| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 1.18 MB | Adobe PDF |
Advisor(s)
Abstract(s)
Harmful algal blooms are responsible worldwide for the contamination of fishery resources, with potential impacts on seafood safety and public health. Most coastal countries rely on an intense monitoring program for the surveillance of toxic algae occurrence and shellfish contamination. The present study investigates the use of near infrared (NIR) spectroscopy for the rapid in situ determination of cell concentrations of toxic algae in seawater. The paralytic shellfish poisoning (PSP) toxin-producing dinoflagellate Gymnodinium catenatum was selected for this study. The spectral modeling by partial least squares (PLS) regression based on the recorded NIR spectra enabled the building of highly accurate (R2 = 0.92) models for cell abundance. The models also provided a good correlation between toxins measured by the conventional methods (high-performance liquid chromatography with fluorescence detection (HPLC-FLD)) and the levels predicted by the PLS/NIR models. This study represents the first necessary step in investigating the potential of application of NIR spectroscopy for algae bloom detection and alerting.
Description
Keywords
phytoplankton FTIR spectroscopy monitoring harmful algae bloom saxitoxin