A weighted-averaging regression and calibration model for inferring lakewater salinity from fossil diatom assemblages in saline lakes of the Vestfold hills: a new tool for interpreting Holocene lake histories in Antarctica

The relationship between surface sediment diatom assemblages and measured limnological variables in thirty-three coastal Antarctic lakes from the Vestfold Hills was examined by constructing a diatom-water chemistry dataset. Previous analysis of this dataset by canonical correspondence analysis revealed that salinity accounted for a significant amount of the variation in the distribution of the diatom assemblages. Weighted-averaging regression and calibration of this diatom-salinity relationship was used to establish a transfer function for the reconstruction of past lakewater salinity from fossil diatom assemblages. Weighted-averaging regression and calibration with classical deshrinking provided the best model for salinity reconstructions and this was applied to the fossil diatom assemblages from one of the saline lakes in the Vestfold Hills in order to assess its potential for palaeosalinity and palaeoclimate reconstruction.