Forecasts to blame, 'not Wivenhoe Dam operator'

THE operator of Brisbane's major dam has cleared itself of responsibility for significantly contributing to the devastating January floods.

THE operator of Brisbane's major dam has cleared itself of responsibility for significantly contributing to the devastating January floods, instead blaming inaccurate weather forecasts for underestimating rainfall volumes.

In the wake of studies by senior engineers that the river flooding in Brisbane was largely avoidable, the 1180-page technical report by SEQWater was released about 8pm last night by the Queensland government, owner of the Wivenhoe Dam.

The report finds the operating manual for the Wivenhoe Dam should be thoroughly reviewed and a number of changes are canvassed subject to extensive engineering and hydrological testing.

The report makes little comment on the decision made since the January floods to keep the dam levels 25 per cent lower to give the dam more capacity for another flood. If the decision had been made late last year, when the weather bureau was forecasting extreme rainfall, the flooding in the Brisbane River in January would have been significantly less.

Public recriminations over the operation of Wivenhoe Dam in the days before the flooding in the Brisbane River have raged since senior engineers began questioning the strategies of SEQWater to keep significant volumes of water in the dam over the weekend of January 8-9 and into Monday, January 10.

The effect of not releasing large volumes of water from Wivenhoe Dam on those crucial three days meant that, when extreme rainfall occurred soon after, the dam filled quickly and the operators were forced to make huge and sudden releases.

Senior engineers have identified these releases, on Tuesday, January 11, as the source of most of the flooding in the Brisbane River.

The report, released last night, states that in the eight three-day forecasts between Thursday, January 6, and Sunday, January 9, "the forecasts underestimated rainfall on all but two occasions".

"In hindsight it can be seen that these were the best forecasts provided during the event. In practice, the forecasts could not be used as a basis for decision-making as there was no way of determining that these individual forecasts were more accurate than any others provided.

"Additionally, the large fluctuations between successive forecasts that overlap by two and a half days do not provide a sensible basis for proper and consistent decision-making. These underestimated rainfall forecasts did not support an increase in flood releases above that undertaken."

There are eight instances in which actual rainfall recorded is greater than 200 per cent (highest is more than 1000 per cent) of the three-day forecast rainfall; there are three instances in which the three-day forecast rainfall is greater than 150 per cent (highest is 280 per cent) of the actual rainfall recorded; there are nine instances in which actual rainfall recorded is greater than 300 per cent (over 1000 per cent) of the five-day forecast rainfall; and two instances in which the five-day forecast actual rainfall is greater than 200 per cent (280 per cent) of the actual rainfall recorded.

"These results show that three-day and five-day forecasts only provide an indication of future rainfall and these forecasts cannot be used as a basis of flood operations decision-making where public safety in both rural and urban areas is directly impacted. Future improvements in this area will be examined with interest in order to maximise the flood mitigation benefits of the Dams."

The report states that, when "compared with historical events, flood volumes indicate the volume of the January 2011 Event was almost double that of the January 1974 flood and rivals the February 1893 flood". SEQWater also estimates that without the mitigating effects of Wivenhoe Dam the peak flood height in Brisbane CBD would have "been approximately 2.0m higher than experienced".