Case study: In situ 14C exposure age dating in Antarctica

This page was written by Dr Keir Nichols, who is using 14C exposure age dating to unravel the past behaviour of the West Antarctic Ice Sheet (WAIS).

Given that the ice sheets in Antarctica and Greenland contain enough water to raise global sea level by about 65 metres, it is important to understand how ice sheets have changed in the past. Knowing how they have changed, and under what kind of environmental conditions, means we can better understand how they will respond to future changes in climate.

Millions of people around the world live close to sea level, from places like Bangladesh, the Netherlands, and low-lying islands in the Pacific and Indian oceans, to cities like New Orleans, Jakarta, and London. How quickly have ice sheets melted, and added to sea level, in the past, and how can we work this out? One way is through measuring something called cosmogenic nuclides.

A cosmic ray (in this case a high energy neutron) interacts with an oxygen atom in quartz to produce in situ 14C, just one of many cosmogenic nuclides.

Cosmic rays constantly hit the surface of the Earth, producing rare isotopes known as cosmogenic nuclides within minerals in rocks. Cosmic rays contain enough energy to break apart the atoms that make up rocks, leaving behind new cosmogenic nuclides. You can think of a cosmic ray as a cue ball and the atoms in rocks as a triangle of red balls in a game of snooker. When a rock is exposed to the sky, cosmogenic nuclides are constantly being made within it. 

But what has this got to do with glaciers and understanding their past?

Glaciers block cosmic rays

When a glacier grows larger and covers a rock, cosmic rays can no longer reach the rock and production of cosmogenic nuclides pauses. If a glacier later thins and reveals the rock, cosmogenic nuclides start to be made within the rock crystals again. 

We can measure the amount of a cosmogenic nuclide contained in a rock. If we know the rate at which the cosmogenic nuclide are produced (the production rate), we can convert this number into an exposure age. The age, usually in thousands of years (ka), tells us how long a surface has been exposed to cosmic rays or, in other words, how long ago in the past a glacier thinned to uncover the rock. 

Cartoon illustrating cosmogenic nuclide exposure ages. A glacier transports an erratic boulder, and then recedes, exposing it to cosmic rays. Spallation reactions occur in minerals in the rocks upon bombardment by cosmic rays. By sampling the rocks and separating certain minerals (such as quartz or pyroxene) and calculating the amount of these minerals (as a ratio to other, stable, minerals), we can work out how long the rock has been exposed on the earth’s surface.

The most commonly-measured cosmogenic nuclide is 10Be, which is produced in the mineral quartz. It is radioactive with a half-life of 1.4 million years, which means for every 1.4 million years that passes, the number of 10Be atoms will decrease by one half. When we convert concentrations to ages, we assume everything we measure was made from the most recent time that a rock was uncovered by a glacier. But what happens when this isn’t the case? 

Glacial erosion is needed to ‘reset’ the clock

We rely on glacial erosion to ‘reset’ the cosmogenic nuclide concentration of a rock surface and ensure our cosmogenic exposure ages represent one period of exposure.

When a glacier expands, it can erode the landscape beneath, removing cosmogenic nuclides that built up in the past. Evidence of glacial erosion is found in many areas containing glacial landforms; picture the deep U-shaped valleys of Yosemite National Park or the Scottish Highlands.

The case of minimal glacial erosion

In some places, including Antarctica, glaciers can be cold-based, which means they are frozen to their bed. Rather than eroding the rock beneath them, cold-based glaciers stick to their bed, preserving the landscape beneath (e.g. Stroeven et al., 2002; Sugden et al., 2005), causing problems with exposure dating. 

We know the size of ice sheets waxes and wanes with time; they are larger in glacial periods and smaller during interglacials. If a surface is covered by cold-based ice many times over a long timescale, cosmogenic nuclides from previous interglacials can be preserved. 

The assumption that everything we measure today was only produced during the most recent time a sample was uncovered by ice is not met, resulting in exposure ages that are much older than the true time that a surface was most recently exposed.

Left: Exposure ages sourced from Johnson et al. (2019). Right: Bedrock at the Lassiter Coast, southern Antarctic Peninsula. Photo credit – Jo Johnson, British Antarctic Survey.

Exposure age dating in Antarctica

The graph above shows 10Be exposure ages from pieces of bedrock collected by Johnson et al. (2019) from the Lassiter Coast, an area on the southern Antarctic Peninsula. These samples were collected to study the history of the West Antarctic Ice Sheet. Most of the ages are very old, with the majority over 100 thousand years. 

There are two ways we might interpret the exposure ages. Firstly, the ages are accurate and ice has not covered these rocks for hundreds of thousands of years. However, we know that the West Antarctic Ice Sheet was much larger only 20 thousand years ago, during a time called the Last Glacial Maximum (LGM) (e.g. RAISED Consortium., 2014). So, instead, the very old ages might tell us something else; the rocks have been buried and uncovered by cold-based ice many times, including during the LGM. With no glacial erosion during ice cover, the long half-life of 10Be means it can be preserved in rock for millions of years.

How can we identify which interpretation is correct? Was ice thicker at the Lasster Coast recently, or has it been uncovered for a long time? 

Using 14C to unravel complex exposures

Another cosmogenic nuclide, in situ 14C, can help us answer this question. This slightly heavier, radioactive isotope of carbon is used in radiocarbon dating. With radiocarbon dating, we measure 14C in the remains of organisms to work out how long ago they died. 

With exposure age dating we are instead measuring 14C made in the mineral quartz. Quartz contains oxygen (SiO2), and cosmic rays interact with these oxygen atoms to make 14C (see diagram above). To prevent confusing exposure dating using 14C in rocks with traditional radiocarbon dating of organics, we refer to the former as in situ 14C.    

So why do we measure in situ 14C when we can already measure 10Be? In situ 14C has a very short half-life of 5730 years. The short half-life of 14C, compared to the half-life of 10Be (1.4 million years), means that even with no glacial erosion, the 14C concentration in a rock will rapidly decrease through radioactive decay. The animation below shows how the 10Be and in situ14C concentrations change in a sample collected from the surface of a mountain poking out of an ice sheet.   

Animation showing the build up of two cosmogenic nuclides, 10Be (purple) and in situ 14C (yellow), in a sample (red circle). The dashed line shows the in situ 14C concentration if the sample was not covered. Animation inspired by Hippe (2017

The sample is exposed to cosmic rays 46 thousand years ago and the concentration of both nuclides increases. When the ice sheet is thicker at the LGM (26 thousand years ago), it covers the sample. The amount of 10Be in the sample decreases very slowly, whilst the concentration of in situ 14C decreases rapidly. The samples are exposed again 7 thousand years ago. 

The 10Be concentration at the end of the animation contains a lot of 10Be that was made both before and after the sample was covered by ice. 

On the other hand, the vast majority of 14C at the end of the animation was made after the rock was covered by ice. The resulting 10Be age will be much older than the true exposure age of 7 thousand years, whilst the in situ 14C age will be much closer to it.

10Be (purple) and in situ 14C (yellow) exposure ages from the Lassiter Coast. Note the logarithmic scale on the x axis. Data source: Johnson et al. (2019), Nichols et al. (2019).

Using 14C to determine the glacial history of the Lassiter Coast

To shed light on the glacial history of the Lassiter Coast, Johnson et al. (2019) measured the in situ 14C concentration in some of the same rocks that were already measured for 10Be. All in situ 14C ages (above, yellow) are younger than the LGM, contrasting with the older 10Be ages (purple). 

With young in situ 14C and very old 10Be exposure ages from samples collected from ∼400 m above the modern Antarctic Ice Sheet, we now know that (i) the ice sheet was at least ∼400 m thicker at the LGM than it is now, and (ii) the Lassiter Coast has been covered by cold-based ice multiple times over many hundreds of thousands of years. 

Using 14C in data-calibrated computer simulations

So why does this matter? This information on how much thicker the Antarctic Ice Sheet was in the past, how it has changed since, and under what conditions, will be used to test numerical ice sheet models

Models that can reproduce past changes in ice sheets can be used to model the response of ice sheets to future climate scenarios with greater confidence. Perhaps most importantly, these models can provide vital information on how quickly the Antarctic Ice Sheets will contribute to sea level rise (e.g. Levermann et al., 2020). These types of studies then form an important part of reports by the Intergovernmental Panel on Climate Change (IPCC), the most recent of which was published in 2021.

Drawbacks of in situ 14C for exposure age dating

Because 14C is absorbed by living organisms, we can find 14C in organic chemicals used in laboratories. When using in situ 14C exposure age dating, as well as traditional radiocarbon dating, is it especially important to avoid contamination of sample material with modern 14C (e.g. McIntyre et al., 2016; Nichols and Goehring, 2019). 

Different cosmogenic nuclides are useful for studying different periods of time. Whilst the short half-life of 14C provides advantages described above, it also limits the time that we can use 14C in exposure age dating studies. Note the dashed yellow line in the animation above. The concentration of 14C is flattening-out because a balance between radioactive decay and  production from cosmic rays is being achieved. 

After about 35 thousand years (6 half-lives) of continuous exposure, the in situ 14C concentration in a rock will reach what is known as a saturation concentration. If we measure the saturation concentration in a sample, we know that the rock has been exposed for at least 35 thousand years. Whilst this can be a useful result in itself, it also means we cannot measure exposure ages for events older than this time using this cosmogenic nuclide. On the other hand, 10Be can provide exposure ages approaching 6 million years.

Another drawback of in situ14C, which will hopefully change in the future, is that very few laboratories worldwide have the capability to extract carbon from quartz. Laboratory equipment is expensive, and extraction requires close monitoring and use of a lot of liquid nitrogen.

Further reading on exposure age dating


RAISED Consortium, 2014. A community-based geological reconstruction of Antarctic Ice Sheet deglaciation since the Last Glacial Maximum. Quaternary Science Reviews, 100: 1-9.

Hippe, K., 2017. Constraining processes of landscape change with combined in situ cosmogenic 14C-10Be analysis, Quaternary Science Reviews, 173: 1-19. 

IPCC, 2021: Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [Masson-Delmotte, V., P. Zhai, A. Pirani, S.L. Connors, C. Péan, S. Berger, N. Caud, Y. Chen, L. Goldfarb, M.I. Gomis, M. Huang, K. Leitzell, E. Lonnoy, J.B.R. Matthews, T.K. Maycock, T. Waterfield, O. Yelekçi, R. Yu, and B. Zhou (eds.)]. Cambridge University Press. In Press. 

Johnson, J. S., Nichols, K. A., Goehring, B. M., Balco, G. and Schaefer, J. M., 2019. Abrupt mid-Holocene ice loss in the western Weddell Sea Embayment of Antarctica. Earth and Planetary Science Letters, 518: 127-135.

Levermann, A., Winkelmann, R., Albrecht, T., Goelzer, H., Golledge, N. R., Greve, R., Huybrechts, P., Jordan, J., Leguy, G., Martin, D., Morlighem, M., Pattyn, F., Pollard, D., Quiquet, A., Rodehacke, C., Seroussi, H., Sutter, J., Zhang, T., Van Breedam, J., Calov, R., Deconto, R., Dumas, C., Garbe, J., Hilmar Gudmundsson, G., Hoffman, M. J., Humbert, A., Kleiner, T., Lipscomb, W. H., Meinshausen, M., Ng, E., Nowicki, S. M. J., Perego, M., Price, S. F., Saito, F., Schlegel, N. J., Sun, S. and Van De Wal, R. S. W., 2020. Projecting Antarctica’s contribution to future sea level rise from basal ice shelf melt using linear response functions of 16 ice sheet models (LARMIP-2). Earth System Dynamics, 11(1): 35-76.

McIntyre, C., Lechleitner, F., Lang, S., Haghiour, N., Fahrni, S., Wacker, L., & Synal, H., 2016. 14C Contamination Testing in Natural Abundance Laboratories: A New Preparation Method Using Wet Chemical Oxidation and Some Experiences. Radiocarbon, 58(4), 935-941.

Nichols, K. A., Goehring, B. M., Balco, G., Johnson, J. S., Hein, A. S. and Todd, C., 2019. New Last Glacial Maximum ice thickness constraints for the Weddell Sea Embayment, Antarctica. Cryosphere, 13: 2935-2951.

Nichols, K. A. and Goehring, B. M, 2019. Isolation of quartz for cosmogenic in situ 14C analysis. Geochronology, 1: 43-52.

Stroeven, A. P., Fabel, D., Harbor, J., Hättestrand, C. and Kleman, J., 2002. Quantifying the erosional impact of the Fennoscandian ice sheet in the Tornetr sk-Narvik corridor, northern

Sweden, based on cosmogenic radionuclide data. Geografiska Annaler: Series A, Physical Geography, 84(3–4): 275-287.

Sugden, D. E., Balco, G., Sass, L. C., Cowdery, S. G. and Stone, J. O., 2005. Selective glacial erosion and weathering zones in the coastal mountains of Marie Byrd Land, Antarctica. Geomorphology, 67(3–4): 317-334.

About the Author

Keir Nichols

I am a glacial geologist with a focus on applying cosmogenic nuclides to study the glacial history of Antarctica. For my PhD (Tulane University, New Orleans) I used in situ 14C to shed light on the Last Glacial Maximum configuration of the Weddell Sea sector of the Antarctic Ice Sheets.

At present, I am very fortunate to be part of the International Thwaites Glacier Collaboration (ITGC), a US-UK project investigating practically all aspects of Thwaites Glacier. As a postdoc at Imperial College London, I am measuring cosmogenic nuclides in subglacial bedrock cores to study the short- and long-term history of the Florida-sized glacier which is, at present, the single largest contributor to global sea level rise. 

Email address:   

Twitter: @cosmokeir

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

This site uses cookies. Find out more about this site’s cookies.