Mark Arnold, Gavan Hunter and Mark Foster
Following the dam safety risk assessment for Greenvale Dam in 2008, Melbourne Water implemented a 3.0 m reservoir level restriction on the operation of the storage as an interim risk reduction measure. The 3.0 m restriction coincided with the ‘as constructed’ top of the chimney filter in the main embankment. This interim action reduced the dam safety risk to below the ANCOLD limit of tolerability.
Dam safety upgrade works were undertaken in 2014/15 to bring the dam in-line with current risk based guidelines and to enable the removal of the interim reservoir restriction, bringing the storage back to full operating capacity. Greenvale Dam was required to remain operational throughout the works and this required careful consideration of the dam safety risk during construction.
Deep excavations were required within the crest and downstream shoulder of the embankments, that,, without adequate management, had the potential to increase risk to the downstream population. Excavations up to 18 m depth were required into the wing embankments for construction of full height filters from foundation to crest, excavations up to 7 m deep were required in the main embankment to expose and connect into the existing filters and secant filter piles up to 13 m deep were used to connect the new chimney filter of the wing embankments with the original chimney filter of the main embankment.
A key element of the design and construction of the upgrade works was managing dam safety during construction. Dam safety considerations included (i) design based decisions to manage the level of exposure; (ii) implementation of further restrictions on reservoir level by the owner Melbourne Water; (iii) construction methods to manage exposure; (iv) an elevated surveillance regime during the works and (v) emergency preparation measures including emergency stockpiles and 24 hour emergency standby crew. The construction based dam safety requirements were focused on early detection and early intervention, and were managed via the project specific Dam Safety Management Plan.
This paper focuses on dam safety management including the decisions made, actions taken and construction requirements and touches on how these relate to the key project features.
— OR —
David Laan, Kim Matsen
A slip on the upstream face of Hedges Dam was observed during an annual site inspection in late March 2016. At that stage the slip appeared to be largely contained within the right hand third of the embankment.
By early April, the slip area had developed into a head scarp across the entire central portion of the embankment. Multiple other head scarps were observed, indicating multiple or segmented slips. Several tension cracks were also visible on the face of the dam. The toe of the slips was indicated by a poorly defined bulge.
The most recent drawdown of the reservoir level was identified as a potential driver for the initiation of the slip failure. During the most recent drawdown the maximum drawdown rate was approximately 0.6 m/day whereas in the previous 17 years the maximum drawdown rate was approximately 0.2 m/day.
The remedial works proposed are to place a rockfill weighting zone on the upstream face to stabilise the embankment. The strength of the materials along the sheared surface was back calculated from the mechanics of the failure surface. This data was then used to calculate the shape of the weighting zone required to stabilise the slope.
This paper reviews methods used to estimate the MCE in Australia and New Zealand. In the ICOLD (2016), NZSOLD (2015) and proposed ANCOLD (2016) guidelines, the deterministic approach is applicable only to fault sources, whereas the probabilistic approach is applicable to both fault sources and distributed earthquake sources. Although ICOLD (2016) states that the use of a deterministic approach to develop the SEE “may be more appropriate in locations with relatively frequent earthquakes that occur on well- identified sources, for example near plate boundaries,” the proposed ANCOLD (2016) guidelines retain the use of the deterministic approach for critical active faults which show evidence of movements in Holocene time (i.e. in the last 11,000 years), or large faults which show evidence of movements in Latest Pleistocene time (i.e. between 11,000 and 35,000 years ago). In Australia, active faults make a significant contribution to the probabilistic MCE only at near-fault sites, and even in those cases most of the hazard comes from distributed earthquake sources. However, some sites may be close enough to nearby or even more distant identified active faults that a Deterministic Seismic Hazard Analysis (DSHA) produces MCE ground motions that are far larger than those obtained probabilistically even for very long return periods. Conversely, the deterministically defined MCE may be lower than the probabilistically defined MCE for very long return periods at near fault sites in New Zealand, requiring the probabilistic approach.
Kristen Sih, Richard Rodd
Melbourne Water currently manages over 235 stormwater retarding basins. The process of assessing the risk posed by these assets began in 2006, and at the end of 2015 full risk assessments were completed for around 30 of the basins that were estimated to pose the highest societal risk. However, when analysing the results of these risk assessments, there was some concern that the results were inconsistent and often too conservative, given the few incipient or actual failures that had been experienced.
It was found that one of the key areas causing the conservatism was poor documentation of design and construction details, and the fact that the tools used for assessing the Potential Loss of Life (PLL) were aimed at larger storages that cause much higher depths and velocities in dambreak events than these (generally) small storages. To remedy this situation, advice was sought from specialist practitioners to develop guidance notes on the assessment of PLL and failure likelihoods for retarding basins.
On the back of these guidance notes, Melbourne Water initiated an accelerated program of assessing the risk associated with 78 retarding basins over a 6 month period. This paper describes the key recommendations from the guidance notes, compares the results of the risk assessments performed pre- and post-guidance notes and provides a summary of the portfolio risk assessment outcomes, what they mean for Melbourne Water and what the organisation intends to do to manage this risk into the future.
Peter Buchanan, Malcolm Barker, Paul Maisano, Marius Jonker
Kangaroo Creek Dam located on the Torrens River, approximately 22 km north east of Adelaide, is currently undergoing a major upgrade to address a number of deficiencies, including increasing flood capacity and reducing its vulnerability to major seismic loading.
Originally constructed in the 1960s and raised in 1983, recent reviews have indicated that the dam does not meet modern standards for an extreme consequence category dam.
The original dam was generally constructed from the rock won from the spillway excavation. This rock was quite variable in quality and strength and contained significant portions of low strength schist, which broke down when compacted by the rollers. The nature of this material in places is very fine with characteristics more akin to soil than rock. Review of this material suggests that large seepage flows (say following a major seismic event and rupture of the upstream face slab) could lead to extensive migration of the finer material and possible failure of the embankment. However, it is also envisaged that the zones of coarser material could behave as a rockfill and therefore transmit large seepage flows, which may result in unravelling of the downstream face leading to instability.
This paper addresses the design of the embankment raising and stabilising providing suitable protection against both these possible failure scenarios, which tend to lead to competing solutions. The final solution required the embankment to be considered both as a CFRD and a zoned earth and rockfill embankment.
David Scriven, Lawrence Fahey
Paradise Dam is located approximately 20 km north-west of Biggenden and 80 km south-west of Bundaberg on the Burnett River in Queensland. The dam was designed and constructed under an alliance agreement with construction completed in mid 2005. It is a concrete gravity structure up to 52 m high, the primary construction material being roller compacted concrete (RCC).
In January 2013 the flood of record was experienced at the dam with a depth of overflow on the primary spillway reaching 8.65 m following heavy rainfall in the catchment from ex-tropical cyclone Oswald. The peak outflow was approximately 17,000 m3/s. This equated to a 1 in 170 AEP flood event. When the flood receded it was discovered that the dam and surrounds had suffered severe damage in a number of locations including: extensive rock scour downstream of the primary dissipator and the left abutment, damage to portions of the primary dissipator apron, and the loss of most of the primary dissipator end sill.
SunWater initiated a staged remediation program to manage the dam safety risks and by November 2013 had completed the initial Phase 1 Emergency and Phase 2 Interim repairs. Phase 3 of the program was to implement a comprehensive Dam Safety Review (DSR) and a Comprehensive Risk Assessment (CRA). The DSR became arguably the largest ever undertaken by SunWater and included: extensive geotechnical investigations, large scale physical modelling, numerical scour analysis, stability analysis, and an extensive design assessment. This paper describes some of the key aspects of the DSR undertaken related to the flood damage.