Heterocyte glycolipids: novel tools to obtain high-resolution climate and ecosystem data from lacustrine archives

Lacustrine sediment sequences represent highly valuable, continuous records of continental climate and ecosystem change. As such, they provide vital information on both the timing and magnitude of long-term climate trends as well as abrupt climate change events and associated biotic responses. Assessing this information, however, is often not straight forward as many conventional lipid-based temperature proxies are affected not only by temperature but also other environmental parameters. Here, we show that the distribution of heterocyte glycolipids (HGs), which are synthesized by N₂-fixing heterocytous cyanobacteria and abundantly present in modern freshwater environments, are strongly correlated to lake surface water temperatures. We used these components, in form of the novel organic temperature proxy HDI₂₆, to reconstruct climate variations in tropical East Africa and simultaneously study the impact of climate change on the frequency and intensity of cyanobacterial blooms using a sediment record from Lake Tanganyika (Tanzania). HDI₂₆-reconstructed surface water temperatures varied from ~22 °C to 26 °C in Lake Tanganyika over the last 40,000 years. Lowest temperatures were observed during the Last Glacial Maximum, which is followed by a 3 to 4 °C deglacial warming to yield highest temperatures in the Late Holocene. This general warming trend is interrupted by up to 1 °C cooling during abrupt climate change events (e.g. Younger Dryas). Lipid-based reconstructions of past cyanobacterial activity indicate that although continuously present in Lake Tanganyika, cyanobacteria only became bloom-forming in the late Holocene and in particular during the last 200 years. Our data thus provides new insights and quantitative estimates on Pleistocene to Holocene climate changes in tropical East Africa and associated ecological responses. Given that heterocyte glycolipids are present ubiquitously in polar to tropical lakes and that they have been identified in lacustrine sediments of Early Cenozoic age, the HDI₂₆ and other HG-based indices provide valuable novel tools to extract the highly sensitive climate and ecosystem information that is stored in sediments of lakes worldwide.