The Pleistocene Epoch began about 2.58 million years ago with a rapid drop in temperature and, according to the International Commission on Stratigraphy, ended about 11,700 years ago with the start of the retreat of the ice sheet. The Pleistocene is characterised by a much colder and more unstable climate than before, which fluctuated between arctic and warm temperate conditions, with corresponding ice advances (glaciations) and retreats (interglacials).
The number of glaciations and their ages were originally identified by examining successive glacial deposits left after each ice advance. This approach had the inherent disadvantage that each new glaciation largely destroyed the deposits from the previous one, making it difficult to identify individual episodes. Six named glacial periods were originally identified using this method. Each glacial and interglacial period was named after regions in the northern hemisphere, but the same event was often given different names in each region, hindering correlation.
Marine Isotope Stages
In the mid-1950s, a method for estimating temperature changes during the Pleistocene was developed by measuring the ratio of two oxygen atoms with the same atomic number but different mass (called isotopes because they occupy the same place on the Atomic Table). These ratios were measured in shell fragments recovered from cores taken from seabed sediments. This isotope ratio varied with the temperature of the seawater when the shells were formed and provided an indirect measure of terrestrial temperature.
This method has enabled Ice Age temperature changes to be identified at higher resolution and further back in time than the sedimentary technique, and has allowed the recognition of more glaciations. These temperature variations were used to define Marine Isotope Stages (MIS), with odd numbers denoting warm periods and even numbers denoting cold periods. MIS 1 is the current interglacial. Glaciations and interglacials are increasingly referred to by their MIS number rather than by name.
There are currently between ten and twelve major Pleistocene glacial periods recognised within the MIS framework, but higher resolution has also revealed smaller-scale interruptions to the overall trends. These interruptions are called stadials (colder) and interstadials (warmer). 103 MISs have been recognised, extending back to the beginning of the Pleistocene.
The cooling and warming cycles were irregular, both in the magnitude of the temperature changes and in their duration. The MIS temperature profiles also show that the warming phases at the end of each glaciation occurred more quickly than the cooling phases, implying that the ice-melt phases were relatively rapid. This is supported by the prevalence of meltwater sands and gravels in the geological record, which are indicative of meltwater floods. The temperature profiles also indicate that the length of the glacial-interglacial cycle has increased and that the temperature variation has become more pronounced since about 900,000 years ago.
The MIS method has its critics. It provides an indirect indication of surface temperatures and samples only a few locations on the seabed, which may not be representative of global temperatures. A problem recently highlighted is that glaciations and interglacials were not synchronous worldwide, so some question the global validity of the MIS (Gibbard & Hughes, 2020).
Milankovitch Cycles
There is some evidence which suggests a global uniformity and a cause of the temperature changes recorded in the MIS cycles.
A correlation was recognised between the MIS and regular variations in the heat reaching the Earth from the Sun (insolation) caused by changes in the distance of the Earth from the Sun. These are the Milankovitch Cycles. These variations are controlled by three types of perturbation in the Earth’s orbit and spin around the Sun; its eccentricity, its precession and its tilt. The periodicity of these cycles is 21-24,000 years, 41,000 years and 100,000 years.
Milankovitch noted that the effect of changes in the Sun’s heat would become more pronounced with increasing distance from the Equator. Near the poles, where insolation is weak, a small variation in total heat input has a more noticeable effect at the ground surface than in the tropics, where insolation is strong. (Imagine being in a hall lit by ten thousand candles. If a quarter were extinguished, leaving seven thousand five hundred, you probably wouldn’t notice much of a difference. If you were in a hall lit by four candles and one quarter was extinguished, you probably would notice a difference.)
The Milankovitch Cycles correlate with the glaciation cycles. Before 900,000 years ago, the complete glaciation cycle lasted about 40,000 years, matching the 41,000-year Milankovitch Cycle. It now matches the 100,000-year Milankovitch Cycle, with glaciations lasting about 80,000 years and interglacials 20,000 years. Bridgland (2006) correlated the end-of-glaciation River Terrace deposits, using sedimentary techniques, with the 100,000-year Milankovitch Cycle.
During the last 100,000-year glacial cycle, there have been longer periods of smaller-scale interglacials than of glacial conditions, although this pattern is superimposed on an overall cooling trend that peaked about 20,000 years ago (Fig. 5.7, Murton & Ballantyne, 2017). This interpretation is supported by the fossil and archaeological records, which document the movements of groups of large mammals as the ice sheet waxed and waned, alongside hunter-gatherer humans.
The effects of Milankovitch Cycles have been recognised elsewhere in the geological record – see the note on “How were flints in chalk formed”, which describes the distribution of flint bands in the Upper Cretaceous Chalk.
There is evidence of very rapid temperature increases of 15°C over tens of years in the Northern Hemisphere during the Ice Age, with the magnitude decreasing towards the equator and simultaneous but less pronounced cooling in the Southern Hemisphere. These are the Dansgaard-Oeschger Events (Masson-Delmotte, 2013, but not mentioned in Masson-Delmotte et al., 2021).
Andrew Coleman
Rev. 01/07/2026
References:
Bridgland, D. R. (2006). The Middle and Upper Pleistocene sequence in the Lower Thames: a record of Milankovitch climatic fluctuation and early human occupation of southern Britain. Proceedings of the Geologists’ Association, 117(3), 281–305. https://doi.org/10.1016/S0016-7878(06)80036-2
Gibbard, P. L., & Hughes, P. D. (2020). Terrestrial stratigraphical division in the quaternary and its correlation. Journal of the Geological Society, 178(2). https://doi.org/10.1144/jgs2020-134
Masson-Delmotte et al. (2021). IPCC, 2021: Climate Change 2021: The Physical Science Basis. Contribution of Working Group 1 to the 6th Assessment Report of the Intergovernmental Panel on Climate Change.
Masson-Delmotte, V. , et al. (2013). Fifth assessment report of the IPCC, Physical Science Basis, Ch. 5.
Murton, J. B., & Ballantyne, C. K. (2017). Periglacial and permafrost ground models for Great Britain. In Geological Society Engineering Geology Special Publication (Vol. 28, Issue 1, pp. 501–579). Geological Society of London. https://doi.org/10.1144/EGSP28.5
Geomusings 