The most dominant influence governing the observed isotopic values is temperature dependence. This yields strong annual cycles in ice cores, which are important for dating shorter ice cores. In general, isotopic content is determined by the net effects of conditions at the evaporation source, transport history and conditions at precipitation. Further, deposition processes at an ice coring site such as blowing snow and sastrugi can distort the retrieved isotope data. As such, the isotopic content of the ice may be considered as an indicator of the mean conditions in the hemisphere at the time of deposition. At interannual timescales weak signals corresponding to ENSO and the SAO are observed. At longer timescales, glacial/interglacial cycles are seen, from which quantitative estimates of temperature variations are made. Knowing these mean changes in temperatures and estimating changes in the insolation, the mean changes in the hemispheric circulation can be deduced from dynamic principles. However, it is difficult to provide detailed assessment of circulation changes without considering the complex interaction of physical processes and feedbacks that may be crucial to the isotopic content.
Using a General Circulation Model, these complex internal responses are resolved. Within the model many processes that determine the final isotopic content of Antarctic ice may be included to generate a diagnostic estimate of the isotopic concentration of Antarctic snow. The specific causes of isotopic depletion may be examined independently by performing boundary condition sensitivity experiments. It is found that the history of isotopic processes integrated along the transport path from the source region to the Antarctic is both crucial to the final observed content and not constant over the climate history. While this could lead to biased temperature reconstructions, it also introduces a relation to the characteristics of the cyclonic disturbances that dominate the Antarctic climate. This can be shown using an objective analysis technique applied to the model data. Under controlled conditions in the model, a range of climate scenarios can be conceived to indicate the possible influences of these cyclones on the observed isotope record.