A new approach to better assessing whale population data has emerged, led by a research team of marine biologists from Scripps Institution of Oceanography at UC San Diego and statisticians from Cal Poly.
Scientists typically monitor whale presence through a variety of traditional methods such as visual surveys, photo identification, acoustic monitoring, satellite imagery, and, increasingly, genomic methods. But monitoring can be challenging due to a wide-ranging migration area and intermittent surface pop-ups, among other difficulties.
Using an innovative alternative method, the Cal Poly-Scripps team examined microbial "ecological habitats" as highly accurate predictors of how many filter-feeding whales were occupying the California coast between 2014 and 2020 from San Diego to Morro Bay.
Baleen whales, including blue, fin and humpbacks, commonly migrate along the California coastline and feed on dense collections of food sources. Baleen mouth filters function as a sieve after the whale takes a massive mouthful of water. The massive creatures use their tongues to expel the water and trap prey such as krill, zooplankton and other small fish inside the mouth for the whale to eat.
In their analysis examining the ecological habitat, or community of small organisms associated with baleen whales, researchers from the two institutions collaborated to develop data models to predict whale presence from microbial and plankton information. Their study appeared in a May 6, 2026, article published in the journal PLOS One.
The research used data from the world's oldest marine ecosystem monitoring program, the California Cooperative Oceanic Fisheries Investigations (CalCOFI) , now in its 77th year, to examine the relationship between baleen whales and their ecological habitat.
"The concept of this project was to try and find an indirect signal that's based on ecological relationships in the microbial communities that are in the water and how they respond to macro-organisms, such as whales," said Trevor Ruiz, a Cal Poly assistant professor of statistics, one of the study's lead authors. "We developed a tailored approach, and while our methods are not off the shelf, they remain transferrable to prediction from microbial data more broadly, and along with our scientific findings we have provided a portable software implementation of the methods to lower barriers to adoption for other researchers who might be interested in applying our approach to other problems."
Working in coordination with Scripps marine ecologists, who informed the biological aspects of the research, the Cal Poly team developed the statistical modeling using computational data science tools.
"This project gave me valuable experience in data analysis and modeling," said Nick Patrick, a 2025 Cal Poly statistics alumnus and study coauthor. "It had many collaborators across different fields, from statistics to marine ecology to genomics, and seeing everyone from these disciplines coming together to form the final product was one of the most interesting aspects."
Erin Satterthwaite, a marine ecologist affiliated with Scripps Oceanography, said the scientists compared the whale survey data taken from visual observations with seawater samples analyzed for microbial and small plankton communities using environmental DNA, or eDNA. In this approach, seawater is filtered and DNA from organisms in the water is extracted. Specific genetic markers are then amplified and sequenced, allowing researchers to identify the suite of organisms present by matching sequences to reference databases.
"Many approaches rely on indirect environmental proxies that are several steps removed from the actual biology of the whales," Satterthwaite said. "Our work uses eDNA to characterize the structure of the microbial and small plankton community which adds to existing oceanographic information by incorporating ecological habitat information of whales, which improves our ability to predict whale densities."
Findings based on the study suggest that on average, predictions of whale densities based on microbial communities were found to be 53% more accurate than traditional forecasts.
The work adds to a growing body of research using eDNA that gives scientists a window into the biology and ecology of the ocean. With whales, scientists are interested in monitoring their populations as a general indicator of ocean health, to better understand and manage human impacts such as vessel strikes, entanglement, and noise disturbance, and because they are culturally and ecologically important. Satterthwaite said that this work can help scientists to better understand how whales are connected to tiny ocean life, like bacteria, phytoplankton, and zooplankton.
As the cost decreases and the ease of using eDNA techniques improves, eDNA is becoming increasingly accessible for a wide range of applied purposes. More broadly, this approach could also be used to study other large marine animals, like sharks or large open-ocean fish, which could help create more detailed range maps of marine species.
