Occurrence and abundance of 6-methyl branched glycerol dialkyl glycerol tetraethers in soils: Implications for palaeoclimate reconstruction
De Jonge, C.; Hopmans, E.C.; Zell, C.; Kim, J.H.; Schouten, S.; Sinninghe Damsté, J.S. (2014). Occurrence and abundance of 6-methyl branched glycerol dialkyl glycerol tetraethers in soils: Implications for palaeoclimate reconstruction. Geochim. Cosmochim. Acta 141: 97–112. dx.doi.org/10.1016/j.gca.2014.06.013
In: Geochimica et Cosmochimica Acta. Elsevier: Oxford,New York etc.. ISSN 0016-7037; e-ISSN 1872-9533, meer
|  |
| Auteurs | | Top |
|
- Kim, J.H., meer
- Schouten, S., meer
- Sinninghe Damsté, J.S., meer
|
|
| Abstract |
The distribution of branched glycerol dialkyl glycerol tetraethers (brGDGTs) in soils has been shown to correlate with the soil pH and the mean annual air temperature (MAT). This has been used to perform palaeoclimate reconstructions based on brGDGTs recovered from palaeosoils, freshwater, and marine sedimentary archives. Recently described 6-methyl brGDGTs were shown to co-elute with the 5-methyl brGDGTs that are used to calculate the CBT and MBT' indices used as palaeoclimate proxies. The impact of these 6-methyl brGDGTs on the established palaeoclimate proxies is unknown and will depend on their abundance in soils. Using improved chromatography, we quantified the fractional abundance of 6-methyl brGDGTs in globally distributed soils and show that they are abundant components, comprising on average 24% of the total amount of brGDGTs. All penta- and hexa-methylated brGDGTs (i.e. with zero to two cyclopentane moieties) were shown to comprise both 5- and 6-methyl isomers. The fractional abundances of the six 6-methyl brGDGTs correlate positively with each other, suggesting a common biological source in most soils, and correlate strongly with soil pH. The presence of the 6-methyl brGDGTs introduced scatter in the original MBT'/CBT-MAT calibration and caused a dependence on soil pH of the MBT'. Exclusion of the 6-methyl brGDGTs from the MBT', i.e. the newly defined MBT'5ME, shows that it is no longer related to soil pH. The correlation with MAT is improved, reducing the residual mean error (RMSE) from 6.2 to 4.8 °C. Also, the correlation of the CBT after the exclusion of the 6-methyl brGDGTs (defined as CBT5ME) with soil pH slightly improved. Furthermore, the separate quantification of the 6-methyl brGDGTs allows the definition of new indices. The CBT', which comprises the 6-methyl brGDGTs, is a substantially improved alternative to the CBT5ME, reducing the RMSE from 0.8 to 0.5 pH units. Also, the accuracy of MAT reconstructions can be improved using a multiple linear regression, the MATmr, decreasing the RMSE further to 4.6 °C. Furthermore, we introduce an index that allows the reconstruction of MAT in sites where only the ubiquitous brGDGT Ia, IIa and IIIa are present, the MATmrs. Our results imply that separate quantification of the 6- and 5-methyl brGDGT is essential for accurately quantifying brGDGTs in environmental samples and results in substantially improved MAT and soil pH reconstructions |
|
