Investigating paleotemperature proxies based on archaeal hydroxylated tetraether lipids

Reconstructing past sea surface temperatures (SSTs) is crucial for understanding Earth’s climatic history and validating climate models. Biomarker lipids, such as hydroxylated isoprenoidal glycerol dialkyl glycerol tetraethers (OH-isoGDGTs), produced by marine ammonia-oxidizing archaea, are emerging as promising paleotemperature proxies. This thesis investigated the applicability, environmental controls, and limitations of OH-isoGDGT-based proxies—%OH, RI-OH, RI-OH’, and the newly developed TEX86OH—to improve SST reconstructions across diverse marine and terrestrial settings. A global analysis of OH-isoGDGTs in 575 marine surface sediments revealed that OH-isoGDGT-0 correlated strongly with temperature, while OH-isoGDGT-1 and -2 were influenced by water depth in tropical regions, causing deviations in RI-OH and RI-OH’ proxies. TEX86OH, a new index, showed enhanced temperature sensitivity, particularly in cold environments (annual mean SSTs <15 °C), improving SST reconstructions in temperate and polar regions. Laboratory culture experiments with Nitrosopumilus piranensis D3C and Nitrosopumilus adriaticus NF5 showed that variations in the lipid compositions of non-hydroxylated isoGDGTs and OH-isoGDGTs were not linked to growth phases. However, lipid distributions varied with temperature, as evidenced by the increase in RI-OH’ with increasing temperature, while %OH and TEX86OH exhibit species-specific variability, underscoring the complexity of physiological controls on OH-isoGDGT distributions. The distribution of OH-isoGDGTs in non-marine settings, such as rivers and soils, was analyzed to assess the impact of terrestrial organic matter on various marine environments near river mouths (Kara Sea, Iberian Margin, northern Gulf of Mexico). OH-isoGDGTs showed lower %OH and higher RI-OH values, potentially leading to temperature overestimations in marine environments receiving substantial amount of terrestrial input. This was particularly evident in the Kara Sea sediments, where the influence of terrestrial inputs, combined with a large salinity gradient, affected OH-isoGDGT distributions, highlighting the need for regional studies of these proxies. The OH-isoGDGT distribution in surface sediments from semi-enclosed basins, such as the Mediterranean and Red Seas showed unique patterns. In the Mediterranean Sea, %OH decreased with increase in water depth, while TEX86OH increased, suggesting potential variations in OH-isoGDGT abundance among archaeal communities inhabiting different water depths. In contrast, RI-OH and RI-OH' were found to be primarily temperature-dependent. In the Red Sea, the low abundance of OH-isoGDGTs limited the applicability of OH-isoGDGT-based proxies in this region. The impact of water depth on OH-isoGDGT- and isoGDGT-based temperature proxies was assessed by comparing proximally located sediment cores from deep and shallow sites from the Chilean and Angola margins. Results showed generally similar proxy records of TEX86, RI-OH and RI-OH’ at both locations, indicating similar proxy source depths (with peak abundance at <350 m) at deep and shallow sites. However, regional differences in temperature estimates for RI-OH and RI-OH’ highlighted the need for improved calibrations. OH-isoGDGTs were also detected in sediments as old as ~36 Ma. Overall, this thesis demonstrated that OH-isoGDGT-based proxies, particularly TEX86OH, offer significant potential for paleotemperature reconstructions, especially in colder climates and regions with lower SSTs. However, further studies on archaeal lipid physiology, species-specific adaptations, and proxy calibrations are essential for refining their application as reliable climate indicators.