Calibration of marine autonomous acoustic recorders

Abstract

Absolute measurement of sound in the ocean is becoming increasingly ubiquitous, a common driver being the requirement for monitoring marine noise in support of regulation for the protection of the marine environment. In these measurements, the performance of the measurement system is a crucial factor governing the quality of the measured data. In recent years, there has been an increasing use of autonomous acoustic recorders for absolute in-situ measurement of sound in the marine environment. The technology has developed rapidly utilising recent improvements in mobile microprocessors and data acquisition systems, and currently there are a number of commercial off-the-shelf units available to the user. Whilst offering the enhanced ability to monitor acoustic signals autonomously for extended periods, such recorder units introduce a number of measurement and calibration challenges in addition to those associated with the calibration of individual hydrophones. These include the need to treat the autonomous acoustic recorder as a complete system, providing a traceable calibration which includes the hydrophone, hardware signal processing, the digital-to-analogue conversion, and software processing used to produce the sound data file. Because recorders often use archival storage of digital data, there is typically no access to “live” analogue electrical signals, requiring significant modifications to the standardised calibration procedures adopted for individual hydrophones. An additional problem can occur for recorders where the hydrophone is fixed rigidly to the end of the recorder body. In such cases, diffraction and scattering of the sound around the recorder body may influence the frequency response and directivity at kilohertz frequencies. At lower frequencies, the recorder performance may also be influenced by resonances in the recorder body. Other key performance characteristics include the system self-noise (important for measurement of low-level signals such as may be present in ambient ocean noise), and the dynamic range (important for measurement of high-amplitude impulsive sound sources). In this paper, methodologies are presented for the calibration and characterisation of autonomous recorders to determine the key acoustic performance characteristics, including the absolute system sensitivity as a function of frequency and direction, and the self-noise of the hydrophone and system. Consideration is given to effects due to the proximity of the recorder body to the measuring hydrophone on the frequency and directional response of the overall system. Examples of the results obtained are given and a discussion is presented of the implications of system performance on the quality of the measured data. The work described here aims to provide the traceability for the absolute measurement of sound in the ocean using autonomous recorders so that noise monitoring strategies and in-situ source characterisations are underpinned by robust metrology. The need for enhanced traceability is particularly acute at frequencies below 1 kHz where high-amplitude anthropogenic sources of greatest concern radiate much of their sound energy, and NPL now offers a calibration service to provide traceability for users of marine acoustic recorders over this vital low frequency range. Finally, this paper provides a discussion of how the work described here feeds into a European initiative to provide improved traceability and more robust metrology infrastructure to catch up with the rapidly evolving legislative framework.