Marine megafauna are difficult to observe and count because many species travel widely and spend large amounts of time submerged. As such, management programmes seeking to conserve these species are often hampered by limited information about population levels. Unoccupied aircraft systems (UAS, aka drones) provide a potentially useful technique for assessing marine animal populations, but a central challenge lies in analysing the vast amounts of data generated in the images or video acquired during each flight. Neural networks are emerging as a powerful tool for automating object detection across data domains and can be applied to UAS imagery to generate new population‐level insights. To explore the utility of these emerging technologies in a challenging field setting, we used neural networks to enumerate olive ridley turtles Lepidochelys olivacea in drone images acquired during a mass‐nesting event on the coast of Ostional, Costa Rica. Results revealed substantial promise for this approach; specifically, our model detected 8% more turtles than manual counts while effectively reducing the manual validation burden from 2,971,554 to 44,822 image windows. Our detection pipeline was trained on a relatively small set of turtle examples (N = 944), implying that this method can be easily bootstrapped for other applications, and is practical with real‐world UAS datasets. Our findings highlight the feasibility of combining UAS and neural networks to estimate population levels of diverse marine animals and suggest that the automation inherent in these techniques will soon permit monitoring over spatial and temporal scales that would previously have been impractical.

Several studies have suggested that significant embryo mortality is caused by microbes, while high microbial loads are generated by the decomposition of eggs broken by later nesting turtles. This occurs commonly when nesting density is high, especially during mass nesting events (arribadas). However, no previous research has directly quantified microbial abundance and the associated effects on sea turtle hatching success at a nesting beach. The aim of this study was to test the hypothesis that the microbial abundance in olive ridley sea turtle nest sand affects the hatching success at Ostional, Costa Rica. We applied experimental treatments to alter the microbial abundance within the sand into which nests were relocated. We monitored temperature, oxygen, and organic matter content throughout the incubation period and quantified the microbial abundance within the nest sand using a quantitative polymerase chain reaction (qPCR) molecular analysis. The most successful treatment in increasing hatching success was the removal and replacement of nest sand. We found a negative correlation between hatching success and fungal abundance (fungal 18S rRNA gene copies g-1 nest sand). Of secondary importance in determining hatching success was the abundance of bacteria (bacterial 16S rRNA gene copies g-1 g-1 nest sand). Our data are consistent with the hypothesis that high microbial activity is responsible for the lower hatching success observed at Ostional beach. Furthermore, the underlying mechanism appears to be the deprivation of oxygen and exposure to higher temperatures resulting from microbial decomposition in the nest.

Although sea turtles face significant pressure from human activities, some populations are recovering due to conservation programs, bans on the trade of turtle products, and reductions in bycatch. While these trends are encouraging, the status of many populations remains unknown and scientific monitoring is needed to inform conservation and management decisions. To address these gaps, this study presents methods for using unmanned aerial systems (UAS) to conduct population assessments. Using a fixed-wing UAS and a modified strip-transect method, we conducted aerial surveys along a three-kilometer track line at Ostional, Costa Rica during a mass-nesting event of olive ridley turtles (Lepidochelys olivacea). We visually assessed images collected during six transects for sea turtle presence, resulting in 682 certain detections. A cumulative total of 1091 certain and probable turtles were detected in the collected imagery. Using these data, we calculate estimates of sea turtle density (km−2) in nearshore waters. After adjusting for both availability and perception biases, we developed a low-end estimate of 1299 ± 458 and a high-end estimate of 2086 ± 803 turtles km−2. This pilot study illustrates how UAS can be used to conduct robust, safe, and cost-effective population assessments of sea turtle populations in coastal marine ecosystems.

Sea turtle hatching success at mass nesting beaches is typically lower than at solitary nesting beaches, presumably due in part to high rates of microbial metabolism resulting from the large input of organic matter from turtle eggs. Therefore, we tested the hypothesis that hatching success varies across areas of the beach in conjunction with differences in the physical nest environment and microbial abundance of in situ olive ridley sea turtle nests at Ostional, Costa Rica. We marked natural nests in high-density, low-density, and tidal-wash nesting areas of the beach and monitored clutch pO2and temperature throughout the incubation period. We quantified hatching success and collected samples of nest sand during nest excavations. We quantified microbial abundance (bacteria and fungi) with a quantitative polymerase chain reaction (qPCR) analysis. Hatching success was lower in nests with lower pO2, higher temperatures, higher organic matter content, and higher microbial abundance. Our results suggest that the lower oxygen within the nest environment is likely a result of the high microbial abundance and rates of decomposition in the nest sand and that these factors, along with increased temperature of clutches in the high-density nesting area, are collectively responsible for the low hatching success at Ostional.

No abstract