Total-load sampling has been a perpetual problem in sediment monitoring. Usually a combination of bed-load sampling devices, suspended load suction samplers, and some kind of flume, for total flow rate, is used. Total-load, sediment-sampler-design concepts that can perform all three of these functions are proposed. The resulting designs would require installation at sites that can provide a step-overfall height about equal to the maximum channel flow depth. The concepts are simple, but appear to have been overlooked or ignored for the past many decades, and are based on a moving conveyor belt that is long and wide, with many slots, all of the same size, onto which the stream to be sampled discharges. All flow drops through the slots, and with equal sized slots each must catch a similar proportion of the total flow. Hence, only one slot needs to be collected. As a practical extension it is proposed to replace the conveyor belt with a rack having several slots that represent a short section of the total conveyor belt that is large enough so that the flow does not notice the missing belt parts. This rack is then traversed back and forth on a track through the falling nappe. Laboratory tests of this proposed sampling-assembly rack indicated that the number of the required slots is related to the channel depth and the sum of the slot openings. When the rack is composed of sufficient slots so that the slot-width sum is more than half the channel overfall depth, the system undersampled from 0 to 2% but when there are insufficient slots whose sum represents less than one-third of the overfall depth, the system undersampled by over 8%. The concepts are extended to the condition with a stopped belt where several sampling-slot groups are equally spaced beneath the overfall. A “test of concept” sampler assembly of the stopped-belt idea was built and tested. The sample catch across the stream was within about 4% of expected, offering a total load sampling system where motorized equipment is difficult to install, or electric power is not available. Design and construction suggestions are presented. The catch rate can be small enough to facilitate convenient flow measurement of the catch, which can be converted to total streamflow without the need for separate channel flow measurements.

Tony L. Wahl, P.E., M.ASCE; Albert J. Clemmens, Ph.D., P.E., M.ASCE; John A. Replogle, Ph.D., D.WRE, P.E., M.ASCE; and Marinus G. Bos, Dr., Ir. Hydraulic Engineer, U.S. Dept. of the Interior, Bureau of Reclamation, Water Resources Research Laboratory, Denver, CO. E-mail: twahl@ do.usbr.gov Research Leader and Research Hydraulic Engineer, USDA-ARS, U.S. Arid Land Agricultural Research Center, Maricopa, AZ. Research Hydraulic Engineer/Collaborator, USDA-ARS, U.S. Arid Land Agricultural Research Center, Maricopa, AZ. Professor, International Institute for Geo-Information Science and Earth Observation/ITC, The Netherlands.

Clemmens, A.J., T.L. Wahl, M.G. Bos, and J.A. Replogle. 2001. Water Measurement with Flumes and Weirs. International Institute for Land Reclamation and Improvement, Wageningen, The Netherlands, ISBN 90-70754-55-X This manual describes how to design, construct, and operate long-throated flumes to measure and control water flow in open channels. Broad-crested weirs with a streamlined flow contraction are members of the long-throated flume family and can be analyzed with the techniques described herein. Broad-crested weirs are particularly well adapted to irrigation canals, while long-throated flumes are well adapted to natural streams. The range of conditions over which flow can be measured with long-throated flumes and broad-crested weirs is unlimited. A large variety of examples are presented. The manual includes a revised mathematical model for designing a flow-measuring structure in any open channel to satisfy user-specified hydraulic boundary conditions and design requirements. Upon completion of the design, the model will compute the head-versus-discharge relation and the hydraulic energy losses for the flume or weir. The computer program presented in this publication can accommodate a wide variety of structure and channel shapes as well as many different input and output units. This version greatly expands upon the capabilities of previously published programs.

Many water users have been strongly encouraged to reduce the amount of their diversions through improved water management practices. To identify opportunities for improving a user's water management practices, an estimate of the accuracy of the flow rates and accumulated volumes withdrawn by the user must be made. Typically, water users measure flow at a given interval (e.g., daily) and then integrate them over time to estimate the volume measurement. In the past, estimates of the uncertainties for individual flow measurements were obtained using standard statistical methods; however, estimates of the error for the accumulated volumes have been difficult. In this study, a procedure is outlined that identifies the random and systematic error components for individual flow measurements. These estimates of the individual error components can be combined to yield an estimate of the uncertainty for an individual flow rate measurement. The uncertainty for the accumulated volume that passes through a given site is then estimated by using 1) the uncertainty associated with the individual flow measurements, 2) the errors associated with integration over time, 3) and the errors introduced into the system during the time between individual flow measurements. These uncertainty estimates can then be used to improve water users' flow measurement strategies.

Long-throated flumes provide economical and flexible water measurement capabilities for a wide variety of open-channel flow situations. Primary advantages include minimal headloss, low construction cost, adaptability to a variety of channel types, and ability to measure wide ranges of flows with custom-designed structures. Long-throated flumes can be calibrated using computer programs that apply proven hydraulic theory, thus eliminating the need for laboratory calibration. This paper describes recent advances in the computer software available for design and calibration of long-throated flumes, and highlights two potential flume design issues that have recently come to light. Specifically, there have been field observations of an apparent suction effect below flumes having a vertical drop at the end of the control section; this can lead to significant differences between the actual and theoretical head-discharge rating curve. Second, in flumes that are primarily width-contracted—as opposed to those with a sill that creates a bottom contraction—there is potential for non-modular flow in the throat section when the width to crest length ratio is large, despite the fact that traditional hydraulic theory predicts critical flow. Again, this can cause a significant discrepancy between the actual and theoretical rating curves of a structure.

Pumps are widely used to lift water into canals, and usually spill directly into the canal. Upstream pipe fittings frequently produce a distorted how profile that is detrimental to the proper installation and operation of common pipe meters. Thus, meter calibration may be necessary when the pipe length upstream of the meter is less than recommended. Meter calibrations under field conditions can be difficult and expensive. A simple pitot tube system that can be clamped to the outlet of the pump discharge pipe was built and tested for measuring and calibrating pump discharges under field conditions. It is used to detect the velocity at several points across the pipe diameter; distorted profiles can be measured. Using this information, the meter technician can determine whether a correction in the meter coefficient will suffice and if flow conditioning equipment is working effectively. The system can be constructed using common shop techniques and standard small pipe fittings. Most previous pitot systems required special ports in the pipe or used special concentric tube constructions that are difficult to build.

A brief historical look at the origins of flow metering in the last millennium is offered that touches on some of the developments used today in open-channel and pipeline flows. While the basic physical principals for measuring flows have remained basically unchanged, the range and accuracy of monitoring these physical effects have been vastly improved by developments in electronics and computer technology. For example, the ultrasonic properties of a fluid medium have long been recognized, but only in the last decade have practical and inexpensive means to exploit these properties become available. Some of the newer developments in the last quarter of the past century include the computer-calibrated, long-throated flume. A recent extension to the computer-calibrated flumes repertoire include the adjustable-throat flumes that aid in placement in earthen channels because it virtually eliminates concern for vertical elevation of the throat, which can be adjusted to accommodate ditch flow conditions after installation. Other recent developments discussed include the vortex-shedding meters; ultrasonic flow meters, both the Doppler type and the transonic types.

Practical technologies can encourage farmers to adopt practices that support sustainable irrigated agriculture. Important among these are convenient water measurement and control techniques. Many simple constructions or operating procedures are available that can bring considerable convenience to farmers and irrigation delivery system operators. Some are new technologies and some are improvements on older technologies. Many can be implemented with small expense. Some are superior replacements for current practices. The techniques and devices discussed included: (a) accurate and convenient zero setting for weirs and flumes (b) pressure-transducer field checks, (c) easy-to-use scales for orifice and Venturi meters, (d) flow-profile improvers to assist accurate meter operations in irrigation pipelines, (e) floor sills and wave suppressors for canals that usually flow at variable depths of flow, (f) water surface slope measurements–based on static-pressure tubes, and (g) field checks of flow velocity profiles to evaluate flow conditioning using rising-bubble techniques for flow-profile visualization. Many of the concepts are demonstrated in a summary illustration showing several items in a typical stilling well and broad-crested weir (long-throated flume) that need attention, and offers suggestions for correcting the deficiencies.

The utility of developing vertically-adjustable flumes for canal flow measurement systems, is important to several problems in field practice. One involves the perception by some canal water users that flumes and weirs significantly and harmfully restrict flow. These perceptions, correct only sometimes, occur often enough to cause resistance to flow measurements and impedance to proper irrigation management. Parshall flumes and Cutthroat flumes require ponding depths upstream equal to about 40% of head reading while long-throated flumes and the related broad-crested weirs require only 10 to 15%. The actual head drop through all of these flumes is greater than hydraulically necessary for all but the maximum design discharge. The highly obvious excessive ponding is often misunderstood as a harmful restriction to flow. Also, the velocities at the low flows are reduced by this excessive ponding, which can aggravate sediment accumulation. The system described herein allows control of the ponding restriction from nearly zero to just enough restriction to gain measurement control of the flow at nearly all flow rates in the design range of a particular size. This reduces the amount of visible restriction. For a small structure size, with a control section less than 1 m wide and flowing under a head of less than about 25 to 30 cm, this restriction is about 10% to 15% of the head reading, or about 3 to 5 cm at maximum head. This maintains relatively high velocities in the approach channel for assisting sediment movement. The device is applicable to measuring flow rates in unlined and lined canals. The system described permits adjustment of the canal flow levels, reduces the perception of ponding, and minimizes the induced sediment problems.

In van den Broek, B. J. (Ed.), Dutch experience in irrigation water management modelling. Wageningen, Netherlands: DLO Winand Staring Centre

Overshot gates are becoming increasingly popular for controlling water levels in open channels because of the ability of the gates to handle flow surges with limited depth changes. While water level control is useful, operators also need to know the flow rate at each gate to better operate the system. Empirical equations were developed from laboratory testing to predict the discharge over leaf gates. These equations can ac curately determine the flow rate in the field of a properly ventilated free-flow overshot gate to within 6%. These equations are valid for values of measured head to weir height ratios less than 1.0 and for gate angles between 16 and 63, spanning the typical oper ating range. Additional equations can be used to pre dict the discharge of a submerged overshot gate; how ever, the degree of accuracy was not clearly defined.

Long-throated flumes and broad-crested weirs are becoming the preferred measuring devices for use in irrigation canals. A computer program for calibrating these flumes and weirs has been available since the early 1980s. But because the flume dimensions were program inputs, design was largely by trial and error. FLUME 3.0 is a new computer program that not only provides flume calibrations, but also assists the user with flume design. Design criteria include submergence, accuracy, freeboard, and Froude number limitations. The first two criterion are examined at both high and low flow rates to assure proper performance of the structure throughout the range of flows to be expected. FLUME 3.0 is menu driven for easy use and maintains a database of flume designs and as-builts. The program also has graphical-data entry screens so that profile and cross-section dimensions can be visually verified. This helps the user to avoid errors in data entry. FLUME 3.0 is available along with a user's manual through the International Institute for Land Reclamation and Improvement or through Water Resources Publications.