New calibration of terrestrial brGDGT paleothermometer deconvolves distinct temperature responses of two isomer sets

Abstract

Bacterial branched glycerol dialkyl glycerol tetraethers (brGDGTs) are emerging as one of the most promising tools for quantifying past temperatures across a large variety of continental settings. However, temperature-independent parameters such as the ratio of 6-methyl vs. 5-methyl brGDGTs and bacterial community composition, might contribute to the high uncertainty of the brGDGT paleothermometer. In this work, we investigated brGDGT distributions in surface soils surrounding Lake Qinghai, and reanalyzed previously published brGDGT data. The results show that tetramethylated brGDGTs can be produced along with 6-methyl brGDGTs in the absence of 5-methyl brGDGTs. The additional production of tetramethylated brGDGTs related to 6-methyl brGDGTs, in contradiction to the view that tetramethylated brGDGTs are only correlated with 5-methyl brGDGTs, can potentially bias the MBT′5ME paleothermometer. Accordingly, tetramethylated brGDGTs were hypothetically separated into the 5-methyl and 6-methyl sets, and the temperature responses of the two brGDGT sets were separately quantified. Based on theoretical calculation, we further developed a new temperature calibration model that effectively mitigates the mixing effect of the two sets likely associated with bacterial community shift. As demonstrated in downcore applications, the new calibration is promising for brGDGT-based growth temperature reconstruction in loess and lake sediments, especially when the relative abundance of 6-methyl vs. 5-methyl brGDGTs varies substantially.