Diving into the data: UAA grad student analyzes sea otter behavior

by Keenan James Britt  |   

A sea otter floating on its back
A sea otter in Homer, èƵ. (Photo by James Evans / èƵ)

“We are fortunate to know a lot about sea otter behavior, but there’s also a lot that we don’t know,” said Ana Velasquez, a master’s candidate in biological sciences at UAA and an èƵ Sea Grant State Fellow for the Central Council of the Tlingit and Haida Indian Tribes of èƵ.

Velasquez defended her master’s thesis, ‘Short-term Effects of Tag Implantation in Sea Otters,’ on Oct. 28. While attaching collars or internal tags to animals is a common method in wildlife biology for gathering data on the location and behaviors of wild animals, Velasquez sought to understand how the act of tagging itself can affect that behavior and health of sea otters.

Otters present a special challenge for tagging efforts. “Tags are placed externally either by attachments like collars or backpacks or adhesives [on other animals],” said Velasquez. “Unfortunately, these methods pose a challenge for sea otters.” Previous studies found that external tags weren’t a viable option for sea otters long-term, and scientists turned to surgically implanted devices to monitor the species. However, scientists have been cautiously aware this procedure likely has some effects on the otters’ health and behavior.

“We know that there is a knowledge gap regarding the short- and long-term impacts of implantation [on otters],” Velasquez explained. “There has been increasing awareness of the necessity of such studies to ensure animal well-being and data integrity for researchers tasked with downstream analysis.”

New methods for old data

a sea otter raft in Kachemak Bay, èƵ
A sea otter raft in Kachemak Bay, èƵ. (Photo by Ana Velasquez. USFWS permit #740507)

While Velasquez has studied marine mammals in the field with the U.S. Geological Survey and the U.S. Fish and Wildlife Service, she made use of an existing data set for her thesis project. Originally collected on a group of sea otters from Prince William Sound in 2003 and 2004, the dataset recorded both the otters’ dive behavior and their body temperature. 

After analyzing the data, Velasquez found that several otters experienced fevers within three days of having tags implanted, likely as a result of “an acute-phase response or inflammatory reaction” from the surgery. She also determined from the data that the otters were diving less following the surgery, “showcasing a change in their ability to acquire food which would have been energetically costly given their high metabolism.”

The scale of the datasets Velasquez worked with required specialized techniques for analysis. “The dive data was collected every two seconds for 19 sea otters for a year or more. Body temperature data was collected every two minutes for every sea otter for a year or more,” said Velasquez. “So when you think about it, that's thousands of data points, so it's really hard to look at with just a spreadsheet.”

In order to make sense of all the information, Velasquez processed the numbers using the R programming language, which is often used for statistics and data analysis. Velasquez credits her advisor, UAA’s Amy Bishop, Ph.D., for introducing her to R and advanced statistical analysis. “There's no way around it for marine biologists to get away from stats and coding. Coding is definitely not easy, but is a straightforward and streamlined and reproducible way to do [statistics]. And R is an open access platform, so it's a super great resource for researchers to use and learn on.”

An ethical and pragmatic approach

While Velasquez’s thesis research involved heavy number crunching, otters are more than numbers on a spreadsheet to her. A concern for the well-being of animals is fundamental to her research interests. 

“I grew up with this narrative of ‘You should protect where you play. You should protect what you love.’ I've always had a fascination with animals, and to some extent, ethics as well,” Velasquez said. 

By making use of existing data sets for her thesis project, Velasquez did not need to conduct further capture studies on otters. “As scientists, we have a role to not do more harm than what is the greater benefit of the scientific process, and to increase our understanding,” she explained. “And while we’re trying to always learn more about the system, we don't want it to be at a detriment. There's not a need to go back and re-influence the system by capturing [animals] again if we already have the data available.” 

In addition to animal ethics, analyzing existing data also has pragmatic benefits. “These kind of capture studies are incredibly expensive. You need a specific dive team that has the rebreather certifications and all the gear for that. You need boat time, you need personnel time, you need the drugs to sedate the otters, you need the tags to get the data. There's just tons of cost associated with this kind of study,” Velasquez explained. “There's something to be said of reusing data when you can to answer questions so that you're not duplicating efforts.”

Having completed her thesis defense, Velasquez is currently focusing her efforts on her role with the Central Council of the Tlingit and Haida Indian Tribes of èƵ. “I was hired on to support mariculture work,” said Velasquez, who aims at “increasing the participation of tribal citizens and tribes within the mariculture sector” while working with the Indigenous Stewardship Program team on holistic ecosystem stewardship.

"Diving into the data: UAA grad student analyzes sea otter behavior" is licensed under a .