Within the framework of risk assessment and risk management, determination of the failure probability of dams plays an important role. In this paper, the probabilistic approach to hydraulic failure elaborated by the author in the nineties will be developed further and refined. Furthermore, it is shown how the random event “gate failure” can be taken into account when calculating the overtopping probability of barrages. / Im Rahmen von Risikobetrachtungen spielt die Bestimmung der Überflutungswahrscheinlichkeit eine wichtige Rolle. Im vorliegenden Beitrag wird der probabilistische Ansatz, welcher in den neunziger Jahren vom Autor erarbeitet worden ist, weiterentwickelt und modifiziert. Außerdem wird gezeigt, wie das zufällige Ereignis „Versagen von Verschlüssen“ in die Berechnung der Überflutungswahrscheinlichkeit von Stauanlagen einbezogen werden kann.

Thesis (Ph. D.)--Brigham Young University. Dept. of Civil and Environmental Engineering, 2004. / Includes bibliographical references (p. 223-229).

All life depends upon water. For man, however, water can supply the power necessary to relieve him from the drudgeries of life and give the time and means to enrich his existence beyond the bounds set by the mere need to survive. Large scale hydropower makes a significant contribution to the total generated power of the developed world. The means of harnessing water power for large flows and heads above three metres are efficient and well established. There remains however, the fact that many people, especially in developing countries live adjacent to water courses and tidal estuaries where heads of 2 metres or less exist but are not exploited. The reasons for this are often that practical machines do not exist or are insufficiently well developed to be used. Much effort has been made in recent years to develop equipment suitable for small, low-head sites in remote places. The writer has worked on three such devices in both the laboratory and on site - these are the Salford Transverse Oscillator, the AUR Water Engine and the Underwater Motor. This thesis describes these machines in detail together with other contemporary devices. The initial chapter contains an assessment of machines used in the past to exploit low head sites. There is little doubt that some of these older machines, suitably brought up to date in design and materials, would be more suited to the needs of developing peoples than many of the esoteric devices described in later chapters. The thesis is concerned mainly with the need to provide a means of generating power rather than the assessment or suitability of a site for power generation. The writer draws on experience gained in the laboratory, and as a Mechanical Engineering Consultant for small installations in the United Kingdom and overseas.

Appraisal and design of small-scale hydro power plants requires a knowledge of hydraulics, hydrology, civil, mechanical, and electrical engineering, and basic economics. Further, small hydro is site specific in nature and marginal from an economic view point. Methods of appraisal and design are required therefore that will keep engineering fees to a minimum and yet still achieve a reliable evaluation of scheme potential and economics. In this context it should be appreciated that small hydro is not large hydro scaled down, and that small hydro needs its own experts (Ref. 1). This thesis considers techniques for appraisal of small hydropower schemes, the selection and specification of scheme components, their costing and economic evaluation. These appraisal techniques are subsequently applied to regional assessment of small-scale hydro-electric potential in the U. K, and to the development and application of a new type of ultra low-head hydropower generator called the Salford Transverse oscillator (STO). Although this work is predominately concerned with assessment of scheme potential in the U.K., it also draws on experience gained by the writer during short visits to India and Nepal, and during a six month design appraisal for rehabilitation of mini-hydro schemes in Sri Lanka (Ref. 2).

Although flow deflectors are routinely placed in streams to create fish pool habitat, there is dissent in the literature as to what constitutes for them an appropriate design height to ensure their success. Our objectives were to (1) examine the temporal evolution of bed topography around submerged deflectors in a laboratory flume, for different deflector heights and discharge; and (2) quantify the flow dynamics in their vicinity. Results indicate that, all else being equal, higher deflectors (characterized by less overtopping) produce a greater volume of fish pools. In contrast, over a fixed deflector design, a higher discharge (synonymous with greater overtopping) results in larger pools, because flow velocities are then faster. We also observed differences in the temporal behaviour of scour: that next to high and medium height deflectors is self-similar, but not that for the shorter model. Our results challenge the notion that low deflectors are ineffective, provide information on the development of scour next to submerged structures, and reveal key differences between the geometry of the scour holes engendered by the different deflector height models.

Deep interceptor drains are commonly used to control canal seepage in southern Alberta, Canada. Recently, shallow grid drainage was introduced. A study was initiated in 1987 to assess the effectiveness of grid drainage to intercept canal and natural groundwater seepage and reclaim the resulting saline affected land. / Using a groundwater flow model, MODFLOW, it was found that a single deep interceptor drain would have failed to intercept all canal seepage and maintain the water table downslope of the canal below the 1.0 m design water table depth. Conversely, simulations indicated that with a grid drainage system, all canal and natural groundwater seepage would be intercepted and the water table would remain below the design water table depth, with or without irrigation recharge that would maintain a steady state salt balance. / The benefits of fall irrigation were demonstrated using three test plots near the canal.

A three-dimensional numerical model (PHOENICS) was used to investigate the role of stream deflector angle and length on the flow field in a rectangular laboratory flume. Subsequent bed topography surveys were performed to examine the role of obstruction angle on scour hole development over time. The model was capable of predicting laboratory velocity and turbulent kinetic energy measurements, performing better for flow over a flat stable bed than over a deformed sand bed. A new method of incorporating complex bed topography into a structured Cartesian mesh was developed in the process. Flow field properties such as dynamic pressure, velocity amplification, separation zone length and width, and downwelling extent and magnitude were found to be strongly dependent on deflector geometry. Equilibrium scour hole depths and geometry are also angle-dependent. A predictive equation was produced explaining the rate at which scour holes reach equilibrium, and compared well with existing literature. Finally, a method was developed whereby characteristics of the flow field over a flat, stable bed could be used to predict equilibrium scour hole geometry.

This dissertation investigated the impact of the Surficial Fine-Grained Laminae (SFGL) upon stream biogeochemical cycles to constrain stream C and N budgets. Collection and analysis of 8 years of transported sediment elemental and isotopic signatures, weekly, from a SFGL dominated stream, a novel dissolved C and N dataset, statistical and time-series analysis of sediment and dissolved data, and development of a comprehensive modeling framework that couples hydrodynamics, sediment, C and N biogeochemistry, and stable isotope sub-models to simulate fluvial C and N budgets was used. SFGL C modeling suggests benthic particulate C stocks and transport vary seasonally and annually but are in a state of long-term equilibrium which is governed by negative feedback mechanisms whereby high POC export due to extreme hydrologic events and high frequency hydrologic events reduces benthic particulate C stocks and inhibits benthic particulate C growth. Model distribution fitting suggests transported particulate C in SFGL streams is Gamma distributed; in which statistical moments are governed by variability of the SFGL. Stable isotope un-mixing of the bed source suggests that the SFGL has varying levels of carbon quality seasonally and annually, in which non-equilibrium conditions stem from extreme depositional events. Coupling stable isotope mass balance and SFGL fractionation processes into water quality modeling frameworks, reduced uncertainty of the C budget by nearly 60%, suggesting algal sloughing constitutes nearly 40% of the total organic C budget, shifting the balance from dissolved C to particulate C dominated. Time series analysis of the eight year dataset suggest nitrogen dynamics in the SFGL dominated stream were consistent with existing conceptual models when algal biomass is the prominent organic matter source in the SFGL, but contradicts conventional wisdom in winter through late spring when abiotic sorption appears prominent. The development of a new numerical model to simulate the fluvial N budget couples this new conceptual model of SFGL stream N dynamics to isotope mass-balances and C dynamics in order to provide a comprehensive management tool for restoration engineers. Meta-analysis and upscaling of results for regional to global scales will enable researchers to place the role of the SFGL in a broader context.

This dissertation describes investigations of fully turbulent decelerating hydraulically roughbed flow over gravel and the development of technology to measure turbulence and associated sediment transport in streams. Theory is developed for predicting velocity distributions in simple uniform flow using the asymptotic invariance principle and tested using laboratory and field collected data. A mixed scale is developed that accounts for bed derived turbulent structures throughout the flows depth and is used to parameterize the external boundary’s effect on the flow for the logarithmic and outer layers. The asymptotic invariance principle and similarity analysis is conducted for the equations of motion in the outer region of decelerating flow over gravel to define equilibrium conditions for this class of flows with the velocity scale is the freestream velocity. The combination of time series and time averaged statistical analysis of turbulent flow is used to elucidate the structure of flow under decelerating conditions. Time averaged statistical measures of turbulence confirm results of others for higher Froude number approaching transcritical and time series analysis shows the effects of decelerating flow on turbulence to be frequency dependent. Wireless velocity sensors were developed and found capable of measuring time averaged velocity and able to resolve macroturbulence from time series data. A semi-theoretical model of elastic deformation of cantilever beams under hydraulic forcing was coupled with circuit theory to develop a calibration procedure for the VBS that requires only three measurement points, one of which is at zero velocity. Light based sensors are developed to estimate light attenuation in water for ecological research or estimating sediment concentration in water. A semi-theoretical scaling of light attenuation and sediment properties was developed which predicts light attenuation from sediment properties. The combination of new theory on open channel velocity, turbulent structure and field sensors for measuring turbulence and sediment offers the possibility to extend our laboratory knowledge to realistic flow situations.

Water distribution modeling for hydraulics and water quality is an important tool for managing system performance of water utilities. An important component of a water distribution model is the calibration of a network model with field data in the real world system. The calibration effort requires a protocol or selection criteria for the location of field measurements that best support the calibration effort. A water distribution model was constructed at the University of Kentucky hydraulics lab for the purpose of investigating the performance of water distribution models. The lab model contains numerous hydraulic (pressure, flows) and water quality (concentrations) sensors for measuring system characteristics. This research work utilizes the lab model to compare hydraulic calibration using pressure heads from hydraulic data, velocities from water quality data, and combinations of both as the basis of calibration. It also presents an example of a small experimental system where velocity data as a basis for a calibration effort and pressure based data as a basis doesn’t converge to the same solution. The results of the research demonstrate the necessity of using both velocity & pressure data for hydraulic calibration to avoid compensating errors.