36 eco JANUARY/FEBRUARY 2017

FEA

TU

RE S

TO

RY

Monitoring NestingSea Turtles using aThermal Camera

37eco JANUARY/FEBRUARY 2017

By: Rebecca Snyder, BioScience Group, Seiche Limited

Every year, female sea turtles come ashore to nest on tropical and subtropical beach-es around the world. The nesting period is an important time for sea turtle researchersbecause it represents one of the only opportunities to obtain the counts used to esti-mate sea turtle populations, many of which are endangered or threatened. Existingmonitoring methods include nightly patrols of nesting beaches, however these patrolscan be difficult to complete in remote locations and may require several individuals ifnesting takes place along long stretches of beach. Additionally, beach patrols havethe potential to unintentionally disturb turtles, causing them to return to sea withoutnesting, producing a false crawl, abandoning the nest, and, in some cases, leading tothe release of eggs while at sea rendering them unviable.

An alternative to beach patrols may come in the form of enhanced monitoring tech-nologies such as thermal cameras.

Sea turtles are ectothermic reptiles (their body temperature depends on the externalenvironment), and the use of thermal imaging may seem counterintuitive. However,studies have demonstrated that sea turtles have some capacity to retain heat and,therefore, maintain body temperatures a few degrees above that of the surroundingenvironment (Standora et al., 1982; Sato, 2014). Leatherback sea turtles may havebody-environment temperature gradients as high as 18°C (Bostrom and Jones, 2007).Body temperatures are also elevated when female sea turtles emerge from the seaand journey up the beach to nest (Sapsford and Hughes, 1978; Sato, 2014).

A loggerhead sea turtle emerges from the ocean and onto the beach before nesting. The image was captured during a July 2016 field studyby Seiche Limited that tested thermal cameras for this application.

Unlike low-visibility cameras and night-vision devices, which operate in the near infrared (NIR) portion of the spec-trum, long wavelength infrared (LWIR), also known as thermal infrared) devices do not require light and can operate incomplete darkness. Thermal cameras (cameras operating in the LWIR) produce images based on temperature differ-ences between objects.Therefore, thermal cameraspositioned on nesting beachesshould be able to detect sea tur-tles by sensing the small tem-perature differences in theirbody temperature compared tothat of the surrounding envi-ronment, whether that be air,water, or sand.

Low-light technologies havebeen used to monitor sea tur-tles and hatchlings; however,little research has been con-ducted on the use of thermalcameras. To evaluate the effi-cacy of thermal camera tech-nology to detect and monitorsea turtles, Seiche Limited con-ducted a field trial during the2016 sea turtle nesting seasonin Florida under the permissionof the Brevard CountyEnvironmentally EndangeredLands (EEL) program.

MethodsThe field trial was conducted at the Barrier Island Sanctuary in Melbourne Beach, Florida. The Barrier IslandSanctuary is located near the center of the 20.5-mi Archie Carr National Wildlife Refuge, an important nestingsite for loggerhead, green, and leatherback sea turtles. Monitoring was conducted between the hours of 2030and 0200 (EDT) from 5 to 20 July 2016, with the exception of the evening of 11 July (and early hours of 12July) due to weather.

38 eco JANUARY/FEBRUARY 2017

FEA

TU

RE S

TO

RY

Location of the Barrier Island Sanctuary(red pin) on the Archie Carr WildlifeRefuge (highlighted by the red line on theinset of the Florida coast).

Electromagnetic spectrum.

39eco JANUARY/FEBRUARY 2017

The camera system, which consisted of a thermal cam-era (uncooled, 640 x 480 resolution), tripod, batterypack, laptop, and associated cables, was installed on theobservation deck of the Barrier Island Sanctuary. Theobservation deck sat just above the dunes, providing aclear view of beach for approximately 1 mile to thenorth and south of the sanctuary.

The thermal camera was positioned to view down thebeach at the start of each monitoring session and wasrepositioned as needed when sea turtles were observedon the beach. Real-time monitoring of the video feedwas accomplished using a laptop computer. The bright-ness of the laptop display was reduced to its lowest set-ting, the display covered with a red cellophane sheet,and the laptop placed inside a hood to ensure light emis-sions were minimised (high levels of light can preventnesting by females and disorient hatch-lings). The camera software was pro-grammed to record and save video filescontinuously while the camera was inoperation.

ResultsApproximately 51 hours of thermalcamera data were collected over thestudy period. Sea turtles were record-ed each day, with a total of 74 turtlesobserved. Turtles in the surf, emerg-ing from the sea, excavating the nest(both body cavity and egg chamber),covering and camouflaging the nest,returning to the sea, and in the waterwere all captured on the thermal cam-era. Hatchlings crawling down thebeach for the ocean were observed onone occasion.

The turtle crawl or track produced asthe turtle completed the journey fromocean to nesting site and back againwas clearly visible on the thermalcamera. When viewed at close range,a high level of detail was observed forthe crawls.

DiscussionThe results of the field study are promis-ing and provide valuable insight onpotential applications of thermal imag-ing for the monitoring of sea turtles.

The instantaneous spatial coverage of asingle thermal camera is greater thanwhat a turtle observer would see withthe naked eye during a beach patrol atnight. Therefore, strategic positioningof multiple cameras at a nesting sitewould significantly reduce the fieldeffort required to record nightly num-

bers of nesting turtles over the same area when com-pared to nightly beach patrols. In addition, fewer indi-viduals on the beach conducting nightly patrols wouldlikely result in less disturbance to nesting turtles. Whenused in combination with beach patrols, thermal cam-eras could provide details on the level of disturbancenightly patrols (those conducted on foot and/or with all-terrain vehicles) may have on the behaviors of nestingturtles, which could then be used to inform future sur-vey protocols to minimize impacts.

The camera system used for the field study is representa-tive of a temporary set up. A semi-permanent installationwould incorporate a high capacity computer to run thecamera software and store data files, as well as a teleme-try link that would enable data transmission to a centrallocation for real-time monitoring and system checks.

Thermal camera with laptop (in thinkTANK hood) set up on the Barrier IslandSanctuary’s observation deck.

Auto-detection algorithms couldfacilitate detection and reduce thetime needed to produce counts. Thelevel of detail displayed from thecrawls is quite high and may be suf-ficient to support species identifica-tion. Crawls are unique to speciesand are often used to identify andconnect nests to species when theturtle itself is not observed.Additional algorithms could be usedto identify turtles to species basedon the turtle crawls.

The use of thermal cameras may beparticularly beneficial for monitor-ing nesting turtles on remote anddifficult to access nesting siteswhere daily beach surveys are notpossible. These sites are often inareas where data for population esti-mates are deficient and the informa-tion obtained could inform conserva-tion and management plans as wellas provide insight on population sta-tus. Only occasional site visitswould be required for system mainte-nance, such as battery replacementand hard drive exchange; however,these visits may be further mini-mized by using a renewable powersource such as solar.

Thermal imaging cameras on nestingbeaches would also provide greaterdetails on the predation of eggs andhatchlings, poaching activities, andinteractions between casual beachwalkers and turtles. If the camerasare monitored remotely in real-time,environmental enforcement officerscould be promptly alerted of poach-ing or other instances of turtleharassment. The cameras could alsoserve as a security measure forresearchers and enforcement officerspatrolling beaches with high levelsof poaching.

In addition to the shore-based applica-tions, thermal cameras may be used asa night-time detection tool for sea tur-tles on shipboard surveys. Currently,offshore surveys for sea turtles arelimited to visual observations duringdaylight hours. Mitigation and monitoring programs for industrial activities (dredging, coastal construction, and activ-ities emitting sound into the ocean) often include measures for sea turtles. The inclusion of thermal cameras maymake mitigation programs more efficient for sea turtle detection. Detection of sea turtles at sea should be further eval-uated, however, due to the low number of turtle detections in the water during this study.

40 eco JANUARY/FEBRUARY 2017

FEA

TU

RE S

TO

RY

A turtle pops its head out of the water as it swims away from shore.

Loggerhead turtle crawl (red arrow) showing the alternating flipper pattern character-istic of the species.

41eco JANUARY/FEBRUARY 2017

Shipboard surveys for marine mam-mals often incorporate passiveacoustic monitoring for the detectionof marine mammal vocalizations,however there is potential for missedacoustic observations during periodsof reduced visibility due to variabilityof call rates between species andpotential masking of vocalizations byvessel noise. Incorporating thermalimaging may improve the detectionmarine mammals as well as sea turtles.Seiche has conducted several at-seatrials for marine mammal detectionand the results have been positive.

A key advantage of the use of thermalcameras for monitoring nesting seaturtles is a low level of disturbance. Inaddition to light emissions, sea turtleshave demonstrated sensitivities to thepresence of people while emergingfrom the ocean and excavating thenest, both of which often lead to nestabandonment. Thermal imaging isone technology among several innova-tive methods for studying marinewildlife and, especially for turtles,may prove to be a less intrusivemethod and enable the monitoring ofsea turtles in locations previously toodifficult to access—a positive solutionfor understanding sea turtle popula-tions across the globe.

AcknowledgementsEmma Harrison, Raymond Mojica, Jr.(Environmentally Endangered LandsProgram), and the Barrier IslandSanctuary for organizing access to thestudy site. A special thank you toChristina Mehle and LynnHenneberger who provided valuableassistance in the field.

Literature CitedBostrom, B.L. and D.R. Jones. 2007. Exercisewarms adult leatherback turtles. ComparativeBiochemistry and Physiology, Part A(147):323-331.

Sapsford, C.W. and G.R. Hughes. 1978. Bodytemperature of the loggerhead sea turtle(Caretta caretta) and the leatherback sea turtle(Dermochelys coriacea) during nesting.Zoologica Africana 13(1):63-69.

Sato, K. 2014. Body temperature stability achieved by the large body mass of sea turtles. The Journal of Experimental Biology 217:3607-3614.

Standora, E.A., J.R. Spotila, and R.E. Foley. 1982. Regional endothermy in the sea turtle, Chelonia mydas. Journal of Thermal Biology 7:159-165.

Three turtles are visible in the image: one emerging in the foreground, one returning tothe sea near the two people, and one emerging to the left of the people.

Pair of racoons patrolling the beach.