Determining the age of the ice in an ice core can be done in a number of ways. Counting layers, chemical analysis and mathematical models are all used.
Annual layers of snowfall recorded in an ice core can be counted — in much the same way that tree-rings can be counted — to determine the age of the ice.
This method can present challenges. Many cores come from regions where the yearly snowfall accumulation is too small for the annual layers to be distinguished. Even in cores where the yearly snowfall produces thick layers, the nature of glacier flow stretches and thins layers as they get buried deeper. This flow-thinning means that annual layer counting eventually becomes impossible in all deep cores.
Layers in ice cores can become apparent when the core is analysed for a chemical signal that varies with the seasons. The clearest dating is obtained when several seasonal signals are examined and compared.
Where layer-counting is not possible, dating generally relies upon mathematical models of ice flow.
Another useful technique is to identify events that are verified by other types of climate records, such as historical, tree ring and sedimentary records. These all have independent methods of dating, and so the timing of a major climate shift or volcanic eruption can be used to synchronise the age scales.