18 January 2011

Large Balls

[Image from The Australian, see the original here in PDF]

Wivenhoe Dam near Brisbane, Australia is at the center of controversy in its role in the recent flood. The dam, as is commonly the case, is expected to serve two seemingly contradictory functions.  On the one hand it is a buffer against drought, meaning that it is desirable to keep it more full in the eventuality of low precipitation.  On the other hand, the dam is a buffer against floods, meaning that it is desirable to keep it more empty in the eventuality of heavy precipitation.  Since keeping the reservoir full and empty are not simultaneously possible, it then is necessary to balance these objectives.  Since future precipitation is uncertain, the dam's management is thus a matter of decision making under uncertainty (where risks are known) and ignorance (where they are not).

The Queensland government has initiated a high-level investigation of the dam's management during the flood.  It will focus on the decisions related to storage and release.  According to The Australian, publicly available evidence points to the dam management as a key factor in the magnitude of the Brisbane flood:
More than 80 per cent of the flood in the Brisbane River at its peak last Thursday was the direct result of the release from Wivenhoe, the city's flood shield, of up to 30 per cent of its capacity, according to official data obtained by The Australian. The data shows that, without the unprecedented and massive release at a peak rate of 645,000 megalitres a day from the dam on Tuesday, January 11, the flooding in Brisbane would have been minimal.
Andrew Dragun, an adjunct professor in economics at the Australian Rivers Institute, Griffith University and editor of the International Journal of Water, critqued the management of Wivenhoe as follows:
In the days before the flood, the BOM warned of an upper level low pressure system dumping a large amount of rain over southeast Queensland. The warning came late in the first week of January and was visible on the BOM interactive weather and wave forecast maps. Ironically, the system slowed off the coast for a few days, giving operators plenty of time to make any adjustments to capacity levels. But the gates remained shut on the 100 per cent capacity.

As the low system dumped rain, the operator opened the gates, releasing about 116,000 megalitres on Friday-Saturday, with releases of 100,000ML over the next two days. Despite these releases the dam level rose to 148 per cent by Monday last week.

By Tuesday, with the dam at 176 per cent, the operator released a phenomenal 645,000ML. The result was bound to be bad. Significant flooding and tears, all while Dannien expressed his "confidence . . . that everything happened the right way".

Consequently, the flood peaked in Brisbane on Wednesday at 4.46m. However, the inflows from the catchment were surging, and the dam reached a capacity of 191 per cent.

Anna Bligh admitted that the operators nearly lost control, with water only 90cm from spillway fuse plugs. If triggered, the plugs would have released a torrent of water to the system. The results could have been catastrophic.
Why didn't the dam operators empty the reservoir further in the days and weeks before the storm?

There appear to be at least two important  reasons. One has to do with the psychology of decision making:
Retired engineer Ian Chalmers, a key project supervisor in the construction of Wivenhoe Dam between 1977-85, defended the decisions of the operators in the past week, adding they will do a better job next time.

"These questions are all valid, but put it this way - you would have to have very large balls to [significantly reduce the dam's volumes in the months after the weather warnings] after 10 years of drought, because if you had got it wrong you would be accused of wasting the water," Chalmers said.
In other words, after 10 years of drought it would have been very easy to see a reservoir full of water not so much as a risk of flooding but instead as a buffer to risk of drought.  As well, imagine the outcry if the flood had not occurred and in the next prolonged dry spell, observers point back to all that water was allowed to flow downstream.  Chalmers is suggesting that it is understandable that the dam's operators erred on the side of protecting against drought rather than floods.

Of course, the gets us back to the dual purposes of the dam, and the need to trade off risks of drought with risks of floods.  The only way to make that trade off go away it to have enough storage capacity to make that trade-off go away.

The second reason for not emptying the dam is that according to Professor Dragun the dam operators did not consider the state of the ENSO cycle, which has profound impacts on the Australian climate:
What have the Wivenhoe Dam operators been doing for the past couple of months? According to SEQ Water Grid chief Barry Dennien, dam levels were managed according to the rules and strictly by the operating manual. Dennien is comfortable that "everything happened the right way".

It seems the manual and the operator do not differentiate between the weather outlook of an El Nino (dry drought) and a La Nina (rain, flooding). After the drought, Wivenhoe reached 96 per cent of its supply capacity on March 16, 2010, and has been maintained at that level or higher since.
If it is in fact the case that Wivenhoe is managed without regard to ENSO, then this would be a case of decision making under willful ignorance, rather than decision making under uncertainty, as the ENSO signal is extremely strong in Australia and has a demonstrable influence on the probabilities of extreme (high and low) precipitation.  Thus, when Wivenhoe Dam operators say that they did everything by the book, they may indeed be correct, but at the same time "the book" may have led them astray.

While the Queensland flood inquiry will focus on hydrology and the dam's management, there will be deeper issues here of decision making under uncertainty and ignorance, and how such decisions should be made in the future.


  1. The ENSO cycle is a good starting point to set a default level the dam should be maintained at. For shorter term decision making, why did the managers not utilize the BOM's forecasts? Unlike climate predictions, weather forecasts are made for timescales where skill can be measured. I don't have numbers for Australia handy, but the US GFS model in the N hemisphere has shown 5-day forecast skill for over 20 years. In recent years it is approaching showing skill up to 8 days out today. Unless the BOM and Australian forecasting is much worse off, it doesn't seem like it would take large balls to use their 5-day forecast over the average climate for the last decade. It would take much larger balls to use one's gut feeling over the predictions of experts that can be unfairly blamed if they are wrong.

  2. Reminds me a bit of the Glen Canyon Dam on the Colorado which was almost undercut because of a policy of keeping the dam as full as possible combined with heavy precipitation and a late runoff in 1983:

    All the early flood flow in 1983 was contained behind Hoover because, unlike Powell, Mead Reservoir is required to maintain at least five million acre feet of flood storage space. But as the flood persisted, the Lower Basin bosses at Boulder, Nevada, were obliged to open the spillways on their old art deco dam. Lowland flooding occurred from below Bullhead City to beyond the Parker Strip as the discharges from Hoover rose to over 40,000 cfs.

  3. "In other words, after 10 years of drought it would have been very easy to see a reservoir full of water not so much as a risk of flooding but instead as a buffer to risk of drought."

    or another view . . .

    In other words, after 10 years of unrelenting hysteria and ridiculous threats over global Warming, it would have been very easy to see a reservoir full of water not so much as a risk of flooding but instead as a buffer to risk of drought.

  4. @Fred:

    Thoughts about climate change may or may not have played a role here - who knows at this point? - but it's really not necessary for them to have.

    I'd think most people who have lived downriver from a reservoir, particularly in a drought-prone area and/or where a reservoir's capacity is basically only good for one season, would be familiar with this dilemma, and it's been around for longer than climate change has been in most people's consciousness.

    Complicating things, at least here in the western US, agricultural interests are very influential in this sphere, and while I don't think they're completely uncaring about homeowners' concerns, they'd certainly like to err on the side of a full reservoir. And much like is noted in the article and Andy's comment above, I remember my parents pulling their hair out on more than one occasion in the 80's and early 90's when, in murky sets of circumstances, the Army Corps (of Engineers) fell back on hard-coded regulations rather than approaching the situation more critically.

  5. Kmye . . . the belief in AGW was the driving element in the State changing policy about the dam useage.

    It is well documented, the State Premier said so on TV as late as last week . . . although sh is probably savvy enough to not repeat her statements.

    Since many insurance companies are refusing to pay out claims of policy holders in Brisbane because they claim the floods were an Act of God, the implications of knowing the floods were the result of political policies and human actions will be huge.

  6. -5-Fred

    If the dam was indeed not being managed for ENSO, I find it hard to believe that it was being managed for climate change. Please do share the documentation that you refer to ... Thanks.

  7. We've had quite an extensive discussion of this over the past week or so over at Bishop Hill, where quite a bit of information has been provided by various posters. See:
    plus earlier posts.

    The original intention was to remove restriction on water consumption once the water supply part of the storage (about 1.15ML) reached 60%. But then the Premier announced that water was so precious that it should not be wasted whatever level the dams were at (belief in warming-as-drought meme).

    The is some controversy over whether the additional 1.45ML of flood mitigation storage was managed adequately. It was allowed to reach 190% full and then releases were increased markedly at a time when some are even saying they caused the flood peak in Brisbane.

    For me the key question is less whether the operators followed the manual, but whether the manual had been rewritten after the commissioning of the Tugun desalination plant (which was completed and then mothballed) and the water network that allowed trade between storages. This should have changed markedly the tradeoff between the value of water in storage for consumption and the value of flood mitigation capacity. In choosing to mothball Tugun, the government effectively made a choice that avoided cost at Tugun had greater value than a marginal unit of flood mitigation.

    From a decision-making perspective, SEQ Water is controlled by both the state government and about a dozen loacl government authorities. Brisbane City Council (the only one to gain from flood mitigation) has only 45% control. Other than 20% by the state government, the rest all have an interest only in water supply, so the Premier's statement is k=crucial. ONe for the public choice theorists, perhaps.

  8. but the US GFS model in the N hemisphere has shown 5-day forecast skill for over 20 years. In recent years it is approaching showing skill up to 8 days out today. Unless the BOM and Australian forecasting is much worse off, it doesn't seem like it would take large balls to use their 5-day forecast over the average climate for the last decade

    My understanding is that Nth Hemisphere weather forecast are signifigantly better due to vastly more data inputs in the Nth hemisphere. The south is mostly ocean and we have much less in the way of reliable data inputs.

  9. I don't know whether it was Warren Buffet who said it, but the statement "forecasting by looking in the rear view mirror" was made about how people perceive likely investment returns.

    A poll around the late 1990's in the USA showed th expectation of annual return on investment was 18%.

    In the late 1970's Time magazine had a cover "Are equities dead?"

    Investment returns were very poor from the early 1960's to the late 1970's so "everyone" believed equities were a lousy investment.

    Investment returns were excellent from the early 80s to the late 90's so "everyone" believed equities were a fantastic investment.

    Part of the story is that people's expectations can change equity pricing - unlike the weather - but more importantly the plain obvious facts get pushed to one side by what people see in the rear-view mirror.

    In the case of equities, interest rates soared from the early 60s to the late 70s, then dived from the early 80s to the late 90s. It's a plain obvious fact that the interest rate changes had a major effect on equity pricing.

    Except for everyone looking in the rear-view mirror.

    Politicians get judged by people.

    What's a politician likely to choose for a dam storage policy after many years of drought?

  10. The 60% figure is of interest.

    I understand from Aynsley (above) that water restrictions were meant to be lifted in SE Queensland if the storage capacity of the dams exceeded 60%.

    However in the mothballing of the Tugun desalination plant Queensland Natural Resources Minister, Stephen Robertson stated, "It (Tugun)will return to full-time operation if the region's dam capacity drops to 60 per cent."

    It appears that Queensland authorities have been caught in two minds.

    1. No matter the level of storage or capacity of the water grid system in place every drop of water was said to be precious and should not be wasted.

    2. The water grid had enough redundancy, dams and desalination plants, in the system to lift water restrictions completely in SE Queensland.

    There seems to be two competing themes that resulted in this double thinking, namely the fear of a 10 year drought being due to AGW and the impact of rising water charges - 1 litre of desalinated water is 12 times more costly than 1 litre of dam water.

    This suggests that the 60% figure is erroneous for determining the lifting of water restrictions and/or the switching on of desalination plants. This points out that SEQ Water could not have been following the manual on water storage.

    If they were not following the manual on water storage how likely is it that SEQ Water were fully prepared to deal with flood mitigation despite warnings on the onset and impact of La Nina and local weather forecasts of downpours.

  11. It also transpires that the SE Queensland water grid of desalination and recycling pants would all be fully operational if the storage capacity declined below 40%.

    So the water grid system appears to be designed to maintain a minimal storage capacity of 40%, where by 40 to 60% capacity means water restrictions, above 60% means the lifting of water restrictions, above 100% means the draining of capacity as part of flood mitigation. Is that what it states in the manual?

  12. I'll be off-line for a couple of days, but I'll leave you with a further reference:
    Apparently an engineer warned of the operating rules a decade ago. That would be valid in spades once water security was greater with desal and network.
    As someone has pointed out, the state government's climate change report for SE Queensland mentioned 'drought' 24 times and 'flood' not once.

  13. It is an earthen dam, and its storage components were likely determined when it was built. It appears they added an additional spillway in the 2000s, which indicates they were very concerned with flood mitigation. Looking at a photograph of the spillway, the floodgates are mounted, as expected, near the top of the dam. Funny how that works. It drains to the bottom of the floodgate, and then guess what happens? Nothing. The earthen tub below likely has a very limited capability of being quickly drained. So in a year in which it is raining, getting the lower 1.16 million ML out of tub would not be easy as water intake from the basin would likely offset the meager amounts they could pipe out through power generation tunnels and water supply piping.

    If the dam's initial purpose was flood mitigation, why on earth would they place the gates at the top of the dam, and leave 1.16 million ML of water without any means of rapid release? You have to have tremendous release capacity to get water out of a dam when it is raining in the basin. Makes no sense. The Wivenhoe was multi-purpose from birth.

    Face it, they planned for 1.45 million ML of flood mitigation capacity, and they had it available. Once into 2010, there was no time to alter the dam, and "we guess rainfall might exceed Wivenhoe's 1974-era flood design" would hardly qualify as a reason to blow a hole through the impermeable base (genius move alert) for a lower lying floodgate. Lite the dynamite and holy crap, a Toowoomba-level tsunami arrives Wivenhoe basin without notice! Then you would have a Royal Inquiry with some real work to do.

  14. The Wivenhoe Dam is useless for water storage. It is a very large very shallow dam with over 2000mm annual evaporation. More importantly the dam will only fill during periods of very heavy rain. Steady rain of 100mm/month will not increase the levels.

    The other problem is that the dam will go from empty to overflowing with only 300mm of rain over a week.

    So the only way to safely operate Wivenhoe is to make sure water levels are kept as low as possible during the wet season (November-May).

  15. Re: JCH

    Wivenhoe was never intended as a storage dam. It was built purely for mitigation immediately after the devastating floods of 1974. It only fills once every 5-10 years due to erratic rainfall and very high evaporation.

    Wolfdene Dam on the Albert River was intended to be Brisbane's main water supply. It was scrapped by former prime Minister Kevin Rudd when he was a senior Queensland public servant.

    The floodgates aren't at the top of the dam. They open to the bottom of the dam wall as clearly shown in this photo. http://commons.wikimedia.org/wiki/File:Wivenhoe_dam_wall.jpg

  16. re: ScottGA.

    Yesterday morning the BOM predicted 5-10mm of rain for Brisbane. We got 50mm.

    Weather forecasting for South East Queensland is basically just guesswork. Conditions can change radically within a few hours.

  17. The earthen wall of the Wivenhoe is 160 feet tall. The flood gates are at the top, and are 54.4 feet tall. That means the mud tub of the earthen dam is more than 100 feet deep. The entire area below the bottom of the floodgate is called storage. When the water behind the mud wall reaches the bottom of the spillway entrances, whether gated or not, it has reached 100% of its drinking water storage. That is the function of the lower part of most earthen dams, and that includes the Wivenhoe. There are no holes in it large enough to allow rapid release of its contents, especially when it is raining in the basin, so that portion of the dam's capacity is by definition, not available for flood mitigation in what dam folks here in the states call the "Noah Flood". All flood mitigation, including for the Noah, is handled by the upper-levels of a dam as there are massive release gates.

    The gates are at the top of the Wivenhoe dam. It was built that way, so it was built to store 1.16 million ML of drinking water as every drop of the 1.16 million ML is below the bottom of the gates and the auxiliary spillway. It's in the mud tub, and getting it out of there in a hurry is not possible in a rainy season. When it's been raining hard in the basin, it's impossible: water intake faster than water outflow. Remember that water continues to aggressively flow into a dam even on the sunny days of a rainy season.

  18. Embankment overtopping protections system and earth dam spillways

    Chanson, Hubert (2009). Embankment overtopping protections system and earth dam spillways. In Walter P. Hayes and Michael C. Barnes (Ed.), Dams: Impacts, Stability and Design (pp. 101-132) Hauppauge NY, USA: Nova Science Publishers.

    ABSTRACT: A common type of water storage structures is the embankment reservoir and levees. But, during the 19th century, numerous embankment dams failed in Europe and North-America (Table 1-1, Fig. 1-1). The two most common causes of failures were dam overtopping and cracking in the earthfill. The former was often caused by inadequate spillway facility. The latter resulted from a combination of bad understanding of basic soil mechanics, poor construction standards, and piping at the connection between bottom outlet and earth material. In recent years, the design floods of a number of dams were re-evaluated and the revised flows were often larger than those used for the original designs. In many cases, the occurrence of the revised design floods would result in dam overtopping because of the insufficient storage and spillway capacity of the existing reservoir. In recent years, a number of overtopping protection systems were developed for embankments and earthfill dams. These include concrete overtopping protection systems, timber cribs, sheet-piles, riprap and gabions, reinforced earth, Minimum Energy Loss weirs, embankment overflow stepped spillways and the precast concrete block protection systems developed by the Russian engineers. After a brief discussion of the embankment overtopping breach process, several overflow protection systems are presented and detailed in the next paragraphs.

  19. Yes Malcolm, when I was in college I had a summer very close by to a small earthen dam that was overtopped during a deluge. The combination of the dam's sudden outflow and the extreme rainfall killed more people than were killed by the January rain events in Queensland.

  20. Hi Roger and fellow commentators,

    Like Rogers psych analysis a lot.

    Would be interested in your further analysis in the setting that the dam is back to 100%, and lots more rain/cyclones possible flooding ahead. What pressures on operators now? ... they have "hindsight" and everyone is watching.

    The problem is not over by a long shot.

    Andrew Dragun

  21. Since Wivenhoe is a hydro-electric facility, the five flood control gates cannot be the only non-emergency drains in the dam.
    Additionally, if the large flood control gates were the only water exit, the river would dry up in low water periods.

  22. As I understand it, the Wivenhoe is actually an auxiliary water source for power generation. There is a dam located above it in the hills, and they pump water uphill to that dam when water is needed there.

    I did read about a second generating facility, and it may be downhill. Don't know.

  23. Belated observations. Some commenters are badly misinformed.

    Wivenhoe was always intended as a water supply facility with a a flood mitigation function. History has proven that it can and does serve these two functions.Misdirection occurs when people come to believe that the dam can contain and neutralise all projected flood events .It cannot,and its designers and builders always acknowledged this.

    Any claims that lowering the dam below 1165GL/100% for extra mitigation is the way forward have to realise that buffer volumes gained,while significant,are not huge in terms of potential inflow volumes and speeds [10% is only 116.5GL] and loss of no failure yield is a significant consideration in a dam that supplies the great bulk of SE Queensland's drinking water.

    The spillway arrangement was upgraded recently to meet the new specifications of a nationwide review [ANCOLD] of large dams and their ability to cope with projected floods.

    The gates do NOT extend to the bottom of the dam,they extend down to EL57m/450GL/40% full level.Water CAN be removed quickly below the EL67m/100% full/1165GL water supply maximum level of EL67m,as the 10m of capacity below 67m is held behind movable gates. Below EL57m water removal is slow,via hydro station and pipe. The dam also has a pump storage facility east of the lake.using a small upper reservoir on Splityard Creek

    Wolffdene dam was NEVER intended to be Brisbane's primary water supply,had a catchment size 10% of Wivenhoe's and was a poor,shallow option. The proposals on the Albert River that superseded the Wolffdene idea and got closer to development were further upstream beyond valuable agricultural land,were smaller in size and yield and still failed to meet cost/benefit targets. Wolffdene is a political myth raised when convenient to attack the Labor party which quite properly canned the proposal.

  24. Again belatedly.

    "Why didn't the dam operators empty the reservoir further in the days and weeks before the storm?"

    Weeks out is unforecastable,though the general weather set-up was known to be wet season and wet season-La Nina. Does not guarantee rainfall,let alone flood generating rainfall. The dam has a generous 1450GL set aside as a flood buffer,more than that reserved for water supply. If policy is to change ,how much extra buffer will be factored in? 5%/58GL,10%/116GL,20%/233GL? Ir would have to be a t least the latter figure to be useful in an event as big as 1974,which may be only a 1 in 50 year flood. This is THE major water supply for nearly 2,500,000 people,and changing policy is a big step. Big Balls,but more realistically ,a lot of discussion. More than a few months.

    Days out,the dam had just been through a lesser flood event at the end of December,and had just been relieved of 250GL of flood water. The days before the heavy rain were not that wet,and while heavier rain was forecast,quantities of 200-300mm/day were not immediately envisaged. Nobody knows the hour to hour information stream that the operators experienced ,yet many are certain they would have chosen to lower the dam earlier than the engineers.

    Really,why wasn't the dam raised in line with detailed discussion papers in the governments possession for several years? A 4m raising,would preserve the vital water supply quotient,while winning another 500 or so gigalitres of flood mitigation. There are another million people downstream,of course not all in the immediate path of a flood,since the dam was built in 1985. This is a failure of decision making at the highest level.

  25. Nick said:

    "The days before the heavy rain were not that wet,"

    They were wet enough to force the operators to allow the dam to rise to 115%. They were forced to do this because the manual required them to keep bridges below the dam open while the level was below 115%. That's 15% that was lost to its main flood mitigation task, mainly just to avoid the cost of building high level bridges below the dam. A great cost-saving strategy that turned out to be.

    "The dam has a generous 1450GL set aside as a flood buffer,more than that reserved for water supply."

    Releases become uncontrolled once the dam exceeds 73.5 m or 835 GL of flood buffer. The rest is for dam protection because the dam will fail catastrophically once the flood buffer exceeds 1450 GL. Once you take off the 175 GL reserve that is given away for the bridges below the dam, there is only 660 GL of real flood buffer left. Not quite as generous as it first appears.