BRP Job Introductions: Research Analyst

A busy day at the "pentagon".
A busy day at the “pentagon”.
Building Marine Autonomous Recording Units (MARU) - our work begins at the engineering bench.
Building Marine Autonomous Recording Units (MARU) – where every detail counts.

Summer is almost here and for me, that means barbeques, picnics, hanging out with old friends and meeting new ones. Inevitably the question “what do you do for a living” comes up and I’ve gotten used to some inquisitive faces when I say that I listen to whales.

My fiancée loves to quip that I am the worst marine biologist. Whether it’s because I live 300 miles from the ocean, have only snorkeled once in my life, or am very much afraid of sharks, I can see where she is coming from. But, while you won’t find me chasing whales in an inflatable zodiac boat with a biopsy dart gun to collect important genetic data, or researching hormone levels in pregnant whales while sharing a boat with a dog that is trained to smell whale poop from miles away (seriously, this is a thing!), the work I do as a research analyst back in the office is a vital part in advancing our understanding in how whales communicate and move throughout their habitat.

Two compact flash cards that were removed from an acoustic buoy and are awaiting data extraction.
Two compact flash cards that were removed from an acoustic buoy and are awaiting data extraction.

Here at BRP, we are fortunate to have a great field team that will go out to sea and deploy and recover our acoustic buoys called MARUs (Marine Autonomous Recording Units). When these buoys are returned, the analysis process begins. Data need to be downloaded, formatted into sound files from binary code, and stretched together so that all the audio channels line up with the right time. All these data have to be archived and stored on tapes and hard drives before we even start having fun looking for whales.

It would take me 6 months of time to listen to 6 months of audio data, so we have to come up with smarter and faster ways to browse through the sounds we collect. We create spectrograms, which are visual representations of the sounds that we can go through and look for patterns. Humans are great at pattern recognition, especially when different species of whales sound and look similar, or are far away and faint, but even then, it takes a long time to visually look through all of that sound data. So, we rely on some powerful computing tools to automatically detect potential whale calls. As we analysts verify or reject the detections, we help train and improve the computer’s capabilities in finding whales.

Using Raven Pro to look at a spectrogram and waveform of a humpback whale vocalizing.
Using Raven Pro to look at a spectrogram and waveform of a humpback whale vocalizing.
Research analyst Karolin Klinck hard at work looking for dolphin clicks and whistles.
Research analyst Karolin Klinck hard at work looking for dolphin clicks and whistles.

Each deployment of sound data we collect and analyze tells a different story. We can use the presence information for different whale species to create seasonal patterns, predict habitat usage by correlating other environmental factors such as sea surface temperature, measure ambient noise levels in the environment and track changes over time, locate vocalizing whales in an area and make recommendations to oil and gas and wind energy developers to minimize danger to migrating animals.

BRP is home to a diverse set of people with many different skills, and we all contribute to the mission of the program and the Lab of Ornithology. Take a look at our Meet the Team page to see more of what others are doing here and next time you meet a marine biologist miles away from the ocean, you might not find it so odd!

– by Daniel Salisbury, Research Analyst