Livia Gavard went scuba diving and snorkeling in the Indian Ocean in Tofo, Mozambique. A pod of dolphins swam alongside her the whole time she was there.
Her pod’s leaders were up ahead, studying the actions of the other dolphins. After getting a glimpse of something out of the corner of her eye, Gavard cocked her head to the right. While doing so, she locked gazes with a baby humpback whale that was swimming nearby. Looking under, Gavard saw a second humpback whale that was much bigger than the one swimming on the surface.
Gavard said, “I will never forget when that baby whale looked at me, and I wasn’t sure who was looking at who,” and he wouldn’t forget it either. “I think the whale was more interested in learning about me than I was in learning about him,” the scientist added.
While Gavard has always loved the ocean, it wasn’t until a learning trip to the south of Spain when she was 16 that she developed an interest in whales.
Gavard said, “And at that moment, I knew that being out on the ocean, photographing whales and dolphins, was my true calling in life.” Gavard continued, “And that’s when I realized that.”

Artificial intelligence is being used to distinguish individual humpback whales.
Over the course of the previous six winters, Gavard has commuted from his home in Vermont to his workplace in Tofo, where he analyses images of whale flukes to determine species. Whales’ tail fins feature scars that look like black and white stripes on the underside. We refer to this as “flukes.”
Each individual baleen plate has its own unique shape that is developed over the course of a whale’s lifetime. When Gavard returns from the boat with pictures of the whales’ flukes, he feeds them into an AI system he developed dubbed “flukebook.” When asked to describe how it works, Gavard said, “effectively working like a fingerprint.” When a picture of a whale’s fluke is taken and posted online, it may be easily identified in any other pictures of the whale’s fluke that are uploaded in the future.
It hasn’t always been this easy for scientists to study humpback whales.
Visual matching was the primary method of whale identification until the widespread adoption of flukebooks. In this approach, a researcher would look at all of the flukes and try to pair them together by hand.
It was a lot of work, though, and it makes you go crazy, as Gavard put it. As one user put it, “after looking at too many of them, you start to get a little blurry in the eyes.” Making a plan based on the whale’s tail
Gavard’s master’s project seeks to answer the questions, “Is there a connection between scar presence and sex of the whale?” and “Is there a connection between scar type and location on the fluke?” For this reason, Gavard uses a second AI system to divide the fluke into its constituent parts. After that, she takes careful inventory of the scars on the fluke and where they are placed.
Gavard’s discoveries could also be utilized to identify a humpback whale’s sex. Because they almost entirely lack sexual dimorphism, this is currently difficult to achieve without a tissue sample. When this happens, the males and females of the species don’t look the same. This could be because of variances in size, shape, or coloration.
Gavard claims that “you can tell the difference between a rooster and a hen just by looking at them because they look different.” The scientist emphasized that this phenomenon does not occur in whales. While females tend to grow slightly larger than males, this size difference won’t help you tell them apart on the water.
Individual and social characteristics have helped researchers in the past identify male and female whales. Gavard is anticipating a similar association between the scarring on a whale’s flukes and its sex because previous studies have found such a connection between the number of scars on a whale’s back and its sex. In response to Gavard’s comment that “but it’s like everything in science,” I confirmed his point by saying, “Yes.” The only way I’ll know for sure if I’m entirely off base is if I actually finish the work.
The first task is to find some whales. Gavard’s research technique begins with a trip to sea in search of whales. After breaching the surface for air, whales often dive further into the ocean. When whales come to the surface to breathe, they typically act in this manner. Occasionally during this habit, they will raise their flukes, allowing observers in the right spot to photograph the animals from below.
This strategy requires some level of expertise. Gavard prefers to see certain pods of whales over others because she thinks they are more likely to display their flukes. She’ll decide to follow the pod and observe its dynamics while trying to guess which whales among them are male and which are female.
Gavard’s perspective is different from that of most other scientists who study humpback whales. While most researchers have a marine biology background, Gavard comes from a veterinary care background. For Gavard, this has benefits and drawbacks.
“As veterinarians, we are trained to see things from a different perspective than marine biologists,” Gavard said. I explained that marine biologists often look at the same whale and then place it in the context of the tides and currents. My conclusion could be that it has a scar there or that it acts a certain manner.
Marine biologists may learn new things about the ocean’s ecosystem, such as the role of currents and the maintenance of coral reefs. Veterinarians could learn a lot by studying the physiological differences between marine species.
To obtain a fuller picture of the situation at hand, “it’s been very interesting to combine skill sets with marine biologists,” Gavard said. To paraphrase, “It’s been very interesting.” Veterinarians study animals on a more individual level than population researchers. improved knowledge of disease, anatomy, and the uniqueness of each living being.
After finishing veterinary school, Gavard plans to make Tofo his permanent home and continue his study of humpback whales. She wishes she could merge her veterinary profession and her work with humpback whales into a single career. She plans to study whales in the future using GPS devices. The whales will have satellite tags attached to them so that their whereabouts can be tracked for the duration of their lives.