don’t know how well the seasonal predictions work in this area but if they help they could be included in the plan. Sooner or later there are going to be events that go the wrong way. When they do, it’s desirable that they aren’t accompanied with a load of if-only fantasies, posturing and knee movements.
On my skim of the dam operation manual there seemed to be too much grey area when a worst case model projection might have driven more water release earlier. (I guess.) Extreme events are will be outside anyone’s familiar territory so a concise book of rules is needed. I’ll be very interested to see what the enquiry comes up with.
I think the only rason the Australian was agreeing with you was because Labor is evil, all Labor governments are illegitimate and incompetent, and Anna Bligh should have been up at the dam control room herself, directing operations.
after all the grief and smelly mess,after years and years the dams are finally full and they want to throw that water away on an if,a but and a maybe?
strike me bluddy roan.
Back in early late December there were of course data that suggested increased rainfall would persist in Qld and northern NSW for the first 4 months of this year. The problem in Australia for both flood control and dams or desalination plants for fresh water provision is that nobody is putting all the data together into a meaningful water strategy for Australia.
In one of my blogs I did suggest that Julia Gillard could show some Statesmanship by setting up an Australia wide group to scientifically and methodically do just that.
Alas I was calling for the impossible. Whatever else Julia may be good for (and that’s not altogether easy to define) her skill set just does not include Statesmanship.
And so the data are largely ignored and remain more as a memento to the Peter Principle of cartography than a guide to proper strategy and policy development.
I agree with you John that it is embarrassing to be lined up (at least to some extent) with one of the campaigns that the Australian is running. But they run so many, I suppose at least occasionally they are going to be partly right. And on the Wheat Board bribery issue to their credit, they kept on digging, even though it was very embarrassing to Howard and Downer.
The Brisbane floods have caused immense damage, so it is necessary to review procedures to see if there are better ways of doing things in the future.
I suggest multiple changes would help.
First, as you suggest, run the dams at lower levels in wet seasons.
Second, take advantage of the high quality bureau of meteorology quantitative rainfall forecasts that are now available eg http://www.bom.gov.au/jsp/watl/rainfall/pme.jsp. The data from these forecasts should be run through various programs to get probabilistic estimates of inflow into the catchments. The release rules in the dam manuals should be modified to take account of this information.
Third, do some better modelling on the tradeoffs between extended minor flooding in the Brisbane River and a short period of major flooding. The existing manual is clearly informed by some of this sort of modelling, but I’m not convinced they have done it in enough detail.
The dam operators have to think about projected water use, projected rainfall in the dam catchment, rainfall outside the dam catchment that affects flows lower down, safe current levels, storage, maintenance cycles and much else. Models help, but this sort of situation is best handled with high levels of group expertise – and an attitude that let’s the experts get on with it unless there are major mistakes. Or we could write a prescriptive dam management manual, and when that’s done we could write the ones for brain surgery and winning the next war. There are limits to our conscious calculating abilities.
Peter @#8 said “In one of my blogs I did suggest that Julia Gillard could show some Statesmanship by setting up an Australia wide group to scientifically and methodically do just that.Alas I was calling for the impossible”. Ever heard of BoM and CSIRO? That is their job, or was, until they became slaves to the ruling party’s ideologies.
Peter T at 10. Unfortunately, on the evidence we have at present, it appears major mistakes were made on the Saturday, Sunday and Monday before the major release on the Tuesday. So letting the experts get on with it didn’t work in this case. Also for legal reasons one needs a reasonably prescriptive manual. That manual should be based on expert views of repeated modelling runs of the different scenarios the dam operators may face. So its a combination of expert views and modelling which gets incorporated into the manual and the procedures that are adopted during high risk periods.
“slaves to the ruling party’s ideologies”
Charlie, your insight is striking as usual. I well remember the sudden shift in the BOM/CSIRO position when Howard was defeated, not to mention the many changes as the Howard government went from pro-Kyoto to anti and back to somewhere in between.
Charlie at 11 and Peter at 8. Bom seems to have the job you are describing and have been given $450 million by Rudd to do it. Below is from http://www.bom.gov.au/water/about/waterRole/index.shtml
‘Water for the Future is designed to secure long-term water supply for all Australians. It includes the $450 million Improving Water Information Program administered by the Bureau of Meteorology (the Bureau) and backed by the Water Act 2007 and key stakeholders.
Under Improving Water Information, the Bureau will provide free web access to consistent, integrated national water information, suitable for a range of uses. …..Since 2007, the Bureau has employed an extra 120 people around Australia—mostly based in Melbourne and Canberra—to deliver water information.’
This could be a big mistake according to the analysis done by J. V. Hodgkinson. He points out that a very large fraction of SE Queensland’s water comes from “uncommon events” such as the one in January. As such it is most likely that the 25% they are planning to release will not be recovered any time soon and there is a good chance that this water will be missed sometime in the future. J. V. Hodgkinson’s position is that much more of Wivenhoe/Somerset should be used for flood mitigation but to maintain the amount of water supply storage, Borumba Dam (60 km from Somerset in the head of the Mary valley) should be greatly expanded to 2,000 GL. This would actually allow more water supply storage than Wivenhoe/Somerset even if they were entirely used for flood mitigation. He has been advised that 2-way pumps between Borumba and Wivenhoe/Somerset could transfer 4,000 ML/day compared with the 124 ML/day available from the Tugun desalination plant.
I also note that Wivenhoe dam operations are already violating the manual that applied in January. During the past week (on the 11th) they allowed the flow to approach 100 m3/sec, thus flooding Twin Bridges even though the dam level was below 67.25 m. The old manual only permits Twin Bridges to be flooded at dam levels above 67.25 m.
@Chris O’Neill
I think a small flood on the Lockyer on the 9th-10th,and local catchment storm runoff, may have been a major contributor to that ‘violation’ of the manual.
I agree with the ‘uncommon event’ theory,but find the idea of pumping water to an enlarged Borumba pretty bizarre,and energy-profligate.It’d be over 50km on the ground. Is it feasible to enlarge Borumba by that staggering [46GL to 2000GL] amount? Would it be fed by tunnel? Or over the range? Somerset is at 100m asl,and the crest of the Jimna Range into the Yabba Creek catchment is at around 500-600m.Sounds very expensive,though it may remove the need for Traveston in improving Sunshine Coast supply. But Traveston was designed as a dual purpose storage like Wivenhoe,and you won’t get much mitigation for Gympie out of simply controlling the 435km2 Yabba catchment.
Raising Wivenhoe by 4m is a much less expensive option.
This 75% thingy is just the government reminding us it is ‘decisive’,though the rationale mentions the de-sal and the availability of Wyaralong water ahead of schedule,so the risk of losing a years water supply is softened. I agree there is no great guarantee that the 25% will be easily replaced,even this year.
@Nick
That was back on the 9th at 11 am at Rifle Range Road. The peak at Savages Crossing didn’t happen until 5 am on the 11th or 42 hours later.
I think the basic idea is that any water directly involved requires a lot less energy than desalination. Why don’t you put your points to Hodgkinson?