Report of the Round Table Session
Cato, D.1*, Prior, M.2*, Anderson, M.3, Binnerts, B.2, Eleman, A.4, Erbe, C.5, Folegot, T.6, Popper, A.N.7, Radford, C.8, Sigray, P.3, and van der Schaar, M.9
1 Defence Science and Technology Organisation & University of Sydney Australia
2 TNO, The Netherlands
3 FOI Swedish Defence Research Agency, Sweden
4 Sualti Arastirmalari Dernegi (Underwater Research Society), Turkey
5 Centre for Marine Science & Technology, Curtin University, Perth, Western Australia
6 Quiet-Oceans, France
7 University of Maryland, USA
8 Leigh Marine Laboratory, Institute of Marine Science, New Zealand
9 Laboratory of Applied Bioacoustics, Technical University of Catalonia, BarcelonaTech (UPC), Spain
This report can be referenced as:
Cato, D., Prior, M., Anderson, M., Binnerts, B., Eleman, A., Erbe, C., Folegot, T., Popper, A.N., Radford, C., Sigray, P., and van der Schaar, M. (2015). Report of the Ambient Noise Session, oceanoise2017, Vilanova i la Geltrú, Barcelona, Spain, 10-15 May (Editors Michel André & Peter Sigray). Retrieved from http://2019.oceanoise.com
“Ambient noise in a world of soundscapes”
The topic of underwater ambient noise received considerable study during and after the Second World War and much of the work done on the subject was motivated by the requirements of the military sonars that played an important role in anti-submarine warfare during the Cold War.
In Principles of Underwater Sound, the textbook perhaps most widely used by underwater acousticians and sonar engineers, Robert J Urick describes ambient noise in the following terms
‘It is that part of the total noise background observed with a non-directional hydrophone which is not due to the hydrophone and its manner of mounting called ‘self-noise,’ or to some identifiable localized source of noise. It is what is “left over” after all identifiable noise sources are accounted for.’
Most ambient noise prediction graphs are derived from the work of Wenz (1962) who stated that he did not include “Obvious noise from marine life, ships, and other sources of intermittent local noise…”
This view of ambient noise as a ‘miscellaneous’ set in the taxonomy of underwater acoustics is associated with a feeling that ambient noise is a nuisance – a hindrance to the reliable detection of signals of interest. In Computational Ocean Acoustics, their definitive text on the numerical modelling of underwater sound, Jensen, Kuperman, Porter and Schmidt reflect this view via their definition of ambient noise:
‘It is the interference with respect to detecting or measuring signals.’
On the other hand, many of those who worked on ambient noise saw it as a fascinating study of sounds in the ocean. Several hundred papers have been published and these provide a substantial resource for studying underwater sound. Over the same period, there were many studies of the sounds of marine mammals, fishes and invertebrates. Some scientists worked on both these and ambient noise, and the two areas sometimes overlapped. Generally, ambient noise studies focussed on the sustained components of the noise rather than individual transient sources. For example, it would include the noise of distant shipping or a chorus of fish calls but not the noise of a passing ship or an individual fish.
In more recent times, underwater sound has become a topic for studies of the effects of human activity on ocean life. Concerns regarding the impact of sonars on marine mammals arose in the last decade of the twentieth century and this now includes other sources of anthropogenic noise such as seismic surveying, pile driving and shipping. In the study of the impact of sound on marine life, while those parts of the noise field generated by human activity are regarded as detrimental, naturally occurring noises are often considered as valuable features of the ocean environment that should be protected against human intrusion. Thus, while a sonar engineer would consider both distant shipping and diffuse marine biota as sources of unwanted noise, researchers concerned with ocean ecology would agree on the former but treat the latter as something to be protected.
It is in this context that ambient underwater sound has begun to be considered in terms of ‘soundscapes.’ This term originated in air acoustics where it is used as an aural analogy of the visual ‘landscape.’ City planners and architects may consider the mix of audible sounds as an important part of the urban environment that has a strong effect on quality of life. In national parks and nature reserves, the absence of loud, anthropogenic noises, such as from aircraft overflight or nearby industrial activity, is a feature of the natural environment that must be preserved through planning and restrictions on development. Some usage of ‘soundscapes’ includes auditory perception of the animals involved, particularly human perception in the case of urban soundscapes.
Although the concept of underwater soundscapes is often considered in the context of the potential impact of sound on life, the situation is complicated by the likely wide variation in perception between species. This might lead to a suggestion that any given ocean environment will have not one but many soundscapes: one for each species present. Ambient noise may also be seen in the same way, since two sonars with different frequency or angular resolutions could be said to experience different ambient noise even when deployed at the same time and place. It might be said that there is general soundscape or ambient noise for a particular location and time from which the specific noise or soundscape for a specific species can be obtained, including the effects of perception.
The concept of the soundscape, however, has much wider interest than just the context of the impact of sound. It is the acoustic environment in which animals live and which they exploit, i.e. the role of sounds in their ecology, and perception plays a part in this. In this respect, the most appropriate analogy appears to be with terrestrial soundscape ecology, an area of study that is more advanced than the underwater equivalent and provides a valuable source of ideas to apply to the marine environment.
As usage of the term ‘underwater soundscape’ grows, a question arises: what is the relationship between the soundscape and ambient noise? Activities such as a draft ISO standard on underwater acoustics terminology have attempted to produce definitions for the terms but no such effort is made here. Instead, some of the issues relevant to the relationship between ambient noise and soundscapes are simply discussed.
Parameters and metrics
Usually, measurements of underwater sound focus on acoustic pressure, but studies of fishes and invertebrates indicate that they sense particle motion as well. Thus proper characterisation of an underwater soundscape must include both pressure and particle motion. Although most studies of ambient noise consider acoustic pressure, the concept can just as easily be defined in terms of particle motion. The same applies to soundscapes, so the need to measure particle motion as well as pressure cannot be considered to be a fundamental difference between the two terms.
There are various ways of measuring and characterising ambient noise and soundscapes. The simplest is just to describe the sound field statistically, such as in terms of the percentile levels. Modelling and experiments to understand the components of the noise or soundscape provide more information and understanding. Modelling can be used to aid in understanding the sound fields. Statistics provide a description like climate statistics but they do not provide understanding or the ability to forecast e.g. if you understand how sea surface noise depends on wind speed you can forecast it from wind speed forecasts.
The interference caused by ambient noise is determined not only by its level but also by its directivity. A strong interferer may have little impact on the detection of a signal if its influence can be reduced by spatial filtering techniques such as beamforming. Thus the directivity of sound in both the vertical and horizontal is an important property of ambient noise. Similarly, the directivity of sound may be an important part of the ocean soundscape and studies have suggested that marine organisms may use sound directivity to navigate.
Environmental and ecosystem characterisation
The soundscape can be considered an important property of the ocean environment and measurements of it may be used to infer an ecosystem’s ‘state of health’. Various measures derived from the acoustics of the soundscape are being developed to categorise the state of health of an ecosystem. These include acoustic complexity, acoustic richness and acoustic entropy. Acoustic measurements can be used to demonstrate the presence of sound-generating species such as marine mammals or fishes. Conversely, consistent failure to detect such signals may be used as evidence of species’ absence. Such acoustic survey data can be extended to infer the presence of species that are connected to the sound-generating animals via predator-prey relationships. For example, a recording of sperm-whale clicks could be used as evidence to support the hypothesis of the presence of the squid on which they feed.
Similarly, it has long been known that measurements of ambient noise can be used to obtain environmental description data. Jensen et al precede the definition of ambient noise quoted above with the additional definition
‘It is the resident acoustic field in the ocean and hence also a diagnostic of the ocean environment.’
and the temporal correlation structure of ambient noise has been shown to contain environmental information that reveals seabed sediment layering and Green’s functions describing waterborne propagation paths.
The field of ‘ocean soundscapes’ is relatively new and no formal definition of the term exists or is proposed here. The field of ‘soundscapes’ in the terrestrial environment is much older and the terrestrial concepts could be applied under water. There are several definitions of ‘soundscape’ and the most relevant to this conference would appear to be ‘soundscape’ ecology. There are also difficulties in arriving at a definition of ‘ambient noise’, though there is less variation. The organisers of conference sessions on ambient noise and ocean soundscapes will, for the foreseeable future, struggle to identify a clear boundary between their domains.