Fine-scale foraging ecology and habitat use of sympatric green and hawksbill turtles in the Western Indian ocean.

Using stable isotope analysis of carbon and nitrogen of turtle tissues and putative prey items, we investigated the diet of immature green turtles and hawksbill turtles foraging in the lagoon of Aldabra Atoll, a relatively undisturbed atoll in the southern Seychelles. Aldabra offers a unique environment for understanding sea turtle ecology. Green turtles mostly consumed seagrass and brown algae while hawksbill turtles mainly consumed mangroves and invertebrates. Green turtles showed a dietary shift with size (a proxy for age). There was minimal niche overlap between species and evidence of small-scale foraging site fidelity with turtle tissue reflecting site-specific prey. This highlights the ecological importance of seagrass and mangrove habitats and suggests that turtles play a role in controlling algal biomass at Aldabra. This study is the first to closely examine the foraging ecology of these sympatric turtle species in the Western Indian Ocean, a globally important region for both species.

Chromosome-level genome assembly for the Aldabra giant tortoise enables insights into the genetic health of a threatened population.

BACKGROUND: The Aldabra giant tortoise (Aldabrachelys gigantea) is one of only two giant tortoise species left in the world. The species is endemic to Aldabra Atoll in Seychelles and is listed as Vulnerable on the International Union for Conservation of Nature Red List (v2.3) due to its limited distribution and threats posed by climate change. Genomic resources for A. gigantea are lacking, hampering conservation efforts for both wild and ex situpopulations. A high-quality genome would also open avenues to investigate the genetic basis of the species' exceptionally long life span. FINDINGS: We produced the first chromosome-level de novo genome assembly of A. gigantea using PacBio High-Fidelity sequencing and high-throughput chromosome conformation capture. We produced a 2.37-Gbp assembly with a scaffold N50 of 148.6 Mbp and a resolution into 26 chromosomes. RNA sequencing-assisted gene model prediction identified 23,953 protein-coding genes and 1.1 Gbp of repetitive sequences. Synteny analyses among turtle genomes revealed high levels of chromosomal collinearity even among distantly related taxa. To assess the utility of the high-quality assembly for species conservation, we performed a low-coverage resequencing of 30 individuals from wild populations and two zoo individuals. Our genome-wide population structure analyses detected genetic population structure in the wild and identified the most likely origin of the zoo-housed individuals. We further identified putatively deleterious mutations to be monitored. CONCLUSIONS: We establish a high-quality chromosome-level reference genome for A. gigantea and one of the most complete turtle genomes available. We show that low-coverage whole-genome resequencing, for which alignment to the reference genome is a necessity, is a powerful tool to assess the population structure of the wild population and reveal the geographic origins of ex situ individuals relevant for genetic diversity management and rewilding efforts.