Capability of pure water calibrated time-of-flight sensors for the determination of speed of sound in seawater

Abstract

Results from a laboratory investigation of commercial time-of-flight sensors designed for oceanographic in situ speed-of-sound measurements are presented. An older analog model and three modern digital units were calibrated in pure water in the temperature range of 1 °C to 50 °C. The speed of sound (w) was measured in salt solutions of varying concentration (NaCl, MgCl2, Na2SO4) and in samples of original and diluted North Atlantic seawater at atmospheric pressure. A high reproducibility of the time-of-flight readings was found, resulting in sound speed standard deviations in pure water between 0.033 m s−1 and 0.015 m s−1, depending on the individual instruments. This depicts the potential of the time-of-flight method. However, although simultaneously calibrated, the measurements revealed systematic speed-of-sound differences between the different sensors which exceeded the reproducibility by about 1 order of magnitude. As the cause of these deviations could not be determined within this study, this exhibits a constraint for the uncertainty of measurements in seawater relative to pure water. In comparison with recent equations this has been estimated at 0.3 m s−1 (200 ppm) in original seawater. In seawater at temperatures >40 °C and in diluted seawater the results indicate relevant differences from the recent Thermodynamic Equation of Seawater–2010 equation of state.