Oh Magadi! Interpreting isoGDGTs and n-alkanes in a saline tropical lake: Lake Magadi, Kenya

The Hominin Sites and Paleolakes Drilling Project (HSPDP) seeks to understand the paleoclimatic and paleoenvironmental context of hominin adaptation and evolution by analysis of paleolacustrine cores taken near key hominin fossil and artifact localities in Kenya and Ethiopia. We present biomarker and compound specific isotope data from a 200 m drill core from Lake Magadi, Kenya. Located 20 km from the Koora Plain in the southern Kenya Rift, and adjacent to the Olorgesailie basin, Lake Magadi is in one of the richest Early-Late Pleistocene archaeological localities in Africa, a region that has been key in debates about the relationship between climate and evolution. Present-day Lake Magadi is a saline pan, a descendant of a series of paleolakes that have occupied its drainage basin and progressively dried for approximately one million years. Nearly 70% of samples analyzed for n-alkanes recorded a robust terrestrial signal. The majority of samples did not contain the complete suite of branched GDGTs necessary to reconstruct temperature from the Methylation of Branched Tetraethers and Cyclisation of Branched Tetraethers (MBT/CBT; Weijers et al., 2007) proxy. The TetraEther indeX with 86 carbon atoms (TEX86; Schouten et al., 2002) temperature proxy was established for 90% of samples analyzed for isoGDGTs, however the Methane and Ring Indices (Zhang et al., 2011; Zhang et al., 2016) suggest that the TEX86 is not applicable to temperature reconstruction at Magadi. Despite this, the Magadi TEX86 temperature reconstruction appears to agree with not only the trends in our n-alkane data but with other regional and global records, including the GRIP-2 δ18O record. We compare our temperature data to other records in the region, and investigate influences on our TEX86 data including microbial community turnover and lake drying.