Although laboratory studies confirm that otoliths incorporate trace elements and stable isotopes from surrounding waters, few studies explore the relationship of otolith chemistry to water chemistry in the field and none include a larger suite of environmental tracers, such as rare earth elements. Using spotted seatrout (Cynoscion nebulosus) as model species, we tested the hypothesis that otoliths record the water chemistry of seagrass habitats in Chesapeake Bay. In summer 2001, we sampled water and juvenile fish in seagrass beds of the bay. Weighted linear regressions showed that [Ba/Ca]otolith and [La/Ca]otolith were best predicted by salinity and were modeled as [Ba/Ca]otolith (µmol·mol–1) = –2.25 ± 0.35 × salinity + 59.47 ± 7.01) and [La/Ca]otolith (pmol·mol–1) = –8.71 ± 0.65 × salinity + 243.87 ± 12.52. [Ba/Ca]otolith increased with [Ba/Ca]water, but the relationship was nonlinear. Salinity did not influence [Mn/Ca]otolith, but this ratio was positively correlated with [Mn/Ca]water. Although the partition coefficient of Sr (DSr = 0.23 ± 0.019) was similar to that in laboratory experiments, [Sr/Ca] in waters and otoliths was decoupled despite equal temperature exposure, suggesting that [Sr/Ca]otolith concentration may not be a simple function of water composition. However, there was a predictive relationship between [δ18O]otolith and [Sr/Ca]water ([δ18O]otolith = 1.18 ± 0.09 × [Sr/Ca]water (mmol·mol–1) – 14.286 ± 0.78) resulting from mixing between fluvial and oceanic waters. Water chemistry showed mixed values as a proxy for otolith chemistry and may not be a surrogate for otolith chemistry in wide estuaries.

We investigated the variability of otolith chemistry in juvenile spotted seatrout from Chesapeake Bay seagrass habitats in 1998 and 2001, to assess whether otolith elemental and isotopic composition could be used to identify the most essential seagrass habitats for those juvenile fish. Otolith chemistry (Ca, Mn, Sr, Ba, and La; δ 13 C, δ 18 O) of juvenile fish collected in the five major seagrass habitats (Potomac, Rappahannock, York, Island, and Pocomoke Sound) showed significant variability within and between years. Although the ability of trace elements to allocate individual fish may vary between years, in combination with stable isotopes, they achieve high classification accuracy averaging 80-82% in the Pocomoke Sound and the Island, and 95-100% in the York and the Potomac habitats. The trace elements (Mn, Ba, and La) provided the best discrimination in 2001, a year of lower freshwater discharge than 1998. This is the first application of a rare earth element measured in otoliths (La) to discriminate habitats, and identify seagrass habitats for juvenile spotted seatrout at spatial scales of 15 km. Such fine spatial scale discrimination of habitats has not been previously achieved in estuaries and will distinguish fish born in individual seagrass beds in the Bay.Extra keywords: estuarine-dependent fish, natural tag, otolith microchemistry, seagrass bed discrimination.

No abstract