Page 32 - Spring 2018

Exploring the Ocean Through Soundscapes
Soundscape Modeling
Thus far we have discussed decomposing the sounds measured at points in the ocean to explore what sourc- es are present and how they shape the acoustic space used by humans and marine life (Figure 1). However, it is also possible to estimate a soundscape by combining the acoustic signatures of regional sources. Soundscape modeling is the process of composing the sounds at a re- ceiver based on assumed sources, source locations, move- ments, and acoustic propagation conditions. Researchers will model soundscapes to test detection, classification, and localization algorithms in controlled conditions or to predict the potential effects of human activities. Simplify- ing assumptions have traditionally been made to reduce the computational burden, especially for the sea surface noise from wind and waves and the acoustic propagation loss. As computer speeds increase, more advanced sound source and propagation models enable increasingly rapid algorithm development and improved understanding of sound propagation, and provide better information for decision makers during the evaluation of permit applica- tions requesting the approval for incidental exposure to marine life during industrial, scientific, or military activi- ties (Aulanier et al., 2017; e.g., Figure 5).
Acoustic Measurements
for Conservation
Understanding the effects of noise on marine life motivates many marine soundscape measurements. The effects of noise are often grouped into four categories: (1) death and injury, (2) physiological effects, (3) behavioral disturbance, and (4) masking of sounds. Protecting marine life from death and injury has been the focus of recent industry and government funding. As a result, we know more about what sounds levels and metrics predict injury, especially to marine mammals, than those associated with behavioral changes and masking. Two threshold measurements are used to estimate the on- set of injury in marine life. The peak sound pressure level of the impulse (e.g., explosion, pile-driving strike) is used to assess the possibility of physiological damage to tissues (e.g., barotrauma in fishes). The amount of sound energy that can damage hearing in marine life is estimated by the SEL, which accumulates sound energy over time (Popper et al., 2014; National Marine Fisheries Service, 2016). The SEL in the marine environment is a complicated dynamic related to the source distance, acoustic propagation conditions, and the overlap between the frequency content of the source
Figure 5. Modeled SPL from a snapshot of automated identification system vessel locations, which was generated as part of research into the cumulative effects of current and additional projected vessel traf- fic at the port of Prince Rupert, British Columbia, Canada. Figure provided by JASCO Applied Sciences, Nova Scotia, Canada, and the Prince Rupert Port Authority.
and the hearing sensitivity of the receiving animals. For hu- man sound exposure, we use the familiar A-weighting, and similar weighting functions are proposed for five groups of marine mammals (Southall et al., 2007). We do not yet un- derstand the hearing of fishes, sea turtles, and invertebrates sufficiently to propose weighting functions for these groups.
Studies of behavioral disturbance and auditory masking are increasing now that the acute effects of noise are better understood. These studies are directly associated with the concept of a soundscape; how does the marine life interpret and react to sound? Most studies of behavioral disruption relate the reaction to the sound pressure level at the time of the reaction. Much additional work is required to better understand what measurements, including particle motion, are appropriate for understanding behavioral disruption on most taxa. Masking occurs when the ability to detect or recognize a sound of interest is degraded by the presence of another sound, the masker (Dooling et al., 2015). Under- water signals can be masked by natural components of the soundscape such as sea ice, wind-wave interactions, rain, and distant animal choruses or vocal bouts as well as by an- thropogenic activities. Whereas studies of sound-induced
30 | Acoustics Today | Spring 2018