Identifying and enumerating salmon lice larvae in plankton is the only way to obtain direct measures of sea lice infection pressure.
However, enumeration and identification by traditional methodology is too resource demanding for routine application and there have therefore been a number of efforts to develop more efficient methodologies including; quantitative real time PCR, eDNA analysis, quantitative fraction PCR and automated image analysis (FlowCam).
The molecular methodologies suffer from being vulnerable to variability between samples and therefore requiring extensive validation while automated image analysis methods previously tested were too imprecise for the purpose. Fluorescence microscopy was suggested as a solution in 2017, but the presented methodology yielded variable results indicating that further development was needed.
The Fluorolice project was designed to develop the methodology by investigating the salmon louse fluorescence profiles and their dependence on treatments, age etc. Flourolice would furthermore investigate the fluorescence profile with respect to other zooplankton organisms and explore the possibilities for further development and automatization of sample processing.
The project identified the most promising fluorescence peak to be at an excitation wavelength of 470 ± 40 nm and an emission wavelength of 525 ± 50 nm and it was found that illumination was required to be even across the entire field of view. The fluorescence was found to be weak in unpreserved samples and to be variable in samples stored on ethanol.
This indicates that live fluorescence will not be readily applicable to live samples. Formalin fixation, in contrast, yielded strong and specific fluorescence that did not overlap with most other plankton species. A few species did overlap, however, and it is, therefore, necessary to consider also morphology in identification.
Tests indicated that the processing speed for plankton samples can be increased by a factor of 10 in manual sample processing. Preliminary tests suggested that automated identification should be possible as well, but the development of this requires algorithms trained on a very large number of pictures which was beyond the reach of the present project.
While the results enable much faster processing of plankton samples, widespread implementation of the methodology requires standardized sampling of large volumes and optional development of algorism for sample analyses.
Fluorolice was coordinated by the Institute of Marine Research in Norway in collaboration with Marine Scotland Science, University of Stirling, OptoScale and Firum. The project was funded by FHF – Fiskeri- og havnruksnæringesn forskningsfinansiering project number 901508
Read the final Fluorolice report here.
Contact person: Gunnvør á Norði
