Item

Abstract

The motion of conductive sea water through the earth's magnetic field generates electromagnetic (EM) fields through a process called motional induction. Direct measurements of oceanic electric fields can be easily converted to water velocities by application of a first order theory. This technique has been shown to obtain high quality velocities through instrumental advances and an accumulation of experience during the past decades. EM instruments have unique operational considerations and observe, for instance, vertically-averaged horizontal velocity (from stationary sensors) or vertical profiles of horizontal velocity (from expendable probes or autonomous profiling floats). The first order theory describes the dominant electromagnetic response, in which vertically-averaged and vertically-varying horizontal velocities are proportional to electric fields and electric currents, respectively. After discussions of the first order theory and deployment practices, operational capabilities are shown through recently published projects that describe stream-coordinate velocity structure of the Antarctic Circumpolar Current, quickly-evolving overflow events in the Denmark Strait, and time-development of momentum input into the ocean from a hurricane. A detailed analysis of the Gulf Stream at its separation point from the continental slope serves as a case study for interpreting EM measurements, including the incorporation of geophysical knowledge of the sediment. In addition, the first order approximation is tested by the many features at this location that contradict the approximation's underlying assumptions: sharp horizontal velocity gradients, steep topography, and thick and inhomogeneous sediments. Numerical modeling of this location shows that the first order assumption is accurate to a few percent (a few cm s-1) in almost all cases. The errors in depth-varying velocity are <3% (1-3 cm s-1), are substantiated by the direct observations, and can be corrected by iterative methods. Though errors in the depth-uniform velocity are <2 cm s-1 (<10%) at all locations except for the upper continental slope, where apparent but unresolved meander events in water shallower than 500 m can generate depth-uniform errors of order 30%, there are not sufficient observations to confirm these errors directly. Errors in the first order approximation at this location show no non-linear increase due to the joint effect of steep topography and horizontal velocity gradients. Using motional induction in the world's oceans, aside from stationary measurements when depth-uniform ocean currents meander across topography, these results suggest that the first order approximation is accurate to within 1-2 cm s-1 or less in almost all regions of the ocean, an error similar to the instrumental accuracy of EM instruments. © 2011 Elsevier Ltd. All rights reserved.