Kinematic and dynamic estimates from electromagnetic current meter data

Abstract

The dynamic response of electromagnetic current meters (manufactured by Marsh‐McBirney, Inc.) has been clarified through a comprehensive laboratory measurement program combined with a thorough literature review. Elucidation of the behavior of these flow meters under a variety of dynamic conditions has been neglected in the past. Since flow past a spherical body has considerable hydrodynamic complexity for different dynamic conditions, a careful laboratory study was carried out for pure steady, pure oscillatory (horizontal plane), and combined steady/oscillatory conditions at two test facilities. Test results indicate that flow meter behavior under pure steady flow is reasonable in the absence of high levels of free‐stream turbulence, with an rms error of 1–5 cm/s. These errors could be reduced with a higher‐order polynomial regression fit. Pure oscillatory response was also reasonable, with rms errors of 1–2 cm/s, and sensitivity which is correlated with the oscillatory Reynolds number (Re)0, and the Keulegan‐Carpenter number (A/d, where A is the oscillation excursion and d is the probe diameter). Combined steady/oscillatory flows degraded current meter performance with larger residual errors (1–6 cm/s) and significant differences in sensitivity (up to 20%). Horizontal cosine response showed systematic deviations from pure cosine behavior, with a notable intercardinal undersensitivity and cosine “shoulder” at lower Reynolds numbers. Error analysis shows these current sensors are adequate for many kinematic measurements but may lead to excessive errors when using velocity to calculate dynamical quantities (such as bottom friction, Reynolds stress, or log‐layer friction velocities). A careful error analysis must precede any use of these meters for estimating dynamical quantities. These studies pointed out a potential difficulty in using these meters in areas of large ambient turbulence levels (20% turbulent intensities), which are characteristic of many near‐bottom shallow water environments. Further study is needed to clarify this behavior.

HF radar comparisons with moored estimates of current speed and direction: Expected differences and implications

“…The principle underlying the HF Doppler radar technique has been investigated theoretically and experimentally over the past several decades (see Barrick [1978] and Prandle [1989] for reviews). The basic premise of the scattering mechanism is that energy of radar pulses are backscattered from the movement of the ocean surface by resonant surface waves or "Bragg waves."…”

Measuring water velocity in highly turbulent flows: field tests of an electromagnetic current meter (ECM) and an acoustic Doppler velocimeter (ADV)

“…This study was particularly motivated by the need to assess the reliability of available instruments for the characterization of high velocity, highly turbulent and non-uniform flow environments such as rough step-pools, cascade rivers and gravel-bed rivers in flood. The performance of ECMs has been previously tested in the laboratory (Aubrey and Trowbridge, 1985;Lane et al, 1993) and in coastal (Soulsby, 1980;Guza et al, 1988) and riverine environments (Lane et al, 1998;Roy et al, 1996b). Similarly, ADVs have been tested in the laboratory (Rodriguez et al, 1999;Voulgaris and Trowbridge, 1998;Finelli et al, 1999), in the ocean (Elgar et al, 2001;Smyth and Hay, 2003;Elgar et al, 2005) and in rivers (Lane et al, 1998).…”

“…Similarly, ADVs have been tested in the laboratory (Rodriguez et al, 1999;Voulgaris and Trowbridge, 1998;Finelli et al, 1999), in the ocean (Elgar et al, 2001;Smyth and Hay, 2003;Elgar et al, 2005) and in rivers (Lane et al, 1998). Only two of these tests, those by Aubrey and Trowbridge (1985) and Voulgaris and Trowbridge (1998), were able to control their flow environment such that the performance of the instrument could be assessed against known values. These laboratory experiments, however, were typically conducted in still water tanks or smooth-walled flumes and cannot be directly applied to quantify sensor accuracy in field deployments due to the significant effects of turbulence scales and intensity (Guza et al, 1988).…”

“…Voltage differences captured by the ECM probes are relatively weak, and these voltages are amplified prior to data storage. Amplification can be adjusted during calibration, but differences between sites mean that small offset errors are difficult to eliminate (Aubrey and Trowbridge, 1985), especially for measurements in natural rivers (Guza et al, 1988). At higher values of U, the ECM recorded higher values than the ADV, although only one point has an error larger than 10%.…”

“…ECMs have been shown to have good low-frequency response up to 3·0 m/s (Elgar et al, 2001). Aubrey and Trowbridge (1985) tested for the occurrence of flow separation around the spherical head and found a non-linear behavior in the streamwise component of the flow, with a point of inflection close to U = 0·80 m/s. However, this error was only 1.2% at U = 2·0 m/s and the probes used in that study were three to seven times larger than those utilized for these field tests.…”