Geochemical markers in the otoliths of chinook salmon in the Sacramento-San Joaquin River system, California.

摘要

The purpose of this research is to develop geochemical tracer in otoliths for the reconstruction of fish life history in freshwater systems. In Chapter l, otolith 87Sr/86Sr for juvenile salmon is shown to reflect the 87Sr/86Sr signal for the location where the fish spent the majority of its life. However, for hatchery fish, there is a significant systematic offset in otolith 87Sr/ 86Sr relative to water 87Sr/86Sr that is consistent with approximately 30 percent of otolith Sr coming from the hatchery feed, which is composed primarily of marine fish meal (87 Sr/86Sr_feed = 87Sr/ 86Sr_marine = 0.7092 at all sites). Therefore, it is necessary to distinguish between otolith material deposited while in captivity, as oppose to in the wild, to interpret the otolith 87Sr/86Sr signal.; Chapter 2 presents a new ion microprobe method for reconstructing the dietary sulfur isotope history of a fish using the otolith. This method can distinguish between otolith material deposited while in captivity, as oppose to in the wild, based on the large difference in sulfur isotopic composition between hatchery and freshwater diets. This method allows otolith 87Sr/86Sr to be correctly interpreted. The otolith sulfur isotope chronology that this technique generates also has broader applications to understanding fish ecology and movement.; Chapter 3 presents a new ion microprobe method for making spatially-resolved, high-precision 87Sr/86Sr analyses in otoliths and other carbonates with moderate Sr concentrations. Such a method is necessary to deconvolve the time-averaged otolith 87Sr/86Sr signal for fish that have been away from their natal location for an unknown period of time. The SHRIMP-RG is shown to have sub-permil measurement precision with repeated analyses. Spatial resolution is approximately 20 μm, translating to approximately one week in temporal resolution in juvenile chinook salmon otoliths.; Chapter 4 demonstrates that three geochemical markers (87Sr/ 86Sr, Sr/Ca and Ba/Ca) provide nearly 100 percent site discrimination for 8 of the 10 rivers in this study, and they also discriminate two sites in the Sacramento-San Joaquin River Delta, based on analyzed water samples. This chapter does not include otolith data. The variability of these watershed markers relative to differences between sites is smallest for the 87Sr/86Sr signal.