The influence of soil chemistry on branched tetraether lipids in mid- and high latitude soils: Implications for brGDGT- based paleothermometry

Abstract

Branched glycerol dialkyl glycerol tetraethers (BrGDGTs) are a suite of orphan bacterial membrane lipids commonly used as paleo-environmental proxies for mean annual air temperature (MAT) and pH. Recent calibrations between the Methylation of Branched Tetraethers index (MBT′5ME) and MAT, based on modern surface soils (including peats), show a considerable amount of scatter, especially in mid- and high latitude soils, suggesting that brGDGT signals are influenced by additional environmental and/or biological controls at these sites. Here we test the impact of soil chemical gradients and bacterial community changes (16S rDNA sequence-based) on brGDGT distributions at two grasslands sites (Ossenkampen [NL], ForHot [IS]), and one agricultural site (Craibstone [UK]). In addition to the variation in soil chemistry, the ForHot site experiences belowground warming. Of the studied edaphic parameters, soil pH is the primary factor that explains simultaneous changes in both the bacterial community composition and the brGDGT distribution. Variations in the MBT′5ME at two sites without soil warming indeed correlate strongly to soil pH (r = 0.9–1.0, pH = 4.5–7.3), whereas pH explains part of the variation in the MBT′5ME at the site with soil warming (mean soil temperature ranging between 5 and 14 °C). At all sites, soil pH is positively related with the same brGDGTs (Ib, IIb, IIIb, IIIc, IIa′, IIb′, IIc′, IIIa′, IIb′, IIIc′) and influences the ratio between main brGDGT compounds Ia, IIa and IIIa, impacting the MBT′5ME values. This change in brGDGT distributions coincides with a change in the composition of the bacterial community at all sites. The bacterial clades that vary at the three experimental sites (specifically Acidobacteria subgroups 1, 2, 3, 6, 22) have previously been shown to also respond to soil pH on a global scale. As soil pH changes on geological timescales, the impact of changing pH on the MBT′5ME paleothermometer should be considered when performing paleoclimate studies.