PHYTOPLANKTON CARBON-ISOTOPE FRACTIONATION DURING A DIATOM SPRING BLOOM IN A NORWEGIAN FJORD

The stable carbon isotope composition (delta(13)C) Of particulate orga nic carbon (POC) was measured in 3 size fractions (POCtotal, POC>20 mu m, POC<20 mu m) during a phytoplankton spring bloom dominated by the diatom Skeletonema costatum in Lindaspollene, a land-locked fjord in s outhern Norway. In addition to standard parameters for characterizing the phytoplankton bloom (chlorophyll, nutrient, and POC concentrations , and species composition), simultaneous measurements of delta(13)C Of dissolved inorganic carbon (DIC), total alkalinity and DIC concentrat ion were obtained to determine temporal trends in dissolved carbon dio xide concentration and in carbon isotope fractionation (epsilon(p)) of the POC size fractions. The carbon isotope composition of the >20 mu m size fraction, which was dominated by diatoms, was ca 2 parts per th ousand heavier than that of the <20 mu m fraction, which was mainly co mposed of flagellates. delta(13)C Of both size fractions increased by about 3 parts per thousand over the course of the bloom. A 5 parts per thousand increase in delta(13)C-POCtotal during the bloom resulted pa rtly from a shift in the phytoplankton community from a flagellate- to a diatom-dominated one. Carbon isotope fractionation of all fractions decreased with declining CO2(aq) concentration (14 to >6 mu mol l(-1) ). A positive correlation between epsilon(p) and [CO2(aq)] in the diat om size fraction was obtained for the period of exponential growth. De viation from this correlation occurred after the peak in cell density and chlorphyll a (chl a) concentration, when POC still continued to in crease, and may be related to changing phytoplankton growth rates or t o possible effects of nutrient (nitrate) limitation on epsilon(p). Com parison of these results with those of previous field studies shows th at, while an inverse relationship is consistently observed between eps ilon(p) and the ratio of instantaneous growth rate and CO2 concentrati on {mu(i)/[CO2(aq)]}, considerable scatter exists in this relationship . While this scatter may have partly resulted from inconsistencies bet ween the different studies in estimating phytoplankton growth rate, it could also reflect that factors other than growth rate and CO2 concen tration significantly contribute to determining isotope fractionation by marine phytoplankton in the natural environment.