Coastal Management Conservation

Taiwan creates “Facebook” for sea turtles!

Paper:

Su, C.M., Huang, C.T., Cheng, I.J., (2015). Applying a fast, effective and reliable photographic identification system for green turtles in the waters near Luichiu Island, Taiwan. Journal of Experimental Marine Biology and Ecology. 467: 115-120. doi: 10.1016/j.jembe.2015.03.003

Mark the green sea turtle's new profile pic! (Credit: Roy Niswanger)
Figure 1: Mark the green sea turtle’s new super cute–scute for short?–profile pic! (Credit: Roy Niswanger)

Background:

In order to protect a species, the first order of business is to find out how many there are to begin with, and where do they spend most of their time? The most common technique for identifying individual sea turtles is attaching flipper tags—a mark and recapture technique. While certainly a useful method, there are some challenges: applying the tag is tricky in some environments, the tag might weaken and fall off, it is somewhat invasive, cost of effort can be relatively high and handling the animal when tagging and recapturing can cause stress. An alternative technique is photo identification (ID). By using certain patterns, sizes, shapes and colours of physical characteristics—in this case, facial scales—a sea turtle can be recognized as an individual. The greatest advantage of using photo ID is that researchers do not come into direct contact with these endangered animals, significantly reducing long term stress and injury. In Taiwan, there are several species of sea turtles including: green, loggerhead, hawksbill, olive ridley and leatherback sea turtles.  Green sea turtle are the only species that nest in Taiwan. While capture and tagging sea turtles is legal with a license, researchers often run into a huge barrier – millions of tourists. With a complete overlap of research and recreational sites, researchers Su, Huang and Cheng wanted to establish a photo ID database for green sea turtles as a conservation tool.

 

The study:

Figure 3: Map of study sections: I: Beauty Cave to Vase stone, II: Jun-Au Beach to Taipower Company, III: Hai Mouth to Daliao Fishing Port, IV: Shan-fu Fishing Port to Clam Bay.
Figure 2: Map of study sections: I: Beauty Cave to Vase stone, II: Jun-Au Beach to Taipower Company, III: Hai Mouth to Daliao Fishing Port, IV: Shan-fu Fishing Port to Clam Bay. From Su et al., 2015

Liuchiu Island of Pentung County was chosen as the study site since it is the only area where green sea turtles both nest and forage. Then, the extremely terrible task of photographing green sea turtles by diving and snorkeling was completed from June 30, 2011 to July 19, 2013. In case you didn’t catch that, yes, that was sarcasm. Researchers had to moonlight as paparazzi as they attempted to photograph the turtles from both left and right sides of their faces, and figure out if they are male or female.   While the goal was to photograph the same turtle consecutively from both sides without losing sight of it, this was not always possible. Liuchiu island was split up into four study sections (Figure 2). On a survey day, all sites were reviewed for sea conditions and the most favourable site was chosen. On days when all sites were favourable, the site with the least number of surveys was chosen. In the lab, photographs were examined for the number, size, shape and pattern of facial scales in different areas: behind the eyes, at the temples, on the cheek, and unusual arrangements were noted. Many photos were too blurry and could not be used for ID, leaving some turtles unidentifiable.

Results:

Figure 4: A&B show photos of the same turtle with high similarity on either facial side, C&D show the same turtle with low similarity on either facial side
Figure 3: A&B show photos of the same turtle with high similarity on either facial side, C&D show the same turtle with low similarity on either facial side. From Su et al., 2015

After all the photos were analyzed, 35 turtles had paired left and right side photos, 61 turtles had only their left side photographed, and 46 had only their right side photographed. By comparing characteristics on both sides of the faces, researchers found that 80% had low similarity, meaning population size estimates require photographing both sides of the turtle’s face (see Figure 3). Forty-three turtles were photographed more than once, while 77 turtles were photographed very frequently throughout the study suggesting that these were residents. Six percent of all turtles were male, and the rest were female or undetermined. Lastly, most turtles were found in Section I (see Figure 4).

 

Significance:

Figure 5: Percentage of turtles spotted at four surveyed sections.  Black: both faces, light grey: left face, dark grey: right face.  From Su et al., 2015.
Figure 4: Percentage of turtles spotted at four surveyed sections. Black: both faces, light grey: left face, dark grey: right face. From Su et al., 2015.

Okay, so maybe it’s not exactly like Facebook for turtles, but it is essentially a database, the first animal photo-ID database in Taiwan! By identifying individual sea turtles, researchers are able to get a sense of the number of turtles that visit or live in Taiwan, and where they like to spend their time—a great resource for marine conservation and management efforts. Due to its non-invasive nature and high volume of tourists in the area, this is the best option for population estimates, especially considering that the green sea turtle is endangered. There may even be an opportunity for vacationing divers to contribute their turtle pictures, resulting in data that can be used in conjunction with international databases to compare habitats across the globe, track migration patterns and identify high priority foraging sites, nesting sites and underwater highways for protection for this endangered species.

Do you know of any photo ID databases for animals? Bonus points if you know of a database where the public can contribute their photos. Let us know in the comments below!

 

2 thoughts on “Taiwan creates “Facebook” for sea turtles!

Leave a Reply

Your email address will not be published.