Traditional in situ measurement techniques to determine sea-ice thicknesses include measuring the depth of holes drilled through the sea ice or measuring the length of sea-ice cores. ‘Hot-wire’ stakes have also been used to allow repeated measurements at the same sites over time. These methods are accurate, but in Antarctica such records are extremely limited in spatial and temporal extent because of the need to have personnel present to make the measurements. In situ measurement of the temperature profile across a growing ice–water interface has been used to determine the rate of increase in sea-ice thickness. This method has the advantage that measurements can be made with minimal personnel involvement throughout the growth season, but again records are limited to a few sites by logistical constraints. An ideal method for determining rates of change in sea-ice thicknesses would be able to be carried out remotely and retrospectively, so that any location of interest could be investigated. However, it is currently challenging for remote-sensing techniques to determine sea-ice thicknesses. Existing models to predict sea-ice thickness evolution are empirical and geographically specific, and require ice–air interface meteorological data. Retrospective deduction of sea-ice growth rates from in situ sea-ice salinity measurements is a long-standing technique, but such calculations are particularly vulnerable to natural millimeter- to centimeter-scale variations in sea-ice structure and are problematic when brine drainage occurs, as is typical during summer fieldwork. As a result of these difficulties, previous researchers have proposed a method for determining sea-ice growth rates based on the segregation of oxygen isotopes during the sea-ice formation process. To test the predictive ability of these methods, this presentation will compare rates of increase in sea-ice thicknesses determined from direct in situ measurements by coring, drilling, temperature probes, video camera recordings, and from oxygen isotope analysis for sea-ice and sea-water samples from sites in Antarctica. From these comparisons, suggestions will be made for modifications to existing isotope methods for calculations of rates of change in sea-ice thickness in the vicinity of ice shelves.