K. Chandler, D. Gill, B. Maher, S. Macnish and G. Roads
SEQWater is the major supplier of untreated water in bulk to Local Governments and industry in the South East Queensland region of Australia, through ownership of Wivenhoe, Somerset and North Pine Dams. Wivenhoe Dam (Lake Wivenhoe) is located on the Brisbane River in Esk Shire. The storage provides both flood mitigation and water supply storage to Brisbane and Ipswich. The water supply storage capacity at full supply level is 1,160 GL. An additional 1,450 GL of storage above full supply level is used for flood mitigation.
Changes to the estimation of extreme rainfall events has resulted in significant increases in the estimates of the PMF since the original design of Wivenhoe Dam. To upgrade the flood security of Wivenhoe Dam, SEQWater has formed an alliance with Leighton Contractors, Coffey Geosciences, MWH and the NSW Department of Commerce.
This paper details the alliance delivery method, the latest estimates of the PMF based on the GTSMR method and details of the two preferred options being finalised by the Alliance.
There is a widespread perception among dam engineers that tree root invasion occasions a very serious threat to embankment dams by virtue of its potential to initiate piping failure, with appropriate action invariably recommended. Remedial works can, on occasions, be extensive.
While the principle is ostensibly plausible and scarcely challenged, there has never been, to the Author’s knowledge, a satisfactory investigation to establish any credible scientific basis for it. One case that has attracted some attention in literature (by virtue of the extent of the investigation undertaken), viz a piping accident at Yan Yean Dam, is critically reviewed to show that the accepted view on the role of tree roots in this incident is less than satisfactory. In the course of this review, two physical Laws of Piping are proposed, and applied both to this case and to another nearby Melbourne Water dam that also has a history of piping.
Whilst the consequences of piping in a major dam are such that risk from this source must be kept to a very low level, it is concluded here that piping risk arising from tree root invasion has been considerably overstated and that a more balanced assessment is necessary before determining what, if any, action is required.
A survey of spillway gate systems and operations has recently been completed by dam organisations in Nth America, Australia and New Zealand. The survey sought to identify typical arrangements for spillway gate systems and common features pertaining to reliability such as system redundancy, actuation methods and back-up systems, gate and hoist types, remote and local operation, gate testing programmes, and human factors.
Sixteen organizations responded, covering sixty two dams and nearly four hundred gates. This Paper reports on the preliminary analysis of the data, providing an overview of the industries’ approach to spillway gate operation and control.
Investigations of damaging blowback incidents at the headrace tunnel intake to Rangipo Power Station in the Central North Island of New Zealand are described. The blowback phenomenon is explained theoretically based on evaluation of the evidence available from the incidents and information obtained from the literature. A physical hydraulic model study is described in which this explanation of the blowback phenomenon was verified. The model was also used to devise a solution for the blowback problem.
There are many important dams and other structures on catchments smaller than 1000 km² with response times less than 24 hours, however these catchments have been largely overlooked in previous research into large and extreme floods. This paper is an initial step in “catching up” design practice for short duration rainfall events to the current best practice that is available for estimation of floods from rainfall events with durations of 24 hours and greater.
Two issues are specifically addressed in this paper. Firstly, a regional analysis of short duration rainfall depths is conducted to extend the frequency curve beyond an AEP of 1 in 100. Rainfall frequency curves are estimated for durations between 0.5 and 12 hours, using data from ten pluviograph sites around Australia. Secondly, sets of temporal patterns are derived that could be useful in joint probability analysis of short duration rainfall events. The effects of these new rainfall depths and temporal patterns on flood frequency curves are tested by applying them to rainfall-runoff routing models for three dams with small catchment areas.
By 1976 head loss in the 23 km long 750/900 mm diameter CLMS pipeline from Eppalock Reservoir to Bendigo had increased from 45.7 m to 98.2 m (115%) after only 12 years service. The cause was identified as increased friction from soft voluminous iron and manganese bacterial slime building up on the pipe walls and increasing the friction. Inspection of the drained pipes in the dry gave little indication of the problem since the slime consolidated to an innocuous looking thin smooth coating as it dried.
1960 studies by Tyler and Mitchell at the University of Tasmania for the Hydro-Electric Commission had shown that the micro-organisms producing these slime growths were present in all pipelines. However they required the presence of iron and manganese in the water to flourish and produce flow reduction. Remobilisation from oxygen deficient bottom sediments was shown in the 1940’s by Pearsall and Mortimer in England to be a major source of iron and manganese in reservoir water and this could be controlled if sufficient dissolved oxygen could be provided to convert the reducing conditions at the sediments to oxidising conditions.
An experimental aeration system designed by the author was operated in the 180,000 ML Eppalock Reservoir for 19 days during March 1977. This mixed the reservoir to the depth of the aerators (24 m) increasing the low 10% saturation dissolved oxygen at this depth to a high 94% saturation thereby changing chemical conditions from reducing to oxidising. As a result the iron concentration in the surface water decreased from 2.04 mg/L to 0.54 mg/L but there was little change in the pre-aeration 0.03 mg/L manganese concentration with this short period of aeration. The iron concentration in the water flowing in the pipeline changed from 1.78 mg/l to 0.57 mg/l.
The problem of pipe flow reduction from bacterial slime growth on the pipe walls is discussed in this paper and examples are given of the use of automatic reservoir aeration to overcome the problem including costs and results.