By Turner Johnson
The group of species known as “right whales” was once seen as the quintessential (read: “right”) whale to hunt; they forage on the sea’s surface close to shore, are friendly towards boats, and float when killed. Beginning in the 1800s, right whales were a main casualty in the North American whaling industry, nearly hunted to extinction. Today, however, the name “right whale” refers to three different species: the North Atlantic right whale (E. glacialis), the North Pacific right whale (E. japonica), and the Southern right whale (E. australis). Climate change has exacerbated the visual difference between Atlantic species by contributing to the emaciation of northern right whales.
In the past 10 years, climate change has also affected whale movement patterns, causing them to foray into unprotected shipping lanes and fishing zones, and today only about 410 North Atlantic right whales remain despite extensive protective measures established in the 1970s.
This recent and dramatic change in the distribution of right whales in the North West Atlantic has precipitated the growing need for knowing where these behemoths are and predicting where they will be. Well-established whale tagging methods are incompatible with right whale anatomy, so they must be tracked by acoustic or visual encounter. However, this data is not enough to provide a yearly prediction model. My mentor Nick Record and his collaborators from Bigelow’s Computational Oceanography Lab have surmounted this obstacle by examining lower levels of the trophic food web. Combined with this understanding of the North West Atlantic’s ecology, we use the Maxent model for machine learning to compare environmental conditions, like chlorophyll-a levels and sea surface temperature, to our observational whale data set, and Maxent learns which environmental conditions the whale observations correlate with. Given a set of conditions, the model outputs a map for the likelihood of whale encounters.
This summer I am working on fine-tuning a model that will be included in a grant proposal later this August. The past few months I have been testing the parameters which control how Maxent takes in the observational data and compares it to the condition layers, for not only right whales but also fin whales and possibly humpback whales, which are all included in our dataset.
Nick’s lab also focuses on the incorporation of citizen science, and his lab is planning to eventually incorporate whale watch and other citizen reported data into a future whale model. My experience as an REU intern cannot be mentioned without noting COVID, though I do have a unique frame of reference having worked in an astrophysics lab in a previous REU along similar technical lines but with a very different context (as you would expect). I enjoy working at Nick’s lab, even if remotely, because it’s impossible to forget the motivation behind this work, and it is rewarding to be in a space where a mentor’s imperative for justice extends beyond the environment.
Turner Johnson is a Haverford College student in Bigelow Laboratory for Ocean Science’s Research Experience for Undergraduates program. This intensive experience provides an immersion in ocean research with an emphasis on state-of-the-art methods and technologies.