Reconstructing N₂-fixing cyanobacterial blooms in the Baltic Sea beyond observations using 6- and 7-methylheptadecane in sediments as specific biomarkers

Abstract

Summer cyanobacterial blooms represent a threat to the Baltic Sea ecosystem, causing deoxygenation of the bottom water and the spread of the so-called dead zones. The history of the Baltic Sea cyanobacterial blooms is known from in situ and satellite observations since the early 1980s but is still not well understood. By comparing both weekly resolved sediment trap material and a well-dated sediment core from the eastern Gotland Basin with monitoring and satellite cyanobacterial data of the last ca. 35 years, it is shown here that 6- and 7-methylheptadecane lipids (expressed as 6+7Me-C17 : 0) may be potentially considered semiquantitative biomarkers for diazotrophic cyanobacteria, and more specifically for Nodularia spumigena. Using this organic proxy, it was thus possible to reconstruct the history of cyanobacterial blooms beyond the observational period with a resolution of 2–4 years since 1860. Cyanobacteria were constantly present but in relatively low abundance until 1920, when they started to alternate between periods with high and low abundance. Interestingly, there seems to be no significant increase in cyanobacterial abundance in the 1950s, when eutrophication and deoxygenation of the Baltic Sea increased considerably. While the early increase in cyanobacteria may be related to a small increase in phosphorus loading, decadal to multi-decadal fluctuations are likely related to variability in the Baltic Sea surface temperature and, ultimately, to the Atlantic Multi-decadal Oscillation. A 7000-year 6+7Me-C17 : 0 record from the Bothnian Sea also suggests a relationship with the mean summer temperature in the Baltic Sea region but at a multi-centennial to multi-millennial timescale. The intensity of the cyanobacterial blooms in the Baltic Sea is thus likely mainly related to natural processes such as temperature variability, at least at a multi-decadal to multi-millennial timescale.