Methanesulfonic acid (MSA) datasets from two different Dronning Maud Land (DML) ice cores, B32 and M150, and one from the South Pole are investigated for their potential as proxies for past sea-ice variations around Antarctica. We have applied spectral and scale-space methods on the datasets in order to find out if there is a common signal or if these records are destroyed by post-depositional processes. The results suggest that despite all complicating factors, including low accumulation rate and post-depositional loss, these three MSA records from the polar plateau are surprisingly similar on timescales longer than about a century. Moreover, the reproducibility of the MSA records between the two DML sites is better than for the oxygen-18 records for these cores, which are usualy considered to be indicative of the variations in the local air temperature. Our results thereby strongly suggest that MSA concentrations can be attributed to the same physical processes on a regional scale. At interannual to decadal timescales the variations in MSA from the DML ice cores were found to show weak but essentially negative correlation with maximum winter sea-ice extent in the Atlantic sector of the Southern Ocean. These results are in line with a recent interpretation of MSA in coastal DML ice cores strongly related to atmopsheric circulation changes and a corresponding transport efficinecy of MSA inland. Analysis of longer sea-ice presence records from Signy Island and two marine sediment records from the subpolar South Atlantic suggests, however, the opposite relationship to exist on timescales longer than a few decades. We conclude therefore that higher MSA in these cores reflects generally heavy sea-ice conditions around Antarctica. Such a generally more severe sea-ice condition is also favourably in line with the stable water isotope records showing lower oxygen-18 (i.e. lower temperatures, when increase sea ice is most likely) in times of higher MSA.