Does the Antarctic Ice Sheet transport glacial erratics?

Dear Malcolm,

Thank you for your question on glacial erratics. Below is a summary of the main debris transport pathways in glacial systems.

Debris can be supplied to a glacier system from supraglacial or subglacial sources. Supraglacial sources include rock or snow avalanches from valley sides or mountain peaks that rise above an ice sheet/cap surface (these are known as nunataks). Debris of a smaller grain-size (e.g. volcanic ash, dust) can also be added to glacier surfaces from windfall. Subglacial debris can either be derived from erosion at the glacier bed, or where supraglacial debris is lowered through a glacier (e.g. via crevasses or moulins).

Whether debris is supplied at the glacier surface (supraglacial) or at the bed (subglacial) goes some way in determining its transport pathway. Debris transported on a glacier surface is referred to as high-level debris transport. This material is also commonly termed passively transported debris because there is little modification of rock debris transported on the glacier surface. However, debris that falls on the glacier surface does not always remain there. It may descend to an englacial or subglacial position due to burial within primary ice stratification in the accumulation zone, or it may descend through crevasses or moulins.

By contrast, debris transported at the glacier bed is known as low-level debris transport. This type of debris is often termed actively transported because it is modified (e.g. rock edges are rounded off) during the process of transport. At the bed, debris can be transported via traction, within subglacial meltwater streams, or within the basal ice itself. Once entrained in basal ice debris may be transferred to different levels in a glacier. The folding and thrusting of ice as it deforms (e.g. where squeezed between narrow valley walls) can be enough to elevate basal debris into englacial, or even supraglacial, positions. Similarly, basal debris often emerges at the glacier surface close to the snout, due to compressive ice flow (put simply, slower flowing ice at the margin may be overridden by faster flowing ice immediately up-glacier, bringing with it basal debris).

To return to your question about Antarctic erratics, like all glaciers, debris is transported at the surface, within, and at the bed of the Antarctic Ice Sheet. An interesting point to note is that there is very little terrain above the ice sheet surface in Antarctica, which almost completely submerges the underlying topography (including mountain chains). This provides a natural limit on the amount of debris supplied to the ice sheet surface. However, surface debris does occur, and erratic rocks have been identified from past advances of the Antarctic Ice Sheet, such as those on James Ross Island, and they provide a valuable means of reconstructing ice sheet history through cosmogenic isotope dating techniques.

I hope this answer has been useful in outlining the main transport pathways in Antarctic glaciers, and glacial systems more generally. Do not hesitate to get in touch with more questions!


How much of the Antarctic Ice Sheet is below sea level?

Hi Luke, thanks for your question!

An answer to your question can be found in the recent BEDMAP2 (an ice bed, surface, and thickness dataset for the Antarctic Ice Sheet) paper.

In terms of area:

5.50 x 10^6 km^2 (or 5,500,000 km^2) of ice is grounded below sea level.

The total area of the ice sheet is 12.295 x 10^6 km^2 (or 12,295,000 km^2).

Therefore, ~45% of the ice sheet is grounded below sea level.

What is the source of Larsen C Ice Shelf?

Looking at maps of Larsen C, the area of the shelf is many time larger than the area of the glaciers that I assume would accumulate snow and feed it.  This made me wonder about the source of the Larsen Ice Shelf. Is most of Larsen C’s ice mass originally from the glaciers or was it accumulated in place as snow falls on the shelf itself?

Asked by Todd

Hi Todd,

Great question! The ice shelf receives mass from the glaciers on land that flow into it, and from snowfall directly on to the ice shelf, and from sea water freezing on to its base. It loses mass by submarine melting, sublimation and calving icebergs.

Is the breakaway of the Larsen C Iceberg normal?

my question:
Fact: A massive iceberg weighing more than one trillion tons has broken away from western Antarctica
Speculation: Global Warming

My Question: In 1911 – the Titanic sank due to icebergs in the shipping lanes – was this Global Warming? Probably not?
Speculation: It seems icebergs break away are normal


Asked by Greg,


Hi Greg,

You are completely right that iceberg calving is a completely normal process in Antarctica. Antarctica loses about 1089 Gt of ice per year through calving icebergs – exceeded only by basal melt.

This iceberg is unusually large, but we cannot attribute its calving event directly to climate change. We could only do this if we saw a sustained increase or change in calving behaviour over a long period of time, and were able to exclude dynamic factors that can cause changes in calving behaviour.

We are concerned that the calving of the iceberg might destablise Larsen C Ice Shelf and make it more vulnerable however – more information here.

How do we know how old the Antarctic Ice Sheet is?

The oldest penetrated Antarctic ice is about 800,000 years old. However, I have read that the Antarctic Ice Sheet has been present for several millions of years. Is there any direct evidence for older ice, or is it simply an inference from dC18 values from somewhere, or something like that? Do global sea-level curves support the older ice story?

-Asked by Joe

Hi Joe,

That is a great question. In fact, there are several ways in which we know when ice first began to accumulate on the Antarctic continent. Firstly, offshore seismic surveys and drilling campaigns of sediments on the continental slope and shelf. These sediments are dated using biostratigraphical methods (such as dinoflagellate cysts), strontium isotopes analysis on pristine shells, and by isotopic analysis of volcanic rocks (see Davies et al., 2012). Some of this evidence comes from the Antarctic Peninsula, summarised here.

This is supported by the deep-sea record of del18O, which is a proxy for global ice volume, and global eustatic sea level change. Together, these indices suggest that glaciation began on Antarctica around 35 million years ago.


Davies, B.J., Hambrey, M.J., Smellie, J.L., Carrivick, J.L., Glasser, N.F., 2012. Antarctic Peninsula Ice Sheet evolution during the Cenozoic Era. Quaternary Science Reviews 31, 30-66.

What do I need to study to become a glaciologist?

Asked by Johnny

Hi Johnny,

Thanks for your question! Glaciology encompasses a number of disciplines, from maths, physics and chemistry, geology and geography. A science background is necessary but there is a huge range of specialities within glaciology, ranging from numerical modelling (needs maths and physics and computer science) to geomorphological mapping (needs Geography or Geology). A degree in a science subject, followed by a Masters and most likely a PhD is the route into becoming a professional glaciologist.

Do glaciers recede during summer and advance during winter?

Is there a significant cyclic variation in the position of the glacier terminus along the year?

-Asked by Felix

Hi Felix.

This is a great question. In some areas with glaciers with especially fast response times, this may indeed be the case. Some glaciers in places like Scandinavia and Iceland have small annual moraines, which form each year, as the glacier recedes in the summer and the terminus position stabilises in the winter. However, this is not the case for most glaciers, as the response time of the glacier is typically longer than a single year.

What is iceberg grounding?

Icebergs ‘ground’ on the ocean floor when their keel is deeper than the water depth. They may become stuck until high tide floats them off, or if they are washed into a shallow area, until they melt sufficiently to float away.

How are ice cores dated?

I was wondering how ice cores are dated accurately. I know Carbon 14 is one method, but some ice cores go back hundreds of thousands of years. Would other isotopes with longer half-lives be more accurate?

Also, how much does it cost to date the core? How are samples acquired without destroying the ice? I imagine keeping the ice intact as much as possible would be extremely valuable.

Some of the answers to these questions are available on the Ice Core Basics page.

Ice cores can be dated using counting of annual layers in their uppermost layers.  Dating the ice becomes harder with depth. Other ways of dating ice cores include geochemisty, wiggle matching of ice core records to insolation time series (Lemieux-Dudon et al. 2010), layers of volcanic ash (tephra) (Vinther et al., 2006), electrical conductivity, and using numerical flow models to understand age-depth relationships (Mulvaney et al., 2012), combined with firn densification modeling to estimate the delta-age (Lemieux-Dudon et al. 2010). Usually multiple methods are used to improve accuracy.

Common global stratigraphic markers are palaeo-events that occur worldwide synchronously, and can allow wiggle-matching between ice cores and other palaeo archives (e.g., marine sediment cores). For the ice matrix, these global stratigraphic markers can include spikes in volcanic ash (each volcanic eruption has a unique chemical signature), or volcanic sulfate spikes. For the gas phase, methane, and oxygen-18 isotopic ratio of O2 have been used (Lemieux-Dudon et al. 2010).

Uranium has been used to date the Dome C ice core from Antarctica. Dust is present in ice cores, and it contains Uranium. The decay of 238U to 234U from dust in the ice matrix can be used to provide an additional core chronology. Beryillium-10 has also been used to date ice cores.

Ice cores are expensive to collect, house and keep. They must be stored continuously at a specific temperature. The American National Ice Core Laboratory provides some information on how they store and keep ice cores.

When ice cores are analysed, they may be cut or sectioned, with half the sample remaining as an archive. As the ice must be melted for analysis, the sample is usually destroyed during analysis.

OLYMPUS DIGITAL CAMERATypical CPL cut plan for a large multi-investigator ice coring project such as the WAIS Divide Ice Core project. —Credit: NICL-Science Management Office



Lemieux-Dudon B, et al. Consistent dating for Antarctic and Greenland ice cores. Quaternary Science Reviews 29, 8-20 (2010).

Mulvaney R, et al. Recent Antarctic Peninsula warming relative to Holocene climate and ice-shelf history. Nature 489, 141-144 (2012).

Vinther BM, et al. A synchronized dating of three Greenland ice cores throughout the Holocene. Journal of Geophysical Research: Atmospheres 111, n/a-n/a (2006)