Date of Award

5-2008

Level of Access

Campus-Only Thesis

Degree Name

Master of Science (MS)

Department

Oceanography

Advisor

Collin Roesler

Second Committee Member

Andrew Thomas

Third Committee Member

Howard Patterson

Abstract

Single excitation fluorometers are commonly used to estimate phytoplankton biomass, using fluorescence as a proxy for chlorophyll α, which is, in turn, a proxy for carbon biomass. However the interpretation of in situ fluorescence is complicated because the relationship between chlorophyll α and phytoplankton carbon varies with species, light acclimation, nutrient status and growth phase. Additionally the relationship between in vivo chlorophyll α fluorescence and chlorophyll α concentration varies with intracellular pigment concentration and accessory pigment composition as a function of species, photo acclimation and inhibition, nutrient limitation, and temperature among other factors. If unaccounted for, changes in these proxy relationships can potentially produce significant errors and inconsistencies in biomass estimates. With these factors in mind, we undertook to characterize a new fluorometer with three excitation wavelengths (the 3X1M) paired with a sensor measuring particle backscattering, fluorescence of colored dissolved organic matter and phycoerythrin, and temperature. We hypothesized that this sensor package consisting of two types of optical sensors (ECO Triplets, WET Labs, Inc.) could reduce or eliminate many of these sources of variability and use in situ fluorescence to estimate biomass with <10% uncertainty and to additionally provide estimates of phytoplankton community composition. Temperature was found to influence fluorometer readings in such a way that uncorrected in situ estimates of wintertime or cold region chlorophyll estimates could yield a 100% error. Colored dissolved organic material was found to contaminate fluorometrically measured chlorophyll, causing overestimates of chlorophyll concentration by approximately 0.4 to 4 mg/m3 in typical freshwater and coastal settings (where chlorophyll concentrations are generally >5 mg/m3 and 1-5 mg/m3 respectively). The chlorophyll a fluorescence response of the 3X1M was calibrated with thirteen diverse species of phytoplankton grown acclimated to two light conditions. The largest variability in calibration response factors was due to differences in accessory pigment composition between species. Using the median calibration slope derived from all species and conditions, biomass could generally be estimated within a factor of two. Improved estimates were obtained by selecting specific pigment-based lineage-specific calibration factors which are obtained using fluorescence ratios. The fluorescence ratios between the different excitation channels of the 3X1M were calculated and found to be linearly related to the pigment absorption ratios. Significant differences among pigment lineage groups were found between the 435 nm / 532 nm and 470 nm / 532 nm fluorescence excitation ratios, enabling the selection of a lineage-specific calibration factor. Biomass estimates in China Lake using the median calibration were within 60% of chemically derived chlorophyll concentration, and typically within 30%. These differences were reduced to <30% overall and typically <6% by using our derived temperature and CDOM corrections and selecting the appropriate pigment group calibration factor, determined by using pigment ratios.

Share