Help Guide Disclaimer Contact us Login
  Advanced SearchBrowse





Spectral fluorescence of chlorophyll and phycobilins as an in-situ tool of phytoplankton analysis - models, algorithms and instruments


Beutler,  Martin
Department Ecophysiology, Max Planck Institute for Limnology, Max Planck Institute for Evolutionary Biology, Max Planck Society;

Wiltshire,  K. H.
Department Ecophysiology, Max Planck Institute for Limnology, Max Planck Institute for Evolutionary Biology, Max Planck Society;

There are no locators available
Fulltext (public)

(Publisher version), 3MB

Supplementary Material (public)
There is no public supplementary material available

Beutler, M. (2003). Spectral fluorescence of chlorophyll and phycobilins as an in-situ tool of phytoplankton analysis - models, algorithms and instruments. PhD Thesis, Christian-Albrechts-Universität, Kiel.

Cite as:
Fingerprints of excitation and emission spectra of chlorophyll and phycobilin fluorescence can be used to differentiate 'spectral groups' of microalgae in vivo and in situ , e. g. vertical profiles can be taken within a few minutes. The investigated spectral groups of algae (green group - chlorophyta; blue - blue cyanobacteria; brown - heterokontophyta, haptophyta, dinophyta; red - red cyanobacteria and mixed - cryptophyta) are each characterised by a specific composition of photosynthetic antennae pigments and, consequently, by a specific excitation and emission spectrum of the chlorophyll and phycobilin fluorescence. Particularly relevant are chlorophyll a, chlorophyll c, phycocyanin, phycoerythrin, fucoxanthin and peridinin. In a first approach, a laboratory based instrument and a submersible instrument were constructed containing light-emitting diodes to excite chlorophyll fluorescence in five distinct wavelength ranges to facilitate the differentiation of four spectral algal groups (green, blue, brown, mixed). They were measured under a fixed emission wavelength. Norm spectra were determined for the four spectral algal groups (several species per group). Using these norm spectra and the actual five-point excitation spectrum of a water sample, an estimate of the group-specific chlorophyll concentration is rapidly obtained for each algal group. This was accomplished by the development of a fast mathematical fit procedure. In vivo and in situ measurements based on calibration experiments were compared with results obtained by high performance liquid chromatography and biovolume estimations from the light microscope. Depth profiles of the distribution of spectral algal groups taken over a time period of few minutes were shown. The described method for algal differentiation opens new research areas and monitoring and supervision facilities related to photosynthetic primary production in aquatic environments. Yellow substances (coloured dissolved organic matter) may interfere with the measurement because of an overlap of the excitation spectra with those of phytoplankton. The use of an ultra-violet excitation source (370 nm light-emitting diode) enabled differentiation between algal fluorescence and fluorescence by yellow substances. The resulting six-point excitation spectra were deconvoluted on the basis of norm spectra. A mean norm spectrum for yellow substances was obtained from natural samples. In a new submersible instrument the correction of chlorophyll fluorescence measurements for the influence of yellow substances was tested in in vivo and in situ experiments. Specific problems associated with the principle of a free-falling depth profiler for algae discrimination were also considered. When F0, F and Fm are determined sequentially with one measuring cell, then phytoplankton inside the cell experiences a different light history according to different locations in the cell. This leads to a superposition of different induction curves of chlorophyll fluorescence. Mathematical algorithms were developed that enable theevaluation of the integral fluorescence signal (averaged for 1s) for different velocities of the falling probe. This yields a correction factor which allows the usage of calibration factors obtained from stationary suspensions for the determination of algal concentrations in flowing suspensions. The predictions of the model were compared with measurements in flowing suspensions containing chlorophyta, cyanobacteria, cryptophyta and diatoms. The comparison showed the reliability of the algorithms. The requirement of corrections by the algorithm was high for dark-adapted cells and less important for light-adapted cells. Fluorometric determination of the chlorophyll content of cyanobacteria is impeded by the unique structure of their photosynthetic apparatus, that is, the phycobilisomes in the lightharvesting antennae. The problems are caused by the variations in the ratio of the pigment phycocyanin to chlorophyll a resulting from adaptation to varying environmental conditions. In order to improve fluorometric analysis of algae a simplified energy distribution model describing energy pathways in the cyanobacterial photosynthetic apparatus was conceptualised. Two sets of mathematical equations were derived from this model and tested. Fluorescence of cyanobacteria was measured with a new fluorometer at seven excitation wavelength ranges and at three detection channels (650 nm, 685 nm and 720 nm) in vivo. By employing a new fit procedure it became possible to correct for variations in the cyanobacterial fluorescence excitation spectra and to account for other phytoplankton signals. The effect of energy state transitions on the phycocyanin fluorescence emission of phycobilisomes were documented. The additional use of the phycocyanin fluorescence signal at 650 nm in combination with the previously developed mathematical approach for phytoplankton analysis based on chlorophyll fluorescence spectroscopy allows a more detailed study of cyanobacteria and other phytoplankton in vivo and in situ. The detection of red cyanobacteria with these newly developed methods is not possible because of adaptation processes of the cyanobacterial photosynthetic apparatus and spectral interferences with cryptophyta. To overcome these problems a simplified energy distribution model accounting for energy pathways in the red cyanobacterial photosynthetic apparatus and the apparatus of cryptophyta was designed. Mathematical equations were derived that enabled the calculation of the pigment content in both organisms: cryptophyta and red cyanobacteria. This resulted in the extension of a fluorometer previously developed for phytoplankton, with seven excitation wavelengths and four detection channels (600 nm, 620 nm, 650 nm and 685 nm). An extension of the fit procedure allowed corrections for variations in the fluorescence excitation spectra of red cyanobacteria and cryptophyta in the presence of other phytoplankton signals. The new approach provided correct fluorometric pigment estimation also in the presence of energy state transitions. The combination of fluorescence emission excitation matrices, the fluorescence models and the enhanced fit algorithm provides valuable information for phytoplankton analysis. Finally, the set-up was tested successfully with natural samples. It enabled the determination of chlorophyll in five spectral groups of phytoplankton and of the phycobilins in three spectral types. A correction for yellow substances was included using mean fluorescence spectra of filtered natural samples. The pigment estimation of this method was compared to reference estimates obtained by high performance liquid chromatography and wet chemicalanalysis of natural freshwater samples. Reference and fluorometric methods showed similar results. This measuring principle was installed as a submersible instrument which makes possible the measurement of fluorescence depth-profiles via pigment estimation in situ in a time scale of a few minutes. Profiles obtained with the instrument were compared to those obtained with the stationary instrument. The correlation between both methods was very high. This demonstrates the success and the large step forward in phytoplankton analysis achieved by this method.