Paul Somerville, Gary Gibson
Abstract: This paper describes current methods for seismic hazard analysis and their application at Hinze Dam. Although Southeastern Queensland has experienced significant earthquakes in historical time, none of them are known to have caused surface rupture, and no active faults that could be used to represent earthquake sources have been identified in the region that surrounds the site. Under these conditions, we must estimate the seismic potential of the region using historical seismicity. Two alternative approaches to modelling future earthquake occurrence based on historical seismicity have been used. The first approach is based on the AUS5 source model of ES&S (2005), which uses geological criteria to identify zones of uniform seismic potential, and then uses historical seismicity to characterize the seismic potential of each zone. The second approach, developed by Hall et al. (2007) at Risk Frontiers, is based on the spatial smoothing of historical seismicity without identifying discrete source zones. Previous work by ES&S has shown that the attenuation of strong ground motion in Southeastern Australia is fairly well represented by ground motion models developed using strong motion data from western North America. The recently developed NGA ground motion models based mainly on data from Western North America represent the local site conditions using Vs30, the shear wave velocity averaged over the top 30 metres at the site. This provides a significant advantage over previous models, which were for broad site categories such as rock or soil, and did not provide for the use of more site-specific information. The left abutment, lower tower and valley section foundation at Hinze Dam are characterized by hard unweathered rocks with shear wave velocity of 2.0 km/sec estimated from P wave velocity measurements. The right abutment of the main embankment and the saddle embankment foundation consist of extremely weathered rock, with shear wave velocity of 0.45 km/sec estimated from P wave velocity measurements. This causes the ground motion response spectra estimated for the right abutment and neighbouring foundation components to be significantly larger than for the left abutment and neighbouring foundation components, by factors of 1.4, 2.0 and 2.3 for periods of 0 (PGA), 0.5 sec and 1 sec respectively.
Keywords: seismic hazard analysis.
David M Schaaf, Jeffrey A Schaefer, Rick W Schultz, Jason T Needham
Abstract: As one of the main federal agencies with responsibility to build, operate, and maintain large dams in the United States, the US Army Corps of Engineers (USACE) is developing a risk based framework to better manage their portfolio of 600+ dams in terms of risk management and prioritization of funding. A key element to this effort is the development of risk-based analytical tools to evaluate primary features for applicable failure modes. These are used in conjunction with loading and consequence modules to assess the overall risk associated with the dam in terms of lives and economic damages. The focus of this paper is on the engineering analysis modules used to generate fragility curves for dam features.
The analysis modules are broken into three main categories by engineering discipline: geotechnical, structural, and mechanical/electrical. The risk based assessment tools associated with geotechnical failure modes include Seepage & Piping, Embankment Stability, Seismic Performance, and Erosion of Unlined Spillways. The structural assessment tools include Concrete Monolith Stability, Spillway Gates, Scour of Concrete Lined Spillways, Spillway Training Wall Stability, Performance of Pipes through Dams, Hydropower Superstructures and Intake Towers. The mechanical and electrical are primarily focused on the performance of machinery used to operate dam gates.
This paper gives a broad overview of the main characteristics and methods used for each of these analysis tools. Some of the modules use historical performance to establish failure rates, while others are more analytically based. The context of each within the framework of the overall risk assessment effort of USACE dams is covered.
Keywords: risk based analytical tools, fragility curves, risk assessment, US. Army Corps of Engineers, portfolio, dams.
David Brett, Ben Hanslow. Rob Longey
Abstract: Mine tailings storages are among the largest man made structures in the world and often pose a considerable risk to the aquatic environment due to the nature of the stored materials. In particular, sulphide minerals are prone to oxidation when brought into contact with air and water. This leads to the formation of acidic conditions within the storages leading to dissolution of toxic metals, with seepage from these structures being known as Acid Rock Drainage (ARD). ARD is responsible for pollution of natural waterways in many areas of the world with some significant examples in Australia. Current practice in the mining industry is to attempt to exclude oxygen or water from tailings storages in order to prevent the oxidation process taking place. This involves capping of the storages with sophisticated soil covers or, where sufficient water is available, leaving a permanent water cover.
Mining operations have a relatively short life, usually around 5 to 10 years, although some can operate for over 100 years, as has the Mount Lyell Mine. Normal practice has been for companies to relinquish the mining lease on the cessation of mining, however governments are now realising the extent of liability involved with the “ownership” of large waste storage facilities. Bonds are placed by mining companies during operations, intended to cover the cost of “closure” of the mine. Often the major item covered by the bond is for the “closure” of the tailings storage facility. Following “closure”, the intention is that ownership of the lease, including waste storages, reverts to the State. State governments are now more aware of the potential liabilities in accepting the relinquishment of these leases and need to address the issues of their long-term management.
In Tasmania, Dam Safety legislation covers both water and also soil covered tailings storages, with the legislation requiring each type to meet various ANCOLD guidelines. In other jurisdictions this could well also be the case through common law requirements to meet common best practice. However, the current ANCOLD Guidelines are generally written around water storage dams and interpretation to include a waste storage facility is often not straightforward. As an example a tailings dam during operations with a water storage component is clearly a dam. Due to the environmental impact of failure it could well have a Significant or High-C hazard rating, which would require design for extreme floods and earthquakes. After closure, with say a soil cover and water diverted away, is it still a dam within an ANCOLD definition? Are ANCOLD guidelines relevant? The current ANCOLD (1998) Guidelines on Tailings Dam Design, Construction and Operation does not give specific guidance on these issues.
This paper explores these questions and suggests ways that ANCOLD could provide assistance with more guidance on the long term management aspect of tailings storages to assist designers, owners and regulators consider the closure phase.
Keywords: tailings, acid drainage, mine closure
Matthew Pollard, John Vitkovsky, Richard Priman
Abstract: South East Queensland (SEQ) currently has severe (Target 140) restrictions imposed to help secure supplies during the current drought which is the worst on record. Additionally, a $9 billion water infrastructure program is being fast-tracked to increase the climate resilience of the region and provide for long term sustainable growth.
The Draft South East Queensland Water Strategy (SEQWS) released in March 2008 was prepared by the Queensland Water Commission to reduce the likelihood of ever experiencing such severe restrictions again and to ensure water security into the future. The Strategy includes a Water Supply Guarantee underpinned by advanced analytical techniques for estimating the system yield from surface water, groundwater and manufactured water supplies connected by the SEQ Water Grid. This approach builds on the Level of Service (LOS) Objectives methodology, originally promulgated by the Water Services Association of Australia in their June 2005 paper entitled “Framework for Urban Water Resource Planning”. The approach has led to a significant improvement in our understanding of water supply risks and the associated planning implications for assessed yields/water availability.
To assess yield using the LOS objective methodology and to determine the benefits of the new water infrastructure, a regional water balance model for the connected SEQ Water Grid simulating the water sharing arrangements of all of the SEQ water sources was established. The model uses a logic tree to allocate water using a “proportional storage rule” from multiple sources to meet competing demands. This approach enabled quantification of the increase in system yield resulting from the construction of the SEQ water grid which allows demands from previously disconnected areas to be met by more efficient allocation of water from supply sources. Stochastically generated dam inflow data was used to facilitate a more comprehensive assessment of climate variability and water supply risk than is possible with historic data alone.
The paper discusses the managed hydrologic risk approach adopted in developing the regional water balance model which implements the LOS Objectives approach and improves the understanding of the relationship between the level of service, supply yield and risks associated with climate variability. The result is a far more thorough approach to planning for future water service delivery and water infrastructure.
Keywords: level of service objectives, South East Queensland Water Strategy, SEQ Water Grid, yield, water security, regional water balance model, stochastic modelling, climate variability.
Eric Lesleighter, Erik Bollaert
Abstract: Many of the dams in Australia, and other countries, are potentially unprotected from the occurrence of extreme floods which would discharge either over the dam and/or the abutments due to inadequate spillway capacity.
The paper commences with the presentation of the evaluation procedure that has been applied by the first author for a number of dams in Queensland, initially with a detailed description of dams, the original hydraulics studies, the current hydrology, and the geological information. The procedure then comprises analysis of the extreme flood hydraulics and an erodibility assessment which leads to an estimate of the possible erosion of the rock for a range of flood magnitudes. A description of the components of the final evaluation and the typical conclusions is presented with reference to the Julius Dam in Queensland.
The paper includes a description of an alternative or complementary procedure that comes from the work of the second author. Comparison is made with the Comprehensive Scour Model (CSM) developed by Bollaert (2002, 2004). This model is physically based and allows estimating scour formation in rock or concrete as a function of time duration of discharge. The model compares the resistance against fracturing of concrete or rock layers with the hydrodynamic pressure fluctuations exerted by a turbulent aerated jet impacting in the plunge area. Once the fracture network is formed, dynamic uplift of single rock or concrete blocks in computed. Based on fracture mechanics and air-water hydrodynamics, a detailed time-evolution of scour formation is obtained. The model is generally used for projects involving plunge pool floors, ski-jump spillways, fractured rock and so on, where sufficient data are available on both rock quality and duration and intensity of spill from the dam. In the present paper, the CSM has been used to model scour of non-overflow abutment sections when there is overtopping flows.
Keywords: dams, spillways, extreme floods, rock scour, erosion, dam safety
Gregg A Scott
Abstract: The Bureau of Reclamation has been performing quantitative risk analysis as the primary dam safety decision making tool for well over a decade. This paper summarizes some of the key concepts and basic methodology currently used in the dam safety risk analysis process at Reclamation.
Keywords: dam safety, risk analysis, reliability analysis, event trees, subjective probability.