Measurement of turbulence in ocean boundary layers

This paper is concerned with the acquisition of information on turbulent mixing in the boundary layers of the ocean, that is, near the sea surface, near the sea bed, in coastal regions and in estuaries. In such problems, natural fluctuations with periods ranging from a few tenths of a second to a few weeks are of greatest concern, with specific emphasis on continuous time series over periods of at least 30 minutes. For useful measurements in this area, accuracies of 0.1 cm/sec often are required. Presently, my group is using mechanical, acoustic and electromagnetic meters. Only the mechanical units meet the desired specifications, and they have too high a threshold velocity (1.0 cm/sec), are too large when deployed in groups of three, and are too often contaminated by filaform seaweed to be considered satisfactory. However, it is not just the suitability of the flow sensors themselves but also the nature of the devices on which they are mounted and the care with which they are deployed that determines the accuracy of current measurements in general and turbulence data in particular. Precise flow meters located in the wake of a frame part or deployed in the disturbed pressure field caused by a support arm produce results that are just as inaccurate as judiciously spaced sensors with less desirable response characteristics. Furthermore, unresolved frame motion can render velocity fluctuation measurements useless. When deployed from a properly designed support device and with a good understanding of their assets and shortcomings, instruments such as our pulse output mechanical current meters often can be used to obtain results of the scientifically required accuracy. When used in this fashion, however, it is essential that the investigator have a thorough understanding of the instruments based on a comprehensive testing program. Also he must have a good theoretical understanding of their operation and of the flows to be examined in order to permit exploitation of the test results and to avoid misinterpreting their implications in regard to field deployment.