Journal of Hydraulic Engineering

This paper is written to mark the hundredth anniversary of the birth of Hans Albert Einstein (1904–1973). It casts his career as that of the archetypal researcher protagonist determined to master intellectually the way water flows and conveys alluvial sediment in rivers. In that effort, Einstein personified the mix of success and frustration experienced by many researchers who have attempted to formulate the complicated behavior of alluvial rivers in terms of mechanically based equations. His formulation of the relationship between rates of bed-sediment transport (especially bedload transport) and water flow comprised an innovative departure from the largely empirical approach that prevailed at the time. He introduced into that relationship the emerging fluid-mechanic concepts of turbulence and boundary layers, and concepts of probability theory. Inevitably the numerous complexities attending sediment transport mire formulation and prompt his use of several approximating compromises in order to make estimating bed-sediment transport practicable. His formulation nonetheless is a milestone in river engineering.

Measurements of bed-load and suspended-load transport rates were carried out successfully at four cross sections of the Nile River, in Egypt, along the entire length from Aswan to Cairo using a mechanical sampler called the Delft Nile Sampler. The measured transport rates were compared to similar data sets from two other large scale rivers: the Rhine-Waal River in the Netherlands and the Mississippi River in the USA. The bed-load transport rates in the Nile River and in the Rhine-Waal River are in very good agreement. Comparison of suspended transport rates in the Nile River and in the Mississippi River shows that both data sets are complementary, revealing a very consistent trend of suspended transport against current velocity; suspended transport is roughly proportional to $(V_{\mathrm{av}}{)}^{3\mathrm{to}4}\text{.}$ Three formulas for the prediction of bed-load transport were tested using the Nile data: Meyer-Peter–Muller, Bagnold, and Van Rijn. The prediction formula of Van Rijn produced significantly better results than the other two formulas; the average relative error was about 60%. The formula of Van Rijn was modified to extend it to conditions with slightly nonuniform sediment mixtures by introducing a correction factor for the bed shear parameter. Based on a limited number of flume experiments, the correction factor was found to be dependent on the characteristics of the sediment mixture $(d_{10}\text{,}$ $d_{50}\text{,}$ $d_{90}\text{,}$ and ${\sigma }_g)\text{.}$ Comparison of bed-load transport measured in the Nile River with computed transport rates of the modified formula showed improved results; the average relative error decreased to about 30%. The formulas of Bagnold and Van Rijn were also used to compute the suspended transport rates in the Nile River. The computed transport rates were found to be within a factor of 2 of measured values; the formula of Bagnold performed slightly better. The total load transport formula of Engelund–Hansen was also successfully used (computed values within a factor of about 2 of measured values).

A numerical method based on optimal control theories for identifying Manning’s roughness coefficients (Manning’s n) in modeling of shallow water flows is presented. The coefficients are difficult to be determined especially when the spatial variation is significant, and are usually estimated empirically. The present methodology is applied to determine the optimal values of the spatially distributed parameters, which give least overall discrepancies between simulations and measurements. Through a series of systematic studies to identify the n values in both a hypothetical open channel and a natural stream stretch, several identification procedures based on unconstrained and constrained minimizations are analyzed. It is found that the limited-memory quasi-Newton method has the advantages of higher rate of convergence, numerical stability and computational efficiency. Although the identification of Manning’s n is chosen as an example, the identification methods can be applied to numerical simulations of various flow problems.

Turbulence characteristics of hydraulic jumps with Froude numbers of 2.0, 2.5, and 3.32 are presented. A Micro Acoustic Doppler velocimeter was used to obtain measurements of the velocities, turbulence intensities, Reynolds stresses, and power spectra. The maximum turbulence intensities and Reynolds stress at any section were found to decrease rapidly from the toe of the jump towards downstream within the jump and then gradually level off in the transition region from the end of the jump to the friction dominated open channel flow downstream. The maximum turbulence kinetic energy at each section decreases linearly with the longitudinal distance within the jump and gradually levels off in the transition region. The Reynolds stress and turbulence intensities within the jump show some degree of similarity. The dissipative eddy size was estimated to vary from 0.04 mm within the jump to 0.15 mm at the end of the transition region. The dominant frequency is in the range from 0 to 4 Hz for both horizontal and vertical velocity components.

A detailed experimental study was conducted to investigate the hydraulics, interfacial instability, and mixing in the supercritical region of exchange flows downstream of a sill crest. Measurements of the velocity field and the interface position were obtained using flow visualization and particle image velocimetry. Large periodic fluctuations in the measurements of the flow rate and interface position were caused by the Kelvin–Helmholtz (KH) instabilities at the interface as well as the internal seiche. These KH instabilities caused entrainment of fluid from the upper layer into the lower layer, with the entrainment coefficient considerably larger than the values for gravity currents.

A discontinuous Galerkin method for the solution of the dam-break problem is presented. The scheme solves the shallow water equations with spectral elements, utilizing an efficient Roe approximate Riemann solver in order to capture bore waves. The solution is enhanced by a projection limiter that eliminates spurious oscillations near discontinuities. The main advantage of the model is the flexibility in approximating smooth solutions with high-order polynomials and resolving at the same time discontinuous shock waves. Furthermore, the finite element discretization is capable of handling complex geometries and producing correct results near the boundaries. Both the h- and p-type extensions are investigated for the one-dimensional dam break, and the results are verified by comparison with analytical solutions. The application to a two-dimensional dam-break problem shows the efficiency and stability of the method.

This paper presents a detailed analysis of different deepwater blowout scenarios using the Clarkson deepwater oil and gas model (CDOG). In CDOG, hydrate formation, hydrate decomposition, gas dissolution, nonideal behavior of gas, and possible gas separation from the main plume due to strong cross-currents, are integrated with the jet/plume hydrodynamics and thermodynamics. CDOG takes into account unsteady-state three-dimensional variation of ambient currents and density stratification. Detailed comparisons between CDOG simulations and deepspill field experiments have been published. The model is used to simulate 30 deepwater blowout scenarios based on realistic cases and the results are analyzed in this paper. The scenarios demonstrate the differences in plume behavior due to different ambient conditions, different types of gas, possible hydrate formation, and variations in gas-to-oil ratio. Some key findings of these analyses follow. Oil droplet sizes affect the oil surfacing time significantly. For oil-only blowouts, the ambient conditions do not affect the oil surfacing time significantly, but the location and the size of the slick are affected. For oil and gas mixes, the surfacing time is not sensitive to the type of gas in the mix, but is somewhat dependent on the ambient conditions. In none of the cases simulated here, did free gas reach the water surface. While changing the release temperature had only an insignificant effect on the model results, changing oil type or gas-to-oil ratio did affect the model results. The analyses are useful to engineers/scientists and administrators.

This research aims to advance current knowledge on cluster formation and evolution by tackling some of the aspects associated with cluster microtopography and the effects of clusters on bedload transport. The specific objectives of the study are (1) to identify the bed shear stress range in which clusters form and disintegrate, (2) to quantitatively describe the spacing characteristics and orientation of clusters with respect to flow characteristics, (3) to quantify the effects clusters have on the mean bedload rate, and (4) to assess the effects of clusters on the pulsating nature of bedload. In order to meet the objectives of this study, two main experimental scenarios, namely, Test Series A and B (20 experiments overall) are considered in a laboratory flume under well-controlled conditions. Series A tests are performed to address objectives (1) and (2) while Series B is designed to meet objectives (3) and (4). Results show that cluster microforms develop in uniform sediment at 1.25 to 2 times the Shields parameter of an individual particle and start disintegrating at about 2.25 times the Shields parameter. It is found that during an unsteady flow event, effects of clusters on bedload transport rate can be classified in three different phases: a sink phase where clusters absorb incoming sediment, a neutral phase where clusters do not affect bedload, and a source phase where clusters release particles. Clusters also increase the magnitude of the fluctuations in bedload transport rate, showing that clusters amplify the unsteady nature of bedload transport. A fourth-order autoregressive, autoregressive integrated moving average model is employed to describe the time series of bedload and provide a predictive formula for predicting bedload at different periods. Finally, a change-point analysis enhanced with a binary segmentation procedure is performed to identify the abrupt changes in the bedload statistic characteristics due to the effects of clusters and detect the different phases in bedload time series using probability theory. The analysis verifies the experimental findings that three phases are detected in the bedload rate time series structure, namely, sink, neutral, and source.

This paper offers solutions for drawdowns due to intermittent pumping cycles or cyclic pumping, which are high accuracy approximations of the series of Theis functions superimposed in time. The proposed approximation formulas are an improvement over the earlier works. The earlier approximations are valid only if the number of pumping cycles is greater than 10 and involve gamma functions that are less convenient to evaluate than the rational approximation formulas offered in this paper. The proposed approximations are valid for any number of pumping cycles and involve simple functions that can be computed even using a calculator. The drawdown functions are defined for the drawdowns at the end of pumping or shutoff periods. The proposed expressions for these functions are also suitable for the estimation of aquifer parameters by plotting the observed drawdowns on semilogarithmic paper. Procedures for estimation of storage coefficient and head loss at the well from cyclic pumping drawdowns are not available. This paper also offers procedures for the estimation of transmissivity, storage coefficient, and head loss at the pumped well from the observed intermittent (cyclic) pumping drawdowns.

An equation is presented for estimating the time of concentration of overland flow over bare packed sandy soil surface. Experiments were performed in a field catchment exposed to rainfall during the wet months of April to September 2001. Time of concentration formula was evaluated using the relevant data obtained from the experiments. The time of concentration formula was compared with the time of concentration equation obtained by Morgali and Linsley using computer synthesized overland flow in their study, and found to be very closely correlated. Furthermore, the investigation was extended to a few other catchments with differing slopes, lengths and rainfall intensities (hence Reynolds numbers) and very close agreement was indicated between the two results.

The characteristics method by using the cubic-spline interpolation is comparable to the Holly–Preissmann scheme in solving the advection portion of the advection–diffusion equation. In order to conduct a cubic-spline interpolation, an additional constraint must be specified at each endpoint. In general, four types of endpoint constraints are available, i.e., the first derivative, second derivative, quadratic, and not-a-knot constraints. The goal of this paper is to examine each type of endpoint constraints. Two hypothetical cases are used to conduct the investigation. Among the four types of constraints examined herein, the not-a-knot constraint and the first derivative constraint with high-order finite difference approximation yield the best results. However, as far as accuracy and simple implementation are concerned the not-a-knot constraint should be the best choice in solving the advection–diffusion equation