Journal of Hydraulic Engineering

The U.S. Geological Survey has developed new standard rating tables for use with Price type AA and pygmy current meters, which are employed to measure streamflow velocity. Current-meter calibration data, consisting of the rates of rotation of meters at several different constant water velocities, have shown that the original rating tables are no longer representative of the average responsiveness of newly purchased meters or meters in the field. The new rating tables are based on linear regression equations that are weighted to reflect the population mix of current meters in the field and weighted inversely to the variability of the data at each calibration velocity. For calibration velocities of 0.3 m/s and faster, at which most streamflow measurements are made, the new AA-rating predicts the true velocities within 1.5% and the new pygmy-meter rating within 2.0% for more than 95% of the meters. At calibration velocities, the new AA-meter rating is up to 1.4% different from the original rating, and the new pygmy-meter rating is up to 1.6% different.

The accuracy of cross-channel integrated sediment transport of bed material is determined with an elaborate set of field measurements in the Waal River, The Netherlands. The measurements were done during a discharge wave in the upstream part of the river, which has a bimodal sand-gravel bed. The sampling strategy should take both spatial and temporal aspects into account to obtain maximum accuracy. Presence of moving bedforms, differences in bed-sediment grain size in the cross section, and presence of preferential transport lanes dictate that at least five subsections for sampling in the cross section are necessary. The accuracy of cross-channel integrated bedload transport depends mainly on the measurement strategy. An uncertainty of <20% (bedload) and 7% (suspended load) of cross-channel integrated sediment transport is shown to be feasible if 30 samples of bedload and two vertical profiles of suspended bed-material load are taken in one subsection, provided that the cross section of the river is divided into at least five subsections. The samples in one subsection should be distributed over the length of the bed form. Changes of discharge during the measurements cause systematic differences between the subsections. To minimize this uncertainty a compromise between the spatial and temporal accuracy is necessary. Therefore, when only one vessel with instruments is available for doing the measurements, the number of sampling positions and subsections must be reduced if the rate of change of discharge is large. Based on the results a prediction method is given to estimate the feasible accuracy in the planning phase of future campaigns, and the necessary time and financial investment for that accuracy.

Calibration of a water distribution network is intended to develop a model that mimics field conditions under a range of demand distributions. In this paper, a three-step calibration procedure is developed that considers the uncertainties in measurement and estimation and provides a measure of the quality of the calibration. The approach can also be used to identify preferable conditions for data collection. The procedure's steps are parameter estimation, calibration assessment, and data collection design. Parameter estimation considers input uncertainty and the resulting uncertainty in model parameters. Calibration assessment analyzes the propagation of the parameter errors on model predictions. The trace of the covariance matrix of the predictive heads is used to measure the model uncertainty. Based on this uncertainty and using a sensitivity-based heuristic analysis, data collection experiments can be designed for systemwide tests and critical pipe for individual pipe tests. An example system is analyzed to select calibration demand conditions.

The common field problem of identifying and locating impervious aquifer boundaries is generally solved by analyzing the pump-test data. The estimate of the distance between the image well and the observation well (r′) is required for locating the boundary. Popular methods of estimating r′ either use the concept of fully developed two straight lines or rely on curve matching. Curve-matching methods are subjective and involve errors due to personal judgment. A very long duration pump test is required in order that the second straight line is developed. In many cases, the first straight line is not developed at all, especially when the observation well is at a greater distance from the pumped well. In this paper, a derivative-based robust optimization method has been proposed for the identification of aquifer parameters (transmissivity, T and storage coefficient, S) and r′ from the drawdowns observed at an observation well. It returns unbiased estimates of the parameters (T, S, and r′) even with a poor initial guess. The method can yield the estimates of the parameters from shorter duration pump-test data, thus saving time and money involved in conducting the pump test with the purpose of identifying r′. It is observed that the hydraulic diffusivity (T/S) of the aquifer and r′ can be estimated even when the constant rate of pumping is not known.

The level of stratification within ponds, lakes, and other wetlands is a strong determinant of water quality. It is shown here how the stratification in shallow systems (to a few meters deep) can be described in terms of two nondimensional parameters, which are functions only of the water depth, turbidity, mean wind speed, and mean solar radiation. The functional forms have been derived from flows generated by a coupled atmospheric boundary-layer–water-body model. The results generally provide very satisfactory estimates of the stratification in a shallow lake in southeastern Australia and are also shown to be consistent with an existing criterion for the initial onset of stratification in shallow systems. It is suggested that with further validation, the simple empirical description of stratification conditions provided here might be usefully incorporated into existing wetland management strategies.

This paper presents the results of an experimental investigation on suction removal of sediment from between armor blocks/stones placed on a loose bed. The process of suction has been investigated. It was found that the vortices that form in the holes between the armor blocks are key to the process. The sediment swept into these vortices is entrained into the main body of the flow by these same vortices (the suction removal of sediment from between the armor blocks). The critical condition for the onset of suction was determined. It was found that the onset of suction is governed by two parameters: (1) the Shields parameter (based on the sediment size); and (2) the ratio of sediment size to stone size, d/D. The variation of the critical Shields parameter for suction as a function of d/D was determined for a broad range of the parameter d/D, namely, 0.001 ≲ d/D ≤ 1. The timescale of the suction process and the downward displacement of stones (the general lowering of the armor layer) have also been investigated.

Flows in variably saturated media are of profound interest to numerical analysts, engineers, and scientists because of not only the challenge they pose as a result of their highly nonlinear constitutive relations but also their importance in many fields of engineering such as drainage, irrigation, hydrology, environmental, soil, and petroleum engineering. In this paper, the Picard and Newton-Raphson (N-R) algorithms are incorporated into the Green element method (GEM) to simulate these flows. The GEM offers a viable means of implementing the singular boundary integral theory so that the theory is more generally applicable and computationally efficient. Here GEM discretizes the integro-differential equation in space with suitable polygonal elements and in time with a difference scheme, and the system of nonlinear discretized element equations are linearized by the Picard and N-R algorithms. Calculations carried on three numerical examples of infiltration into unsaturated soils in two spatial dimensions indicate better convergence of the N-R algorithm than the Picard algorithm at comparable computational cost.

This paper deals with the identification of an irrigation system in the presence of not completely known withdrawals. The modeling approach is based on a time-implicit finite-dimensional model of simplified equations governing open-channel hydraulics. The diffusive wave model is used with a parameterization based on large rectangular geometry and solved using the Muskingum-Cunge method. A two-stage iterative method is proposed to identify the dynamic parameters and the withdrawal parameters. In the first stage, only the parameters governing the withdrawals are estimated, and in the second stage, the estimated withdrawals are used to estimate the parameters of the dynamics. The simulation studies performed on simulated and real data show the effectiveness of the identification method.

The aim of this work is to highlight the effects of cross-sectional shape on dam-break wave propagation along channels by the solution of 1D conservative equations assuming a power-law variation of the channel width. An exact Riemann solution that allows a second-order accuracy of the solution for the power-law section shape is provided and is applied to the dam-break problem in valleys with different shapes but the same dam area. The streamflow state variables upstream of the bore and the bore speed for some typical sectional shapes (rectangular, triangular, concave, and convex banks) are determined as functions of variable flow depth differences and of the power law index.