Big Data and the reassembly of reality I

Kurhaus Schreveningen

Data is not much more than snapshots of something we want to understand taken through a familiar lens. Of course this presupposes a few conditions: that we’ve identified a relationship we want to explore, and that the data we capture will somehow help us assemble something meaningful about that relationship.

This comes up because I’ve recently returned from a scientific conference in den Haag, Netherlands focused on Aquatic Noise. This international conference comes up every few years since 2007, bringing together marine biologists, physicists, engineers, academics, regulators, and industry and military stakeholders to present our work, talk shop, and push the scientific conversation down the road a bit more.

The field has changed much in the last decade.  I don’t believe that any of us imagined that the inquiry into marine bioacoustics would have expanded so much. The early meetings – over a decade ago, orbited around noise sources (shipping, sonar, and seismic surveys) and impacts on various taxa – mostly marine mammals. The presenters at this recent meeting explored a cornucopia of ideas and original thinking.

Ships, sonar, and seismic surveys remain the leading noise sources, but presenters at this meeting are digging much deeper into impacts on invertebrates and fish. And while these critters are typically not “regulated” (subject to government oversight), focus on these taxa infers a more systematic approach to the ocean noise issue.

Of course what makes this possible is “Big Data” and big data storage.

In the early days of nature sound recording, researchers would need to locate and identify what they wanted to record, deploy their equipment, and hope to get a good recording on tape. These days they can identify a place they want to understand, deploy the equipment, and return later to retrieve the recordings. The bulk of the work then lies in synthesizing and analyzing the data.

Done ‘manually,’ means listening in real time. If some study requires the deployment of equipment in a number of locations, this can eat up a lot of time. But this is where sound recordings become “data;” elements in the recordings are considered in settings, in relationship with, and contributing to a sound field. These data then can be identified through the use of classifiers, and sorted out for analysis through various computer algorithms.

So in earlier times a bioacoustician might seek out a specific animal and try to capture some representative sounds; they can now record a “soundscape,” providing remote listeners with an experience of a location that would otherwise be inaccessible or improbable without the equipment (our Arctic Soundscapes would qualify for this).

Which brings us around to our “reassembly of reality.” We can only speculate how far or deep this might go, or the socio-psychological impacts of being able to dial up a “virtual” experience of any remote location or habitat. Will we cherish these remote locations curated by scientists if they become part of our common experience, or will we love them less because they are no longer our own?

The Aquatic Noise conference in den Haag, Netherlands was a week of some ~150 presentations, and encounters with some 280 scientists, researchers, regulators, and stakeholders. I am still synthesizing all that I was able to take in.

Needless to say, with a week of this, there was way too many ideas and conjectures to unpack in a single ~500-word essay, so in an attempt to be informative without being too verbose, I’ll be breaking up my takeaways into a number of newsletters we’ll be putting out over the next couple of weeks.