M C N Taylor, Dr H E Cherrill, S F Croft, S F Eldridge
The Stuart Macaskill Lakes are two raw water storage lakes with a combined storage of approximately 3280 ML supplying Wellington City, New Zealand. The lakes are High Potential Impact Category (PIC) earth embankment dams constructed on terrace gravel deposits adjacent to the Hutt River and located within approximately 20 to 50 metres of the Wellington Fault Deformation Zone. Construction of the lakes began in 1982 and they were commissioned in 1985.
In early 2008, the lake’s owner Greater Wellington Regional Council (GWRC), embarked on a programme to supplement Wellington City’s water supply storage. Whilst that study is ongoing, GWRC engaged Tonkin & Taylor (T&T) to investigate the feasibility of increasing the Stuart Macaskill Lakes capacity as an interim measure.
The feasibility study concluded in late 2009 that the lake dam embankments could be raised by up to 1.3 metres in height to gain an approximate additional 450 ML of water storage. An important finding of that feasibility study has been that the seismic requirements have increased significantly since the construction of the lakes. To address this issue GWRC is currently constructing Stage Two of a two stage construction programme to both raise the lakes and to incorporate seismic resistant features into the lakes.
The primary design features are downstream rock buttressing in the critical areas of the lakes and synthetic lining the inside of the lake embankments. The buttressing works were completed in early 2011 and the lining and crest raising works are due for completion in 2013.
This paper summarises the design, laboratory testing and construction to enhance the lakes performance during very strong seismic accelerations (Peak Ground Accelerations of up to 1.08g) expected during a maximum design earthquake originating from the Wellington Fault.
Keywords: Water Reservoir, Seismic Design, Geomembrane, Rock Buttressing, Seismic Risk Assessment, Wellington Fault
— OR —
M. Tooley, D. D’Angelo, B. Priggen, K. Sih, N. Vitharana, R. Mouveri
As the urban sprawl of residential and commercial businesses expand to meet rising population, consideration must be given to the frequency and intensity of storm events and changes in tidal levels, to mitigate the risk of flooding and damage associated with the failure of hydraulic structures.
This paper outlines the design method undertaken to ensure the ageing structure (founded on timber piles) meets modern dam safety criteria, extends the life of the 8 gates operating mechanisms and provides overall inherent reliability for the whole structure. The design method included updated hydrological assessment of the upstream catchment, geotechnical investigation, liquefaction review, consequence category and AFC assessment, hydraulic assessment and stability analysis.
These assessments are being undertaken to introduce inherent reliability in their operation in particular during king tide or storm water events, or a combination of the both, minimising leakage and breakdowns and ensuring the risks of flooding to low lying residential areas upstream of the structure and major airport are minimised. The Glenelg Gates structure is an integral part of a larger regulating system for the catchment.
The findings of the design upgrade would be useful to dam designers and owners faced with the upgrading of gated structures with flooding risks in residential areas.
Keywords: Gated Glenelg Gates structures, upgrade, dam design guidelines.
Gavan Hunter, Robin Fell, Chris Topham
Backward erosion piping is a failure mode that can affect water retaining structures with earthen cores of very low or no plasticity. Backward erosion involves the progressive detachment of soil particles as seepage through a core material exits to a free surface or unfiltered zone. In contrast to other piping failure mechanisms, backward erosion does not require a defect to be present for initiation, and is heavily influenced by the inherent characteristics of the core materials and the available hydraulic head. For dams with non-plastic or very low plasticity core materials, backward erosion can be a material contributor to the overall piping risk and warrants careful consideration during quantitative risk assessments of such dams. However, there is very little published literature for evaluating the potential for backward erosion piping, particularly in broadly graded soils. This paper concerns one such dam where backward erosion of the glacial till core needed to be assessed in the context of a detailed risk assessment for the facility. The backward erosion mechanism was tested in laboratory tests set up to model the situation in the core of the dam at a range of hydraulic heads. The paper describes the core material and objectives for the testing, presents the apparatus used, summarises the findings, and explains how they contributed to the risk assessment for the dam. Recommendations are also made for future similar testing and research needs.
Keywords: Backward erosion, piping, embankment dam, laboratory testing, quantitative risk assessment, glacial till.
The design of tailings dams under earthquake loading is quite challenging due to the nature of the tailings materials which are generally liquefiable under earthquake shaking. The design will be more complicated when the dam foundation is also liquefiable material. While assessment of liquefaction potentials is well developed in practice, assessment of liquefaction induced deformation varies from the simplest Newmark’s displacement method to the more complex effective stress dynamic analysis approach. It is generally accepted that the simplified method can be used for cases involving non-liquefiable materials. However, the use of this method for cases involving liquefaction may generally result in overly conservative designs to cater for the many simplified
assumptions in the method. With the advance of computer technology, time and cost are no longer obstacles for using the more appropriate method for estimating liquefaction-induced deformations of a tailings dams and achieving an optimum dam design.
This paper attempts to critically discuss issues in seismic design of tailings dams and provide an example of the use of the effective stress dynamic analysis method to estimate the liquefaction-induced deformations of a tailings dam and its importance in optimizing the design. The approach used is capable of estimating pore pressure response of liquefiable materials at any given time during the shaking. The effective stress analysis method used herein is embedded in FLAC software using a specially written FISH routine. Using this method, it can be demonstrated that although liquefaction is an issue, it does not necessarily mean that we must prevent its occurrence. As long as the deformation is acceptable, liquefaction is not necessarily a ‘show stopper’ for the project.
Keywords: liquefaction, seismic deformation, tailings dam design.
A.E. Bentley, P.I. Hill, S.M. Lang, M. Freund, A. Richardson
This paper describes the development of a detailed assessment approach using spatial data to estimate the consequences of dam failure across a portfolio of 18 dams in NSW. The assessment is made for potential loss of life; economic and financial losses and a qualitative assessment of environmental and social impacts. The approach is designed around the use and interrogation of spatial databases combined with outputs from hydraulic models. The assessment method is applicable to a wide range of dams in different valleys, each with different downstream characteristics. The paper provides discussion on the advantages of the approach and presents some insights into the effective application to a dam portfolio of significant size and scale.
Keywords: consequence assessment, spatial databases
Graeme Mann, Michael Smith, Louise Thomas
Regular flooding around the coastal town of Busselton, south of Perth, led to the construction of large rural drains in the 1920s to divert two of the major rivers around the town. Hydrologic studies after major floods in 1997 and 1999 showed that the existing drains were providing much less than the desired 1 in 100 AEP flood protection, particularly as subdivisions were being developed along both sides of the Vasse River Diversion Drain (VDD).
Three compensating basins constructed in rural land south of Busselton provide a total storage capacity of 4.4 GL. The banks that form the three basins have a total length of 8.5 km, vary in height up to 6 m and are either zoned earthfill embankments, with a clay foundation cut-off through sandy soil horizons to a depth of 1 to 2 m below ground level or homogeneous earthfill embankments. The spillways are overflow sections on the embankments using concrete revetment mattresses. The outlet works are uncontrolled box culverts with a capacity of up to 14 m3/s. Peak outflows are typically about 30% of peak inflows to the basins.
The paper discusses the Busselton Flood Protection Project and associated diversion drains, including the design of long embankments for the compensating basins on very flat terrain that are required to survive their “first filling period” during each flood emergency.
Keywords: Earth embankments, flood mitigation, flood compensating basins, levee banks, diversion drains, Busselton, piping failures.