AI and Citizen Science Unite to Protect Shark Populations
About one-third of shark species face extinction risks, yet critical data on their habitats and population trends remain scarce. To address this, researchers from Virginia Tech, Stanford University, and other institutions have developed sharkPulse, the largest open database of shark sightings. This platform uses artificial intelligence (AI) to scan photos posted online, automatically extracting data like location, timestamps, and species identification.
Published in Fish and Fisheries on August 11, the research introduces a shift from traditional citizen science, which depends on voluntary submissions, to an automated system that converts everyday online activity into valuable conservation data.
Transforming Data Collection with AI
SharkPulse validates images through a combination of public contributions and expert review before adding them to a searchable database. This method allows scientists to monitor shark populations and distribution patterns quickly and on an unprecedented scale.
"With cameras in nearly everyone’s hands, encounters with the ocean are being recorded more than ever," said Jeremy Jenrette, a Ph.D. candidate involved in the project. "SharkPulse taps into this global stream of images, using AI and data science to monitor shark populations passively."
The platform addresses critical knowledge gaps by identifying habitats and areas sharks frequent, which is essential for informed conservation planning. As Francesco Ferretti, the study’s lead author, stated, "We can’t protect what we don’t know."
Building on Previous Research
Ferretti’s prior work includes the “White Shark Chase,” a project to track critically endangered white sharks in the Sicilian Channel using environmental DNA and baited cameras. This effort detected white sharks across multiple sites and supported plans for a monitoring program to prevent extinction.
During that expedition, researchers tagged a juvenile shortfin mako shark—the first such tagging in the Mediterranean. Tracking revealed the shark traveled over 750 miles in 54 days, highlighting the species’ extensive range and the need for wider conservation strategies.
Virginia Tech also participates in MegaMove, a global initiative tracking more than 100 large marine mammal species to map habitats and migration routes. Findings from MegaMove showed that 60% of vital habitats lie outside protected zones, emphasizing the need for expanded conservation efforts.
Tracking Sharks in Real Time
SharkPulse reduces the need for manual data uploads by automating the collection process, but it still depends on public participation to validate sightings and train AI models. So far, the platform has validated over 91,000 records covering 285 shark species—almost 53% of known species worldwide.
This data has already helped identify new shark hotspots, including white sharks in the Mediterranean. It also supports assessments for the International Union for Conservation of Nature’s (IUCN) Red List by providing dynamic maps of shark distribution and abundance trends.
For example, bull sharks are seasonal visitors to Virginia’s Chesapeake Bay, but their movements and population sizes are not well documented. A 2.6-meter bull shark caught and photographed in 2018 off Cedar Point, St. Mary’s County, represents the kind of local data sharkPulse organizes to strengthen scientific understanding of marine ecosystems.
Collaboration and Expansion
Virginia Tech students and faculty from computer science, wildlife conservation, and data science contribute to sharkPulse. The team is actively seeking grants to expand the platform’s reach and ensure its long-term sustainability.
The researchers plan to scale sharkPulse further by incorporating multilingual data mining and forging international partnerships to close geographic gaps. They are also exploring how to make the data more accessible to policymakers, fishery managers, and conservation organizations.
"This is about creating an always-on pulse monitor for the ocean," said Ferretti. "The more we see, the more we can do to protect."
A Versatile Model for Conservation
SharkPulse serves as a practical, scalable tool for near real-time monitoring of wide-ranging and poorly understood shark populations. It integrates machine learning, big data pipelines, and citizen science into a unified framework.
The team envisions adapting this approach for other species groups, such as sea turtles and bats, to support broader biodiversity monitoring efforts.
The study’s co-authors include experts from institutions such as Stazione Zoologica Anton Dohrn, Stanford University, the Monterey Bay Aquarium Research Institute, and California State University of Monterey Bay. Funding was provided by the Bertarelli Foundation, the Virginia Tech Global Change Center, and the Lenfest Ocean Program.
For those interested in AI applications in conservation and environmental monitoring, this project exemplifies how technology and community engagement can fill critical data gaps. To explore AI training that supports similar initiatives, consider visiting Complete AI Training.
References
- Original study: doi.org/10.1111/faf.70006
- International Union for Conservation of Nature (IUCN): iucnredlist.org
Your membership also unlocks:
