Metric expressions

From Jennifer Amaral et. al (2020) JASA 147:4

I spent last week in Denver at an Acoustical Society of America (ASA) meeting – one of the scientific meetings I have been attending regularly since 1998. The field of acoustics intersects so many disciplines– from speech and language and architectural acoustics, to Non-linear Phononic Materials and Sonochemistry. So the Society meets twice a year just to keep up.

I typically spend most of my time in the Animal Bioacoustics sessions, occasionally dipping into the Underwater Acoustics and Acoustical Oceanography for the physics of the marine acoustical habitats that our work encompasses.

I chaired a session on “Windfarms as an Anthropogenic Noise Source” in which we had a set of papers and presentations from the general to the particular. My overview presentation on wind farm noise, from siting surveys, to installation, and operation – and where above and below water sounds may have impacts on the biological habitats – was in the “general” category.

There were a couple of presentations from Bureau of Ocean Energy Management (BOEM) and the National Oceanographic and Atmospheric Administration (NOAA) on the regulatory agency’s use of data toward assuring best practices in the fields I had identified in my overview, and there were a couple of more-focused presentations on the particulars of monitoring the acoustical environment of turbines and windfarms.

I found all of the presentations informative. My role as Session Chair allows me to “cherry pick” and invite speakers whose research I find intriguing and useful. But there were also a few presentations offered that I hadn’t identified in the literature, advancing our collective understanding in useful ways.

What I find fascinating are all of the ways that data can be gathered, manipulated, and applied to yield informative ways of understanding the various dimensions of any inquiry. Twenty years ago researchers were constrained by computational power and data storage densities.

This was when acoustical sampling equipment involved data storage on hard drives, which had to be spun up to take “acoustical snapshots” of the environment because they did not have the capacity to collect acoustical data continuously for more than a few hours. And one of the problems was how to isolate the noise of the spinning hard disk from the hydrophones…

Another problem back then was how to process and make useful the terabytes of acoustical data being generated. The running joke at the time was that there were not enough undergrads to synthesize all of the data being collected.

Now there are all sorts of methods of grinding up data and spitting out useful information. We are no longer constrained by comparing the dimensions of time, frequency, and amplitude, to display them in visual x, y, and z, graphs. While the graphs are still visual, any quantifiable dimension can be used as a filter or variable to modulate any other dimension.

Jennifer Amaral from Marine Acoustics Inc. illustrated this well in her presentation of impact pile driving noise using two-dimensional scatter plots with axes being Sound Exposure Level (SEL) against Pulse Duration, Rise and Decay Time against distance, and perhaps my favorite, Kurtosis against Distance.

Her graphs yielded intriguing information about pile driving which in turn clarifies the impacts of pile driving on marine habitats.

There were many more sessions presenting information in novel ways. Manipulating time, space, sample count, and energy into visual fields, giving researchers methods to reach our collective imagination – toward a deeper understanding of the world we inhabit.