Abstract

Terminal edges of juvenile Ocyurus chrysurus sagittal otoliths were analyzed using laser ablation- inductively coupled plasma-mass spectrometry (LA-ICP-MS) for a suite of elemental ratios (Mg/Ca, Mn/Ca, Sr/Ca, Ba/Ca, Cu/Ca, Pb/Ca, Zn/Ca, Rb/Ca, and Y/Ca) to identify unique chemical ‘fingerprints’ for three, non-reef, nursery habitats in St. Thomas, USVI. The three juvenile habitats include Mangrove Lagoon (classic mangrove habitat), Sprat Bay (dense Thalassia testudinum habitat), and Brewer’s Bay (sparse seagrass and drift algae habitat). Ten to Fifteen fish in each habitat were collected in Sept.-Dec. of 2000 and May-Jun. of 2001 to test for small scale spatial variability and temporal differences. Significant site differences were found in the otolith microchemistry of fish collected from the three nursery habitats in 2001, but not in 2000. In 2001, Mg/Ca, Mn/Ca, Sr/Ca and Y/Ca all showed significant site differences. Significant differences between years were found for Ba/Ca, Pb/Ca, Rb/Ca and Y/Ca. Quadratic Discriminate Function Analysis (QDFA) was only performed on the 2001 data. Because there were no significant differences in elements between sites in 2000, we felt that performing DFA would force a classification of signatures into one of the habitats, resulting in unreliable interpretation of the data. In 2001, Mg, Mn, Sr, Ba, Pb and Y were selected as predictors and the model was 84% correct at classifying the multi-element signature. On an individual site basis, classifications were 80% correct for Brewer’s Bay, 87% for the Mangrove Lagoon and 87% correct for Sprat Bay. The different environments inhabited by O. chrysurus in this study were separated, at most, by 14 km. Although we were unable to determine unique chemical fingerprints in 2000, because of the lack of elemental differences, we were able to distinguish between sites in 2001 at a relatively high percentage. The ability to distinguish between nursery habitats at relatively small spatial scales is vital if otolith micro-chemical techniques are used as a fisheries tool for small, non-estuarine islands such as St. Thomas. Inhibiting the effectiveness of chemical markers, however, is the inability to replicate similar fingerprints over time. Temporal variation in elemental fingerprints may be one reason for the lack of spatial distinction in the 2000 dataset. Another possible driver in elemental variation is local oceanography. Oceanographic processes, such as general current flow, eddy formation and upwelling may influence the availability and presence of elements in the ambient near-shore waters. Analysis along the marginal edge of juvenile O. chrysurus otoliths demonstrated that, although, ratios of element/Ca differed between nursery sites enough to provide each with a unique chemical fingerprint, they were not consistent between 2000 and 2001.The inability to chemically resolve differences between juvenile habitats in 2000 suggests that some sampling times provide more distinct, or stronger, signatures than others. Similarly, periods of relative homogenous chemical water conditions can make otolith fingerprints, at small spatial scales, indistinguishable. Long-term, temporally replicated studies could elucidate annual or seasonal trends in chemical signatures or show correlations as a response to environmental events

PDF Preview