Using Drones to More Accurately Count Sea Turtles

Caitlin Dempsey


Traditional methods used to count nesting sea turtles has involved manual counts of turtles arriving on beaches to lay eggs and marking shells with non-toxic paint. Turtles are then counted and a population estimate is calculated based on the ratio of painted versus unpainted turtle shells by in situ survey from boats and beaches. This approach to create a year-to-year census of seas is tedious and, as recent research testing a new methodology involving drones suggests, inaccurate.

Each year, the world’s largest population of green turtles nests at Raine Island, located on the northern Great Barrier Reef in Australia. Creating an accurate annual count of nesting green turtle populations is important for understanding the reproductive success and long-term population changes of this vulnerable species.  Nesting beach inundation, climate change, plastic pollution, vessel strikes, coastal development, and commercial fishing are some of the cumulative impacts negatively affecting the health of the green turtle population.

Researchers tested out the accuracy and efficiency of using unmanned aerial vehicles (UAVs) and underwater cameras as compared to historical methods of counting green turtles. In a recently published PLOS ONE journal, the researchers found that compared to traditional counting methods, using UAVs to count sea turtles was a more efficient method time wise, covered a larger swath of area for counting, and could be used in more extreme weather conditions while still providing precise estimates. Researchers found that using UAV and underwater cameras yielded a lower ratio of painted to unpainted turtle counts. This translated into a  1.73x higher population estimate using UAVs and a 1.53x higher estimate using underwater cameras, suggesting that manual counts have been underreporting female green turtle populations.

Screenshot of drone survey at 50m altitude over waters adjacent to Raine Island reef edge showing a painted turtle shell and unpainted shells.  Dunstan et al., 2020.
Screenshot of drone survey at 50m altitude over waters adjacent to Raine Island reef edge showing a painted turtle shell and unpainted shells. Dunstan et al., 2020.

The use of drones and underwater cameras also provides the added benefit of archival video footage that can be replayed to recheck turtle counts at a speed more conducive to manual counting. Researchers haven’t tested for inherent bias in manual counting of painted turtles versus non-painted turtles to better understand the role that plays in factoring out turtle population estimates. Difficulty in counting when high numbers of turtles are present and observer fatigue were also posited as factors that impact the accuracy of traditional survey methods.

In the future, researchers hope to incorporate the use of artificial intelligence to count turtles from drone and underwater camera footage.

The study

Dunstan, A., Robertson, K., Fitzpatrick, R., Pickford, J., & Meager, J. (2020). Use of unmanned aerial vehicles (UAVs) for mark-resight nesting population estimation of adult female green sea turtles at Raine Island. Plos one15(6), e0228524.

Counting turtles is a science. (2020, June 9). Great Barrier Reef Foundation.


Photo of author
About the author
Caitlin Dempsey
Caitlin Dempsey is the editor of Geography Realm and holds a master's degree in Geography from UCLA as well as a Master of Library and Information Science (MLIS) from SJSU.

Free weekly newsletter

Fill out your e-mail address to receive our newsletter!