Modelling of the Effect of Chromatographic Resolution on the Determination of the U^K₃₇’ Index

Abstract

The U^K₃₇^I index is based on the measurement of the relative abundance of C₃₇ alkenones in marine samples and is used to estimate absolute palaeo-sea surface temperatures. The reliability of the data (and the value of their interpretation) depends on the minimization of those factors that give rise to errors in the measurement of U^K₃₇^I. The index is calculated using gas chromatography, and here, the effect of chromatographic resolution on U^K₃₇^I is evaluated through the modelling of chromatograms with Gaussian or exponentially modified Gaussian peaks. It is shown that chromatographic resolution of the alkenones is indeed linked to the accuracy of U^K₃₇^I, although the magnitude of the error depends on the actual values of U^K₃₇^I, the degree of peak tailing, and the relative abundance of coelutants. However, it is not necessary to achieve baseline separation between the alkenones to obtain U^K₃₇^I values with errors that are climatically nonsignificant. Further work should demonstrate if these results can be reproduced experimentally, but in principle, the use of partially unresolved chromatograms with or without peak tailing does not appear to be a major source of error during U^K₃₇^I analysis.