Résumé: In French Polynesia, both currently recognized manta ray species, Mobula alfredi and M. birostris, are observed. Despite being an important cultural asset and generating significant economic benefits through manta ray watching tourism, published data on the ecology and threats to these species in the region are scarce. Based on an 18-year dataset of sighting records collected by citizen scientists and during two scientific expeditions, this study provides the first insights into the population characteristics and regional distribution of the two manta ray species in French Polynesia. A total of 1347 manta ray photographs (1337 for M. alfredi and 10 for M. birostris) were examined for the period January 2001-December 2017, with photo-identification techniques leading to the successful identification of 317 individual M. alfredi and 10 individual M. birostris throughout the Society, Tuamotu and Marquesas Islands. We provide the first confirmation of sympatric distribution of both species in the Society Islands. Our results highlight strong and long-term site fidelity of M. alfredi individuals to certain aggregation sites (> 9 years for 16 individuals) and reveal some degree of connectivity between populations, with 10 individuals recorded moving between islands located up to 50 km apart. Analysis of photographs of individuals bearing sub-lethal injuries (n = 68) suggests that M. alfredi are more likely to be injured at inhabited islands (Maupiti or Bora Bora; 75% of all injured individuals) than at uninhabited islands, with 75% of injuries related to boat propeller strikes and fishing gear entanglements. Our findings emphasize the need for further research to allow for a comprehensive evaluation of population structure, size and threats to manta rays in this region.

Résumé: The oxygen isotopic ratio of fish otoliths is increasingly used as a 'natural tag' to assess provenance in migratory species, with the assumption that variations in delta O-18 values closely reflect individual ambient experience of temperature and/or salinity. We employed archival tag data and otoliths collected from a shelf-scale study of the spatial dynamics of North Sea plaice Pleuronectes platessa L., to examine the limits of otolith delta O-18-based geolocation of fish during their annual migrations. Detailed intra-annual otolith delta O-18 measurements for 1997-1999 from individuals of 3 distinct sub-stocks with different spawning locations were compared with delta O-18 values predicted at the monthly, seasonal and annual scales, using predicted sub-stock specific temperatures and salinities over the same years. Spatio-temporal variation in expected delta O-18 values (-0.23 to 2.94%) mainly reflected variation in temperature, and among-zone discrimination potential using otolith delta O-18 varied greatly by temporal scale and by time of year. Measured otolith delta O-18 values (-0.71 to 3.09%) largely mirrored seasonally predicted values, but occasionally fell outside expected delta O-18 ranges. Where mismatches were observed, differences among sub-stocks were consistently greater than predicted, suggesting that in plaice, differential sub-stock growth rates and physiological effects during oxygen fractionation enhance geolocation potential using otolith delta O-18. Comparing intra-annual delta O-18 values over several consecutive years for individuals with contrasted migratory patterns corroborated a high degree of feeding and spawning site fidelity irrespective of the sub-stock. Informed interpretation of otolith delta O-18 values can therefore provide relatively detailed fisheries-relevant data not readily obtained by conventional means.