Millions of dollars are spent on dam upgrade works which are often undertaken to meet the flood security requirements. Prioritisation of the dam upgrade work is based on portfolio risk assessments in which dambreak modelling is an integral part. Concurrent design flow hydrographs of tributaries downstream of dam are required for the assessment of the incremental effect of a dam break scenario. The Annual Exceedance Probability (AEP) neutral concurrent tributary flows can be estimated using a bivariate-normal distribution approach.
This paper examines the underlying assumptions made in the application of the bivariate normal distribution approach using observed and design rainfall data for Avon Dam and its downstream tributary catchments. Synthetically generated data was used to illustrate the impact of the log-normal distribution assumption on the AEP neutral concurrent tributary rainfalls. This paper suggests a modification to the bivariate-normal distribution approach to estimate more unbiased AEP neutral concurrent design rainfalls. The use of historical gridded rainfall in the estimation of inter-catchment rainfall correlation is also demonstrated.
Now showing 1-12 of 59 2982:
The revised magnitudes of the Geoscience Australia’s NSHA18 earthquake catalogue approximately halve the rate of occurrence of earthquakes of a given Mw magnitude in Australia. This yields probabilistic ground motion levels that are significantly lower than the present design levels at dam sites in Australia that are not near faults, and is expected to result in a general reduction in ground motion levels at dams not near faults estimated for all Risk Assessments, and for Deterministic Assessments for all consequence levels except Extreme Consequence. For the latter, the ANCOLD (2018) guidelines will tend to increase existing SEE ground motion estimates for both of the methods used to estimate the safety evaluation earthquake (SEE). By requiring the use of the Deterministic SEE if it is larger than the probabilistic SEE, and by requiring use of the 85th fractile of the Probabilistic SEE if it is larger than the Deterministic SEE, the ANCOLD (2018) guidelines for Deterministic Assessments are much more conservative than the ICOLD and NZSOLD guidelines for Extreme Consequence dams, especially at those located near faults.
Trustpower’s Mahinerangi Dam in New Zealand’s South Island is a concrete arch and gravity abutment dam built in 1931, subsequently raised in 1946 and strengthened with tie-down anchors in 1961.
This paper discusses a 3D finite element analysis of the dam and the predicted performance of the arch section under Safety Evaluation Earthquake (SEE) loading against identified potential failure modes.
Current guidelines and recent seismic hazard assessments recommend earthquake loadings higher than what was originally accounted for in previous decades. A Comprehensive Safety Review identified stability under SEE loading as a potential deficiency, so a programme of works was commenced to evaluate and better understand the seismic risk by using modern day tools and technology to evaluate the dam against current performance standards.
The final model incorporated the results of extensive laboratory testing, high-resolution LiDAR survey data and dynamic calibration using ambient-vibration monitoring. Motion recordings across the face of the dam during the 2016 Kaikōura earthquake were also used to validate the model. The reservoir has been explicitly modelled together with the opening, closing and sliding of contraction joints and the foundation interface. This allowed the modelling of permanent displacements and the redistribution of loads within the dam under SEE loading, which had been shown to be an important behaviour from the previous stages of analysis.
Since publication in 2003, the ANCOLD Guidelines for Risk Assessment have reached broad acceptance and use in Australia. In practice, dam owners use the principles of risk assessment to drive business investment decisions. As the guidelines undergo revision, it is timely to assess whether our practices need to evolve to more holistically consider all types of consequences, rather than our current focus on loss of life, in decision-making. This paper aims to prompt dam owners and consultants alike to re-assess our focus on loss of life in risk assessment decision-making, and whether we should more meaningfully consider alternative or broader indicators.
An industry survey was undertaken which found that large dam owners are generally happy with the current system of dam safety decision making. However, the survey responses did identify difficulties in relation to justifying investment below the limit of tolerability that are subject to ALARP principles. In a small number of cases, dam owners found it difficult to justify investment when life safety was not important.
Building on the industry survey and subsequent discussions with practitioners, this paper discusses how the current approach to risk based decision making may result in sub optimal decision making. Further it is discussed how there is an important role that economics should play in providing a universally accepted framework for assessing trade-offs and providing consistent evidence to support decision making.
The paper describes the development of UK guidance on reservoir drawdown capacity. The guidance provides for a consistent thought process to be used in determining the recommended capacity. A basic recommended standard is proposed for embankment dams which varies with the consequences of failure of a dam. The drawdown rate for the highest consequence dams is 5% dam height/day with an upper limit of 1m/day. Engineering judgement is used to vary this standard allowing for ‘other considerations’ including the vulnerability to rapid dam failure, surveillance and precedent practice. A different approach is proposed for concrete/masonry dam, which considers the prime purpose of drawdown being to lower the reservoir in a reasonable timeframe to permit repairs rather than rapid lowering to avert failure. The UK approach is compared with that used in Australia and suggestions made for where its use may be appropriate.
This paper will explore the differences in pore pressures resulting from saturated and unsaturated seepage (pore pressure) analysis. It will also evaluate some conventional recommendations, such as the inclusion of essential components of the embankment dam and omission of inessential components. In addition, the identification of inessential components will be discussed.
Finally, pore pressures obtained from these analyses will be compared to monitoring data in order to identify the most appropriate seepage (pore pressure) model.
In conclusion, advantages and disadvantages of each method will be discussed and recommendations will be provided in order to gain the most appropriate results.
The results of this paper can be used for designing new embankment dams or safety reviews of existing dams, particularly when there is lack of reliable monitoring data.