Guarding against repetition

3 November 2011



The floods which devastated the Australian state of Queensland in January 2011 have been described as the worst in living memory. Dam owner and operator Seqwater played an integral role throughout the flood event, and has been key to the subsequent investigation which is now taking place. Suzanne Pritchard reports on how the Queensland flood inquiry is striving to ensure that history never repeats.


Its variable climate has earned Australia the accolade of being a true land of drought and floods. Throughout the country’s history the flood-drought cycle has been a natural part of life, and the recent flooding of December 2010 and January 2011 cannot be described as unique. Eight major floods were recorded in Australia in the 19th century; 23 in the 20th century and six in the first decade of the 21st alone.

Although not unique, the widespread flooding experienced across the states of Western Australia, Victoria, New South Wales and Queensland during 2010-11 were nonetheless devastating. They have been described as the most severe in living memory for Queensland, which is more used to coping with drought than flood.

Indeed from 2001-9 southeast Queensland had been affected by drought. The region’s water supply had been put at risk as the water levels at local dams were well below the full supply level for each structure. It was only when the combined storage capacity of Somerset, Wivenhoe and North Pine dams reached 60% on 20 May 2009 that the drought was declared over.

Ironically, just over 18 months later, the area was subjected to catastrophic flooding. State resources were strained and the consequences were shocking.

An area of Queensland, measuring greater than France and Germany combined, was declared a disaster zone and 35 people died. There was extensive damage to public and private property, over 29,000 homes and businesses suffered from some form of inundation, and more than 2.5M people were affected. The total cost to the Australian economy has been estimated at more than A$30B.

Record rainfall

Typically the Queensland wet season extends from October to April with the initial monsoon onset occurring in late December. However, the 2010/11 wet season was somewhat different.

In June 2010 the Australian Bureau of Meteorology warned that a La Nina event was likely to occur before the end of the year, which historically brings above average rainfall to most of Australia. Coupled with this event, Australia also experienced unusually persistent monsoonal rainfall which lasted longer than usual. Above average rainfall experienced throughout Queensland in late 2010 meant that many catchments were already very wet before the floods occurred. Such soaked catchments just could not absorb excess rain.

July to December 2010 was the wettest period on record for Australia, with December being the wettest on record for Queensland. The Bureau of Meteorology registered record flood peaks at over 100 Queensland river height stations during this period. In many locations the indications were that these were really the most severe floods in living memory.

The extent of the devastation caused Queensland state premier, Anna Bligh, to establish an independent commission of inquiry on 17th January 2011 to examine the unprecedented flooding disaster.

“We wanted to know that our dams were working properly and that is why the premier made a commitment that no stone would be left unturned and the commission of inquiry was established,” said Queensland Minister for Energy and Water Utilities Stephen Robertson. The government said it wanted ‘frank and fearless’ advice and would not back away from the commission’s findings.

Seqwater also welcomed the flood commission’s inquiry. “As the owner and operator of Wivenhoe, Somerset and North Pine dams, Seqwater had an integral part to play in managing and responding to the flood event,” the company’s chief executive officer Peter Borrows said in April 2011. “We welcome this opportunity to tell our story and make an important contribution to the inquiry’s outcome. Seqwater’s staff operated professionally throughout the event, managing a highly complex, rapidly changing and challenging situation, involving multiple variables and scenarios.

“I firmly believe we acted appropriately and in accordance with the flood operations manuals throughout the event,” Borrows added, saying that his company is committed to and will continue to support the commission’s process.

Interim report

On 1 August 2011 the Queensland Floods Commission of Inquiry was required to provide the state government with an interim report. This was to identify recommendations that can be realistically implemented before the next wet season begins, in order to provide safeguards should flooding occur again. It also deals, particularly where dam operations are concerned, with important work which should be started straight away even if it cannot be completed before the onset of rain. Delaying such work until the commission’s final report is published was not considered as an option.

The report states that it hasn’t attempted to catalogue every action undertaken in preparing, planning and responding to the 2010/11 floods, but has concentrated on ensuring an emergency response that will prevent the loss of life and property. Although it has strived to be as comprehensive as the strict time constraint allows, some of the issues raised may require further comment or elaboration in the final report.

“The state government will be in a much stronger position going into the next wet season,” said government minister Stephen Robertson. “The commission’s interim report will act as a blueprint to ensure that we are better prepared and work has already begun. We will provide the necessary support and resources required to implement these recommendations before the next wet season.”

The commission’s interim report examines a range of issues relating to flood preparedness. It also makes 175 recommendations focused on changes which can be implemented before the next wet season.

The report states that Seqwater should review all arrangements for the operation of its dams during flood events for the entire wet season by 30 September each year. The company must also ensure that all parties are adequately prepared in the process and that:

• Seqwater can comply with every aspect of the Wivenhoe and North Pine flood mitigation manuals.

• The flood operations centre is ready and capable of operating during any flood event of whatever duration, including in terms of communications, equipment, rostering of and facilities for staff.

• The flood operations centre must also have available all the tools, studies, equations and data necessary for it to be fully appraised of the consequences of its operations of the dams.

The commission went on to discuss Seqwater’s flood preparedness in some detail. “The prediction of the La Nina wet season by the Bureau of Meteorology (BOM) in October 2010 had (or should have had) implications for Seqwater, as an owner and operator of dams,” the report states.

However in its evidence presented to the commission in March 2011, Seqwater states that the rainfall forecast it receives from the bureau to assist in making operational decisions during flood events ‘are not sufficiently accurate to be used as the basis for making decisions on releasing flood water from dams. Currently a degree of uncertainty exists in all weather forecasts, particularly in quantitative forecasts, and the longer the forecast lead times the higher the degree of uncertainty in the forecast’.

Seqwater also claims that BOM quantitative forecasts for the dam catchments issued over the key days of the flood event underestimated the actual rainfall, which was between 160-340% of the available quantitative forecast for the dam catchment.

Indeed the commission did acknowledge the unpredictable nature of weather forecasting. The report said that it accepts that “primarily because of uncertainties associated with rainfall predictions, the achievement of an ideal strategy is possible only with the benefit of hindsight.”

It was with hindsight that the commission also called for a review of the flood mitigation manuals for Wivenhoe and North Pine dams. Flood engineers make operational decisions about dam releases on the basis of the relevant manual. Until 1 July 2011 government-owned bulk water infrastructure developer SunWater operated the flood operations centre for Wivenhoe, Somerset and North Pine dams on Seqwater’s behalf. As part of this contract Sunwater was required to review the Wivenhoe and North Pine manuals and recommend any improvements or confirm that the manuals remained satisfactory, the commission reported. However no evidence could be found suggesting that such reviews had been carried out and, the commission added, there was no evidence of a request from Seqwater that SunWater comply with this aspect of its obligations.

“This omission,” the report summaries, “assumes some significance when it is acknowledged that the Wivenhoe manual has in important respects, been found to be ambiguous and in need of amendment.” Indeed the manual has been described as lacking precision as to what can be done and how it should be done. The language is also unclear and there is more than one interpretation of certain points.

Seqwater’s tools at the flood operations centre were also under investigation. Flood engineers did not have access to hydrodynamic modelling, which, the report says, “would have given more precise indications of flood levels at particular locations downstream during the height of the flood event.”

The real time flood monitoring system, which had been used in the flood operations centre for 15 years, originally included a hydrodynamic model to determine flow velocities and levels along the river system. However, when the hardware platform was changed, that model was not retained nor replaced. An expert hydrologist called on by the commission considered that having such a model would have been ‘helpful in giving an understanding of the effects of releases from the dam’. The flood engineers did not agree. They used the Brisbane Valley Damage Minimisation Study of 2007 and said it provided them with understanding of the consequences of different flows. They didn’t believe that a hydrodynamic model would have affected how they managed the dams during the January 2011 flood event. The commission, however, remained adamant. The flood engineers should have hydrodynamic models available in the flood operations centre to assist them in determining the downstream impacts of releases from the dams.

“The best approach,” the commission advises, “is to ensure that the flood engineers are guided in their decision-making by a clear, unambiguous manual, based on the best available science, and are equipped with ample and up-to-date modelling tools.”

Criticism was also directed at the fact that Seqwater’s flood preparedness activities ‘do not seem to have extended to matters affecting the practical ability of the flood engineers to carry out their duties’. This was related to conditions under which staff have to work during prolonged flood events and the availability of food, accommodation, contact with family and friends and fatigue management. Plus there was a lack of training for one of the core flood strategies: this flood event was the first time the strategy had ever been triggered in training or in real operations.

“There is no evidence to suggest that the last two matters adversely affected the flood engineers’ performance during the flood event,” the commission said. “The point is that they were matters which should have been identified and addressed by Seqwater prior to the wet season. These four examples [arrangements for flood preparation, manual reviews, tools at the flood operations centre and flood preparedness activities] are not individually significant. However,’ the report claims, ‘they reveal that the process by which flood preparation was undertaken was inadequate.”

Safety concerns

Key safety concerns were also raised at North Pine dam during the flood event of 6-14 January 2011. North Pine had reached a peak lake level that was only 50cm below the electric winch motors which control the dam’s gates. If these become submerged they cannot operate and normal control of the gates is lost which may result in overtopping of the dam. In September 2010 Seqwater had started a project to provide an additional backup system for dam gate operations. The company is now also investigating the adequacy of the system and is reporting to a dam safety regulator on a monthly basis in relation to progress with the work.

Another problem emerged on 11 January when the control area for North Pine dam’s gates was inundated with flowing water. However, the dam operator continued to work here at risk to his own personal safety as it was the only area from which he could observe the gates while opening. Seqwater has since installed a duplicate electric gate control panel which will allow gate operations to be undertaken from a higher position. The commission is satisfied that this will remove safety risks for dam operators in similar future flood events.

Seqwater’s response

Seqwater welcomed the commission of inquiry’s interim report recommendations in relation to future dam operations and affirmed its commitment to working with other state agencies to implement them in full.

CEO Peter Borrows said supporting the commission's enquiries had been an extensive and detailed process and Seqwater had taken the appropriate time to review the report before making its official statement on 8 August 2011.

“We are acutely aware of the impact of the floods and the devastation caused which is ongoing for so many in our community,” he said. “Many of our own staff experienced damage and loss. Seqwater has worked cooperatively and closely with the commission since its establishment to help ensure southeast Queensland is best prepared for any future flood events.”

However Borrows said it was important not to lose sight of the magnitude and rarity of the January event.

“It was the largest flood that southeast Queensland has seen in more than 100 years. We effectively had two 1974 floods hit our dams and catchments less than 30 hours apart, yet the flood levels in the city reaches of the Brisbane River were around 1m lower,” he said. “There is no question that Wivenhoe and Somerset dams did their job and significantly mitigated the flooding.”

Borrows went on to add that the report endorsed the performance of Seqwater's experienced flood engineers during the unprecedented flood event.

“This, coupled with the commission’s own independent expert analysis that Seqwater’s flood engineers ‘achieved close to the best possible outcome’, is welcome acknowledgement of the effort and professionalism of the flood engineers, officers and dam operators who worked tirelessly throughout the 13-day event,” he said.

Many of Seqwater’s own recommendations had been adopted by the commission, Borrows said, and work is already well underway on implementing them. This includes the review of Seqwater’s flood operations manual which is required after every major flood event. The first phase, already in progress, will be an interim review and will take into account any potential change to Wivenhoe dam’s full supply level ahead of the coming wet season and the impact this will have on operations.

This will be followed by a longer, more comprehensive review, which has also commenced. It will involve a number of government agencies, local councils and consultation with the community.

Other key initiatives that Seqwater has been progressing include:

• The opening of a new flood operation centre which became fully operational in July 2011.

• Recruitment of additional duty engineers and flood officers to allow for appropriate succession planning and an increase in operational capacity.

• Investigations into increasing the number of rainfall gauges in the catchment to enhance data collection.

• Technical reviews including investigating enhanced modelling of major rainfall and flooding events.

• Training programmes to reflect a potential change in Wivenhoe dam’s full supply level ahead of the coming wet season.

“The commission has indicated that further submissions may be received before it delivers its final report in February and Seqwater will continue to offer every assistance,” Borrows said. “Seqwater is confident that its approach was appropriate and in accordance with the manual and best practice.”

Given the conclusions of the commission's independent expert, Seqwater notes that speculation that operating the dams differently during the event (including making earlier releases) may have resulted in a better mitigation outcome cannot be maintained.

There also is a popular misconception that Wivenhoe dam could contain all floods emanating in the upper Brisbane River. After considering many of the engineering reports produced about Wivenhoe dam the flood commission acknowledged that, other than for relatively small floods, the dam is only capable of mitigating and not preventing floods.

“We understand the impacts of the flood upon our community, of which we are a part,” Borrows added. “Unfortunately, this devastation is a consequence of the extreme rainfall within the catchment both above and below Wivenhoe dam in January - not the operation of the dams.

“We have to be ready to manage floods every day,” he said. “And we are.”

Final report

Minister for Energy and Water Utilities Stephen Robertson has asked key water agencies, including Seqwater, to accelerate recommendations contained in the interim report. In addition he has instructed his department to fast-track an investigation into raising the Wivenhoe dam walls to increase the flood mitigation buffer.

“We will implement all of these findings, including reducing Wivenhoe dam’s full supply level to 75% of current design capacity, if the Bureau of Meteorology provides advice in the lead up to a wet season that similar rainfall levels are expected,” he said. “We will also legislate to ensure there is greater clarity and understanding about whose role it is to undertake certain actions in relation to our dams in times of heavy rainfall.”

Robertson acknowledged that Seqwater is already reviewing the flood manuals in line with the commission’s recommendations. He also accepted the findings in the interim report and admitted that the state government ‘can do things better in the future’.

“Six and a half months is not a very long time to set up a commission, receive submissions, identify issues, assemble evidence, hold hearings and produce a report,” said commissioner Justice Catherine Holmes. “That it has been possible is a credit to the commission’s energetic and dedicated staff. The report strives to be practical, rather than descriptive; we have not dwelt on the tragedies suffered, because the most helpful thing we can do is to make suggestions to guard against repetition.”

Until the final report of the Queensland Floods Commission of Inquiry is released on 24 February 2012, its investigative work continues.



The flood event

According to Seqwater, in the 28 days prior to 6th January 2011, rainfall in southeast Queensland had been well above the December average and three separate flood events had occurred. Flood releases from Wivenhoe dam occurred on most days and as a result the catchments were near saturation and primed to generate run off. At this time virtually all of Wivenhoe and Somerset dams"™ dedicated flood mitigation storage volume was available for use, prior to the flood event commencing.
In line with the approved Flood Mitigation Manual, which outlines the operational procedures to be followed during flood events, water releases from Wivenhoe dam early in the event took into account the significant flood flows from Lockyer Creek and the Bremer river, and sought to keep the combined flow below the level where urban damage would be sustained below Moggill. Dam outflows that would cause urban inundation were delayed for as long as possible until it became apparent no other option was available without risking the safety of Wivenhoe dam.
Two distinct flood peaks occurred at Wivenhoe dam during the event, 30 hours apart:
"¢ First peak - While the first peak on 10 January was similar in nature and magnitude to the comparable flood flows of the 1974 flood event, the combined mitigation effect of Somerset and Wivenhoe dams ensured this first flood did not result in urban damage below Moggill. However, achieving this meant significant filling of Somerset and Wivenhoe dams"™ flood storage compartments.
"¢ Second peak - Intense rainfall that occurred directly on and near the Wivenhoe dam lake area on 11 January contributed to the second flood peak, which was also similar in nature and magnitude to the 1974 flood flows.
The effects of this intense rainfall, at a critical stage of the event, was exacerbated by the fact it fell on and near the Wivenhoe dam lake area, thereby immediately raising the lake level. Wivenhoe dam rose more than 1.5 m (from 73.4m to its peak of 74.97m) over a 15-hour period on this day. Accordingly, the second flood peak could not be completely contained without risking Wivenhoe dam"™s safety.
Damage in urban areas was unavoidable. Even if no water releases had been made from Wivenhoe dam, damage in downstream urban areas would have still occurred given the volume of flows from the Lockyer Creek and the Bremer River, Seqwater claims.
The flooding at Wivenhoe and Somerset dams can be categorised with an annual exceedence probability (AEP) of 1 in 100 years, to rare (AEP 1 in 2000 years). While North Pine dam can be categorised as rare (AEP of greater than 1 in 100 and potentially with an AEP in the order of 1 in 200 years). The maximum inflow rate was more than double the previous largest flow rate ever recorded at North Pine dam since records began. Total inflow into North Pine dam was 94% of the dam"™s total storage volume.
Seqwater believes that the dams provided clear flood mitigation benefits. The peak of the outflow from Wivenhoe dam was approximately 40% lower than the peak of the inflow, meaning that just below the dam, the maximum hourly flow rate in the Brisbane River was reduced by around 40%. The peak flood height measured at the Port Office gauge in Brisbane was 4.45m. It has been estimated this peak height would have been approximately 2m higher without Wivenhoe dam. This projected reduction in the flood peak height resulted in significant reductions in the potential for the loss of life and 14,000 fewer properties being impacted.
Source: Seqwater fact sheet.



Commission fact file

Established on 17 January 2011 the Queensland Floods Commission of Inquiry is headed by Justice Catherine Holmes. Former Queensland Police Commissioner Jim O"™Sullivan is a deputy commissioner along with Phillip Cummins, current chairman of ICOLD"™s Committee on the Operation, Maintenance and Rehabilitation of Dams.
The Queensland government has allocated A$15M for the inquiry to provide an independent and thorough examination of the chain of events leading to the floods, all aspects of the response and the subsequent aftermath of the 2010-11 disaster. The terms of reference are extensive and include:
"¢ Preparation and planning for the flooding by governments at all levels, emergency services and the community.
"¢ The supply of essential services during the floods.
"¢ The adequacy of forecasts and early warning systems.
"¢ Land use planning to minimise flood damages.
"¢ The performance of insurers in meeting their claims responsibilities.
"¢ A full and careful inquiry of the implementation of the operation plans for dams across the state, in particular the Wivenhoe and Somerset dam release strategy. Plus an assessment of compliance with, and the suitability of, the operational procedures relating to flood mitigation and dam safety.
In the first round of public hearings evidence was obtained from 167 witnesses over 31 days. Community meetings were also held in 13 locations across central and western Queensland and the Lockyer Valley. The second round of public hearings will take place during September and October 2011.
The commission also called for public submissions addressing issues raised by the floods. Written submissions relating to issues of flood preparedness relevant to next summer"™s wet season (particularly dam operations, early warning systems and responses) were requested. To date over 600 public submissions have been received.
The commission was originally required to provide a final report by 17 January 2012. However the date has been extended to 24 February because of the commission"™s extensive public hearing schedule and the volume of evidence that must be considered. Both reports will be available on the commission"™s website www.floodcommission.qld.gov.au



Wivenhoe Dam

* Radial gates and an auxiliary spillway are used for flood events.
* During extreme events when water levels exceed EL 75.7m the damââ"šÂ¬Ã¢"žÂ¢s fuse plugs may be triggered.
* Following an upgrade in 2005 the damââ"šÂ¬Ã¢"žÂ¢s failure level was raised to EL 80m.
* Dam catchment area of 7020km2.
* Full supply level (drinking water storage compartment): 67m.
* Flood mitigation capacity (flood storage above FSL): 1,420,000 ML.



Somerset Dam

* Built in 1959 for water supply and flood mitigation on the Stanley River. Also has a 4MW hydro station.
* Is a 305m long mass concrete gravity structure.
* Has radial gates, sluice gates and regulator valves.
* Dam catchment area of 1340km2.
* Full supply level (drinking water storage compartment): 99m.
* Flood mitigation capacity (flood storage above FSL): 524,200 ML.
* During flood events Somerset and Wivenhoe dams are operated in tandem to maximise the overall flood mitigation capabilities.



North Pine Dam

* Built in 1976 as a 580m long concrete gravity dam with earthfill embankments on the abutment. Located on the North Pine River.
* Provides drinking supplies and no flood mitigation.
* Dam catchment area of 348km2.
* FSL of 214,000ML



Flood Waters around Rockhampton Flood Waters around Rockhampton


Privacy Policy
We have updated our privacy policy. In the latest update it explains what cookies are and how we use them on our site. To learn more about cookies and their benefits, please view our privacy policy. Please be aware that parts of this site will not function correctly if you disable cookies. By continuing to use this site, you consent to our use of cookies in accordance with our privacy policy unless you have disabled them.