CalSound: Visualizing the Sounds of the Ocean
Nine Fung Fellows developed CalSound, a website for visualizing soundscape data, to help marine resource managers understand ocean sound.
By: Leena Elzeiny, Edgar Hildebrandt Rojo, Audrey Kuptz, Anmol Parande
What can you hear right now? Anything from a person walking past your door to the gentle hum of the electronics around you composes your acoustic environment, also called your soundscape. Oceans have a soundscape too, and they include all the biologic, anthropogenic, and geologic noises in the sea.
The protection of natural soundscapes is necessary because of the impact noise has on ocean life. Sounds within a landscape can provide essential insight into the health of the ecosystem. For example, a study published in 2012 observed the effect of shipping and airgun noise on fin whales, finding that their song features decreased and their communication changed to compensate for increased background noise. Noise pollution also creates an avoidance response, as the study observed whales leaving areas with lots of noise.
Because of sound’s impact on the ecosystem, federal organizations like BOEM and the National Marine Fisheries Service regulate man-made sound sources to minimize the disruption of natural habitats for endangered species. However, adding sound to environmental impact assessments is still relatively new, so there is room for growth in understanding how manmade developments impact ocean soundscapes.
This past spring 2021 semester, our team of nine Fung Fellow undergraduates in partnership with the National Oceanic and Atmospheric Administration (NOAA) worked to advance marine soundscape visualization. Our central goal was to discover how we might help marine resource managers see the impacts of noise pollution in the ocean by providing a clear, holistic way of synthesizing information and visualizing ocean soundscapes. By the end of our design process, we created CalSound, a website for displaying and contextualizing soundscape data.
To begin understanding the needs of marine resource managers, we conducted 12 user interviews. Our interviewees spanned several organizations, from National Oceanic and Atmospheric Administration (NOAA) to Bureau of Ocean Energy Management (BOEM) to the California Coastal Commission, and included people with a variety of uses for sound data such as environmental impact assessments, evidence in litigation, and impact mitigation. We had three main takeaways from the interviews:
- There is an unmet need of measuring environmental sound levels versus species-specific sound data. Currently, many datasets and tools focus on particular species, rather than the health of the ecosystem as a whole.
- Both space and time are heavily important. People care not just when a sound was produced, but also where and, if possible, by whom.
- Most of the value of soundscape data comes not from the sound itself, but, rather, the contextual layers, such as species habitats and shipping distributions, that help explain why the data is important and what it means.
These three takeaways drove how we decided to visualize the data. The best way to combine these was to create an interactive website which marine resource managers could use to guide their decision making process when evaluating and approving new ocean developments.
Since both space and time are important, a map-based interface with a sliding scale to mark time form the basis of the tool. The sound data measured by a hydrophone at the given time is displayed in a popup which originates at the place the sound was recorded (Figure 1).
Contextual information such as shipping density, species habitats, and species detections are added as layers on top of the base map (Figure 2). Since the data was collected by drifting buoys, it captures the environmental sound levels which researchers need.
In designing these features, the most critical component was making sure the user experience was as intuitive and clear as possible for marine resource managers. Particularly with applications with several layers of data, one difficulty is ensuring that visualizing one layer doesn’t impact the user’s ability to read and understand data from other layers. Much of our effort went into designing these layers so they were visually distinct, but still useful for the user.
With the primary features in place, our next goal was to get people onto the website. With this in mind, we applied, and got accepted, to the Biennial Conference on the Biology of Marine Mammals, which is dedicated to “enhancing collaboration, sharing ideas, and improving the quality of research on marine mammals within the scientific community.” We also recently applied to the American Geophysical Union (AGU) conference held in December 2021 under the Imagining Ocean Science: Public Engagement Using Visualization and the Arts session. In these conferences, we hope to meet like-minded individuals, passionate about marine ecology, who will find use in our tool and provide us with invaluable feedback on ways to adapt CalSound to fit different people’s needs.
In the meantime, while we wait for the upcoming conferences, we have added new features to the website (information tags, tutorial, small user experience improvements, etc.) and re-interviewed past interviewees to better understand how our tool can coexist with other tools used today. Our next major goal is ensuring the longevity of the website. Ideally, we want users without much technical expertise to be able to use and visualize their own datasets. To accomplish this, we will be including a “Upload Data” feature which will allow users to visualize their own data sets.
If you would like to learn more about soundscapes and how they are analyzed, our team also wrote an extensive literature review on the field of soundscapes and how they can be utilized within our oceans. The literature review, as well as the sources we used, can be found on our website. We are excited to continue working on CalSound and see where it takes us!
As we continue to develop CalSound, we are grateful to the Fung Fellowship Conservation + Tech teaching team for presenting the opportunity to work with NOAA this past semester, as well as for all of the assistance they provided throughout the design process. We’d like to thank our NOAA partner representatives on this challenge, Acoustic Ecologist Anne Simonis and Wildlife Research Biologist Shannon Rankin, for their guidance and connection to key resources as we developed our website and disseminated our findings. We would also like to thank Anne Simonis for all of her expertise and encouragement to continue this project beyond the timeframe of the semester’s design challenge.