Variability in the relationship between in-situ fluorescence and chlorophyll-a concentration in Faroese waters (2002-2019): Recommendations for database management

Phytoplankton form the base of marine foodwebs and sustain productivity throughout all trophic levels, upto and including commercially important fish species. In addition they impact climate through the sequestration of atmosheric carbon dioxide and the production of-active gases. Understanding natural variability and anthropogenic-driven changes in the spatial and temporal distribution of phytoplankton biomass is thus an important objective of both regional and global observing programs. The photosynthetic pigment chlorophyll-a (chl-a) serves as a proxy for phytoplankton biomass and can be either measured directly or inferred from in situ fluorescence. In situ fluorometry is widely-used because it is cost-effective and enables monitoring of phytoplankton bioamss from multiple observing platforms that integrate across the relevant range of spatial and temporal scales. In order to correctly interpret fluorescence profiles it is necessary to characterise the relationship between in situ fluorescence and measured chl-a concentrations, with a view to producing
field-corrected fluorescence datasets. The objective of this technical report was to examine the observed variability in fluorescent yield per chl-a in Faroese waters and provide recommendations for internal database management regarding the storage of field-corrected fluorescence datasets. We analysed
survey data collected by R/V Magnus Heinason during the period 2002-2019 and compared measured chl-a values with in situ fluorescence. There was significant seasonal and interannual variability in fluorescence yield per chl-a covering several orders of magnitude (0.06 – 13.3 Fchl:chl-1). Differences in water-column stratification and nutrient stoichiometry appear to explain some degree of the observed variability, most likley mediated through changes in phytoplankton community dynamics and photophysiological response. A significant shift in fluorescence yield per chl-a and silicate:nitrate ratios was observed from 2014 onwards, indicative of a potential shift in phytoplankton community dynamics on the Faroese shelf. Based on our analysis we conclude that it is not advisable to use a single empirical conversion to produce field-corrected fluorescence datasets in Faroese waters. Consequently we have
modified the FAMRI database such that all in situ fluorescence data are stored as raw data with no field corrections. We have provided guidance that all future data collections should be stored this way. In addition, we recommend that fluorometers should be sent for factory calibration once every three years to carefully monitor changes in sensitivity and analytical performance.