Researchers placed a multi-sensor tag on a whitespotted eagle ray and captured where they go, how they feed, how they move through their habitats, and how they interact with other species \u2013 insights that were virtually impossible to capture before. Credit: FAU Harbor Branch and Kyle Newton
\nWhitespotted eagle ray: a mysterious ocean giant<\/p>\n
Whitespotted eagle rays (Aetobatus narinari) are powerful, fast-swimming predators that glide through tropical and subtropical coastal waters. <\/p>\n
Growing more than 2 metres (6.5 feet) wide and weighing several hundred kilograms, they feed primarily on hard-shelled animals such as clams and conch. <\/p>\n
Despite their ecological importance in marine food webs, little is known about how they behave in the wild. Until now, their smooth skin and lack of a prominent dorsal fin have made it nearly impossible to attach tracking equipment securely.<\/p>\n
While biologging has become a vital tool in studying species such as sharks<\/a> and sea turtles<\/a>, rays \u2013 especially pelagic (open ocean) ones \u2013 have been left behind. This lack of data has concerned scientists, especially as many batoid species (which include skates, rays and stingrays<\/a>) are under increasing threat from habitat loss and overfishing.<\/p>\n How the biologging tag works<\/p>\n The custom-built tag integrates several tools into a single lightweight design, including a motion sensor, underwater microphone, video camera, satellite transmitter and acoustic tracker. <\/p>\n A major innovation lies in how the tag is attached: using silicone suction cups and specially designed straps that secure near the ray\u2019s spiracles \u2013 small openings just behind the eyes. This method proved both fast and minimally invasive, say the researchers, allowing the tag to remain in place for up to 60 hours, even in strong ocean currents.<\/p>\n This marks the longest recorded attachment time for an external tag on a pelagic ray, making it a significant achievement in marine tagging technology, say the researchers.<\/p>\n The detailed data collected by the tag revealed not only where the rays travelled, but also how they fed, interacted and navigated their environment.<\/p>\n Researchers used the tag\u2019s video and sound recordings to identify key behaviours, including ‘browsing’, ‘swimming’ and ‘digging’. <\/p>\n The team trained a supervised machine learning model, known as a Random Forest model, to detect foraging behaviours based on sensor data. This model was first calibrated using labelled video footage and then tested on other data to predict behaviour with surprising accuracy. <\/p>\n \u201cFeeding in the field followed a repetitive sequence,\u201d the researchers document in the study, \u201cdescent to the sediment, ‘browsing’ (raising and lowering the rostrum along the sand while moving forward), occasionally ‘digging’ into the sediment, winnowing sand away… Once prey pickup occurred, the ray immediately ascended out of the sediment, usually >1 m, and glided back downward while processing the prey item.\u201d<\/p>\n Notably, the researchers found that some feeding activities could be recognised using only movement and sound \u2013 without the need for video. This means simpler, more efficient tags could potentially be deployed on a wider scale, enabling long-term ecological monitoring of rays and other understudied species, say the team.<\/p>\n \u201cOur work marks a turning point in how we study elusive marine species like pelagic rays,\u201d says co-author Cecilia M. Hampton. \u201cWe\u2019ve shown that complex behaviours \u2013 like the crunching of clams \u2013 can be identified using sound and movement data alone, even without video. This opens up exciting possibilities for long-term ecological monitoring using simpler, more efficient tags.\u201d<\/p>\n Researchers captured the feeding behaviour of a whitespotted eagle ray with video and acoustics. Credit: Cecilia Hampton, FAU Harbor Branch The researchers say the success of the study demonstrates how biologging technology, when combined with machine learning, can revolutionise the way scientists observe marine animals. With some adaptation, the tag could be used on other ray species, helping to fill important gaps in our knowledge of marine ecosystems.<\/p>\n \u201cAs biologging technologies advance, combining data streams like movement, sound and video \u2013 and applying machine learning for behaviour classification \u2013 could turn rays into mobile surveyors of ocean health and benthic habitats,\u201d adds senior author of the study Matt Ajemian.<\/p>\n![\"Cecilia](\"https:\/\/www.newsbeep.com\/uk\/wp-content\/uploads\/2025\/07\/Cecilia-M.-Hampton-places-a-multi-sensor-tag-on-a-whitespotted-eagle-ray.jpeg\")
Researcher Cecilia M. Hampton places a multi-sensor tag on a whitespotted eagle ray. Credit: FAU Harbor Branch
\nWhat the researchers found<\/p>\n
\nWhat’s next?<\/p>\n