A novel biomarker-based proxy for the spring phytoplankton bloom in Arctic and sub-arctic settings – HBI T25

Abstract The spring phytoplankton bloom is a characteristic feature of mid-high latitudes in modern times, but can be challenging to identify in palaeo records. In the current study, we investigated the absolute and relative distributions of two diatom-derived tri-unsaturated highly branched isoprenoid (HBI) lipids, at least one of which has previously been suggested to be a possible proxy for the productive region of the marginal ice zone (MIZ) in the Polar Regions. Based on a comparison of their distributions in surface sediments from the Barents Sea and neighbouring regions with a range of oceanographic parameters, we identify, via principal component analysis, a strong association between the relative proportion of the two HBIs and satellite-derived spring chlorophyll a (chl a ) concentration. Further, based on agglomerative hierarchical clustering, we identify two clusters of HBI biomarker ratios and spring chl a together with a potential threshold biomarker ratio (termed HBI TR 25 ) for the spring phytoplankton bloom. A modified version of HBI TR 25 (i.e. HBI T 25 ) provides a potentially more straightforward binary measure of the spring phytoplankton bloom. Analysis of HBI TR 25 and HBI T 25 values in a series of short (spanning recent centuries) and long (Holocene) sediment cores from the region provides an initial evaluation of the applicability of this novel proxy in the palaeo record. Outcomes are mainly consistent with the findings from the surface sediments and with other proxy-based reconstructions, including estimates of past sea ice cover, which is well-known to influence primary production in the region. Indeed, we suggest that the new HBI T 25 phytoplankton bloom proxy may also represent an important new tool for characterising the MIZ in palaeo records, especially when used alongside well-established sea ice proxies, such as IP 25 and PIP 25 . Despite the largely empirical nature of the study, we also provide a possible explanation for the observed biomarker ratio-chl a relationship. Thus, a previous laboratory investigation showed that the distributions of the same two HBIs analysed herein in their likely source (viz. Rhizosolenia setigera ) was strongly influenced by culture temperature and growth rate. Confirmation of the generality of our findings and of the causal relationship between HBI T 25 and the spring phytoplankton bloom will, however, require further laboratory- and field-based studies in the future.