Cryospheric Sciences

Ice-hot news

Ice-hot news: The cryosphere and the 1.5°C target

Ice-hot news: The cryosphere and the 1.5°C target

Every year again, the Conference of Parties takes place, an event where politicians and activists from all over the world meet for two weeks to discuss further actions concerning climate change. In the context the COP24, which started this Monday in Katowice (Poland), let’s revisit an important decision made three years ago, during the COP21 in Paris, and its consequences for the state of the cryosphere…

1.5°C target – what’s that again?

Last October, the International Panel on Climate Change (IPCC) released a special report (SR15) on the impacts of a 1.5°C global warming above pre-industrial levels. This target of 1.5°C warming was established during the 21st conference of the parties (COP21), in a document known as the Paris Agreement. In this Agreement, most countries in the World acknowledge that limiting global warming to 1.5°C warming rather than 2°C warming would significantly reduce the risks and impacts of climate change.

But wait, even though achieving this target is possible, which is not our subject today, what does it mean for our beloved cryosphere? And how does 1.5°C warming make a difference compared to the 2°C warming initially discussed during the COP21 and previous COPs?

A reason why the cryosphere is so difficult to grasp is the nonlinear behaviour of its components. What does this mean ? A good basic example is the transition between water and ice. At 99.9°C, you have water. Go down to 0.1°C and the water is colder, but this is still water. Then go down to -0.1°C and you end up with ice. The transition is very sharp and the system can be deeply affected even for a small change in temperature.

As a main conclusion, studies conducted in the context of SR15 show that, below 1.5°C of global warming, most components of the cryosphere will be slightly affected, while above that level of warming, there is more chance that the system may respond quickly to small temperature changes. In this Ice Hot News, we review the main conclusions of the SR15 concerning ice sheets, glaciers, sea ice and permafrost, answering among others the question if achieving the 1.5°C target would prevent us to trigger the potential nonlinear effects affecting some of them.

Ice sheets

The two only remaining ice sheets on Earth cover Greenland and Antarctica. If melted, the Greenland ice sheet could make the sea level rise by 7 m, while the Antarctic ice sheet could make it rise by almost 60 m. A recent review paper (Pattyn et al., 2018), not in SR15 because published very recently, shows that keeping the warming at 1.5°C rather than 2°C really makes the differences in terms of sea level rise contribution by the two ice sheets.

Greenland is a cold place, but not that cold. During the Holocene, the surface of the ice sheet always melted in summer but, in the yearly mean, the ice sheet was in equilibrium because summer melt was compensated by winter accumulation. Since the mid-1990s, Greenland’s atmosphere has warmed by about 5°C in winter and 2°C in summer. The ice sheet is thus currently losing mass from above and its surface lowers down. In the future, if the surface lowers too much, this could accelerate the mass loss because the limit altitude between snow and rainfalls may have been crossed, further accelerating the mass loss. The temperature threshold beyond which this process will occur is about 1.8°C, according to the Pattyn et al., 2018 paper.

Antarctica is a very cold continent, much colder than Greenland, but it has been losing mass since the 1990s as well. There, the source of the retreat is the temperature increase of the ocean. The ocean is in contact with the ice shelves, the seaward extensions of the ice sheet in its margins. The warmer ocean has eroded the ice shelves, making them thinner and less resistant to the ice flow coming from the interior. And if you have read the post about the marine ice sheet instability (MISI), you already know that the ice sheet can discharge a lot of ice to the ocean if the bedrock beneath the ice sheet is deeper inland than it is on the margins (called retrograde). MISI is a potential source of nonlinear acceleration of the ice sheet that, along with other nonlinear effects mentioned in the study, could trigger much larger sea level rise contribution from the Antarctic ice sheet above 2 to 2.7°C.

You can find complementary informations to the Pattyn et al., 2018 paper in SR15, sections 3.3.9,, and in FAQ 3.1.

Glaciers crossing the transantarctic mountains, one of them ending up to Drygalski ice tongue (left side) in the Ross sea. The ice tongue is an example of those ice shelves that form as grounded ice flows toward the sea from the interior. Ice shelves are weakened by a warmer ocean, which accelerates upstream ice flow [Credit: C. Ritz, PEV/PNRA]


Over the whole globe, the mass of glaciers has decreased since pre-industrial times in 1850, according to Marzeion et al., 2014. At that time, climate change was a mix between human impact and natural variability of climate. Glacier response times to change in climate are typically decades, which means that a change happening, for instance, today, still has consequences on glaciers tens of years after. Today, the retreat of glaciers is thus a mixed response to natural climate variability and current anthropogenic warming. However, since 1850, the anthropogenic warming contribution to the glacier mass loss has increased from a third to more than two third over the last two decades.

Similarly to the Greenland ice sheet, glaciers are prone to undergo an acceleration of ice mass loss wherever the limit altitude where rainfall occurs more often than snowfall is higher and at the same time the glacier surface lowers. However, as opposed to ice sheets, glaciers can be found all over the world under various latitudes, temperature and snow regimes, which makes it difficult to establish a unique temperature above which all the glaciers in the world will shrink faster in a nonlinear way. There are, however, model-based global estimates of ice mass loss over the next century. The paper from Marzeion et al., 2018, shows that under 1.5-2°C of global warming, the glaciers will lose the two thirds of their current mass, and that for a 1°C warming, our current level of warming since pre-industrial times, the glacier are still committed to lose one third of their current mass. This means the actions that we take now to limit climate change won’t be seen for decades.

You can find complementary informations in SR15, sections 3.3.9, and in FAQ 3.1.

Sea ice

As very prominently covered by media and our blog (see this post and this post), the Arctic sea-ice cover has been melting due to the increase in CO2 emissions in past decades. To understand the future evolution of climate, climate models are forced with the expected CO2 emissions for future scenarios. In summer, the results of these climate model simulations show that keeping the warming at 1.5°C instead of 2°C is essential for the Arctic sea-ice cover. While at 1.5°C warming, the Arctic Ocean will be ice-covered most of the time, at 2°C warming, there are much higher chances of a sea-ice free Arctic. In winter, however, the ice cover remains similar in both cases.

In the Antarctic, the situation is less clear. On average, there has been a slight expansion of the sea-ice cover (see this post). This is, however, not a clear trend, but is composed of different trends over the different Antarctic basins. For example, a strong decrease was observed near the Antarctic peninsula and an increase in the Amundsen Sea. The future remains even more uncertain because most climate models do not represent the Antarctic sea-ice cover well. Therefore, no robust prediction could be made for the future.

You can find all references were these results are from and more details in Section 3.3.8 of the SR15. Also, you can find the impact of sea-ice changes on society in Section

Caption: Sea ice in the Arctic Ocean [D. Olonscheck]


Permafrost is ground that is frozen consecutively for two years or more. It covers large areas of the Arctic and the Antarctic and is formed or degraded in response to surface temperatures. Every summer, above-zero temperatures thaw a thin layer at the surface, and below this, we find the boundary to the permafrost. The depth to the permafrost is in semi-equilibrium with the current climate.

The global area underlain by permafrost globally will decrease with warming, and the depth to the permafrost will increase. In a 1.5°C warmer world, permafrost extent is estimated to decrease by 21-37 % compared to today. This would, however, preserve 2 millions km2 more permafrost than in a 2°C warmer world, where 35-47 % of the current permafrost would be lost.

Permafrost stores twice as much carbon (C) as the atmosphere, and permafrost thaw with subsequent release of CO2 and CH4 thus represents a positive feedback mechanism to warming and a potential tipping point. However, according to estimates cited in the special report, the release at 1.5°C warming (0.08-0.16 Gt C per year) and at 2°C warming (0.12-0.25 Gt C per year) does not bring the system at risk of passing this tipping point before 2100. This is partly due to the energy it takes to thaw large amounts of ice and the soil as a medium for heat exchange, which results in a time lag of carbon release.
The response rates of carbon release is, however, a topic for continuous discussion, and the carbon loss to the atmosphere is irreversible, as permafrost carbon storage is a slow process, which has occurred over millennia.

Changes in albedo from increased tree growth in the tundra, which will affect the energy balance at the surface and thus ground temperature, is estimated to be gradual and not be linked to permafrost collapse as long as global warming is held under 2°C.

The above-mentioned estimates and predictions are from the IPCC special report Section, and

Slope failure of permafrost soil [Credit: NASA, Wikimedia Commons].

So, in summary…

In summary, what can we say? Although the 1.5°C and 2°C limits were chosen as a consensus between historical claims based on physics and a number that is easy to communicate (see this article), it seems that there are some thresholds for parts of the cryosphere exactly between the two limits. This can have consequences on longer term, e.g. sea-level rise or permanent permafrost loss. Additionally, as the cryosphere experts and lovers that we are here in the blog team, we would mourn the loss of these exceptional landscapes. We therefore strongly hope that the COP24 will bring more solution and cooperation for the future against strengthening of climate change!

Further reading

Edited by Clara Burgard and Violaine Coulon

Lionel Favier is a glaciologist and ice-sheet modeller, currently occupying a post-doctoral position at IGE in Grenoble, France. He’s also on twitter.




Laura Helene Rasmussen is a Danish permafrost scientist working at the Center for Permafrost, University of Copenhagen. She has spent many seasons in Greenland, working with the Greenland Ecosystem Monitoring Programme and is interested in Arctic soils as an ecosystem component, their climate sensitivity, functioning and simply understanding what goes on below.



Clara Burgard is a PhD student at the Max Planck Institute for Meteorology in Hamburg. She investigates the evolution of sea ice in general circulation models (GCMs). There are still biases in the sea-ice representation in GCMs as they tend to underestimate the observed sea-ice retreat. She tries to understand the reasons for these biases. She tweets as @climate_clara.


What’s on at POLAR18?

What’s on at POLAR18?

Next Tuesday (19th June) the POLAR18 Open Science Conference kicks off in Davos, Switzerland. We have put together a quick guide about events that might be of interest to you during the week! Conferences are about the science, of course, but the social side is just as important 🙂

What is POLAR18?

The eagle-eyed among you will have spotted that the POLAR18 conference is, in fact, a collection of different meetings held between the 15th-26th June, it’s quite confusing at first glance, so here is a summary of what is going on!

  • 15 – 18 June – SCAR and IASC/ASSW Business & Satellite Meetings (i.e. Side meetings and workshops) – details here.
  • 19 – 23 June  – SCAR/IASC Open Science Conference & Open COMNAP Session (i.e. the main event!)
    • Main program here – this will be the most important part for most of you!
    • Side meetings program here
  • 24 – 26 June – SCAR Delegates Meeting & 2018 Arctic Observing Summit – details here.


The conference and side meetings are held at the Congress Centre Davos which is in the middle of town (see map below). It is easy to walk around Davos, but if you want to use the local buses you get a free “Guest Card” bus ticket included with most hotel, hostel and apartment bookings.

Needless to say, Davos is a great place to be if you like biking, hiking, trail running and just generally being outside – for ideas on what to do, check out the Q&A section of the POLAR18 website.

Events for ECSs

There is a lot going on during the week – below we have listed just some of the social and networking events we think might be of particular interest to ECSs.

APECS World Summit – Sunday 17th and Monday 18th June

The Association for Polar Early Career Scientists (APECS) is excited to invite members and other early career professionals from around the globe to our 2nd APECS World Summit 2018! Hosted directly before POLAR2018 – the theme for this two-day event on 17-18 June will be “Connecting the Poles”. Please check out this link for more information and very important – YOU NEED TO REGISTER!

Southern Ocean Data Hack 2018 – Sunday 17th June, from 8:00 AM – 4:00 PM

Every wished someone had combined all the measurements of this or that for you into one handy dataset? Well….someone has! Pop into the Southern Ocean Data Hack on Sunday 17th June in Room B Strela to see these collected data sets and talk to the creators behind them. The workshop is supported by the NSF-funded SeaView project ( and the Southern Ocean Observing System (

Introduction to and use of the datasets will be on an informal, drop-in basis from 8am – 4pm. Contact: Steve Diggs ( or Pip Bricher ( ) if you want more info!

Celebrate the Arctic! – Monday, 18th June 2018, from 7:00 PM – 10:00 PM

This is a social networking event to highlight successes of the Arctic research community, organised by ARCUS on Monday 18th June (evening before the official start of the open science conference). It starts at 7pm in the Greenroom at the Hard Rock Hotel Davos. It is a free event with complimentary catering, door prizes, and a cash bar.


EGU Cryosphere ECS Team MeetupTuesday 19th June from 7:00PM

A relaxed social meet-up of the EGU Cryosphere ECS (early career scientist) team – that’s the folks that write this blog!

We are always looking for new members to get involved with the blog, our social media team and organising events and courses at the EGU General Assembly. So if you are interested in knowing more about the EGU Cryosphere Team come along to our meet-up to find out more 🙂

Please email Emma ( for details!

Queers + Allies Meetup – Friday, 22nd June 2018 at 18:30 PM

There will be a Queer/LGBT + Allies meetup at POLAR18 in the Rinerhorn/Strela room at the Congress Centre Davos (conference venue) on Friday, 22 June at 18:30 (after the poster session). The meeting is designed as a meet-up to discuss community goals and get to know people – after the meeting the evening will move to a social location downtown! 

Back to the Front – Larsen C Ice Shelf in the Aftermath of Iceberg A68!

Back to the Front – Larsen C Ice Shelf in the Aftermath of Iceberg A68!

Much of the Antarctic continent is fringed by ice shelves. An ice shelf is the floating extension of a terrestrial ice mass and, as such, is an important ‘middleman’ that regulates the delivery of ice from land into the ocean: for much of Antarctica, ice that passes from land into the sea does so via ice shelves. I’ve been conducting geophysical experiments on ice for over a decade, using mostly seismic and radar methods to determine the physical condition of ice and its wider system, but it’s only in the last couple of years that I’ve been using these methods on ice shelves. The importance of ice shelf processes is becoming more widely recognised in glaciological circles: after hearing one of my seminars last year, a glaciology professor told me that he was revising his previous opinion that ice shelves were largely ‘passengers’ in the grand scheme of things and this recognition is becoming more common. Slowly, we are coming to appreciate that ice shelves have their own specific dynamics and, moreover, that they are the drivers of change on other ice masses.

The MIDAS Project

In 2015, I joined the MIDAS project – led by Swansea and Aberystwyth Universities and funded by the Natural Environment Research Council – dedicated to investigating the effects of a warming climate on the Larsen C ice shelf in West Antarctica (Fig. 1). My role was to to assist with geophysical surveys (Fig. 2) on the ice shelf – but more about that later!

Figure 2: Adam Booth overseeing seismic surveys on the Larsen C ice
shelf in 2015 [Credit: Suzanne Bevan].

Larsen C is located towards the northern tip of the Antarctic Peninsula, and is one of a number of “Larsen neighbours” that fringe its eastern cost. MIDAS turns out to have been an extremely timely study, culminating in 2017 just as Larsen C hit the headlines by calving one of the largest icebergs – termed A68 – ever recorded. On 12th July 2017, 12% of the Larsen C area was sliced away by a sporadically-propagating rift through the eastern edge of the shelf, resulting in an iceberg with 5800 km2 area (two Luxembourgs, one Delaware, one-quarter Wales…). As of 14th October 2017 (Fig. 1), A68 is drifting into the Weddell Sea, with open ocean between it and Larsen C. See our previous post “Ice ice bergy” to find out more about how and why ice berg movement is monitored.

The aftermath of A68

As colossal as A68 (Fig, 1) is, its record-breaking statistics are only (hnnngh…) the tip of the iceberg, and of greater significance is the potential response of what remains of Larsen C. This potential is best appreciated by considering what happened to Larsen B, a northern neighbour of Larsen C. In early 2002, over 3000 km2 of Larsen B Ice Shelf underwent a catastrophic collapse, disintegrating into thousands of smaller icebergs (and immortalised in the music of the band British Sea Power). Rewind seven years further back, to 1995: Larsen B calved an enormous iceberg, exceeding 1700 m2 in area. An ominous extrapolation from this is that large iceberg calving somehow preconditions ice shelves to instability, and several models of Larsen C evolution suggest that it could follow Larsen B’s lead and become more vulnerable to collapse over the coming years.

The enormous mass of the intact ice shelf acts like a dam that blocks the delivery of terrestrial ice into the ocean, and the disappearance of the ice shelf removes so-called ‘backstress’ – essentially ‘breaking the dam’.

Then what? Well, ice shelves are in stress communication with their terrestrial tributaries, therefore processes affecting the shelf can propagate back to the supply glaciers. The enormous mass of the intact ice shelf acts like a dam that blocks the delivery of terrestrial ice into the ocean, and the disappearance of the ice shelf removes so-called ‘backstress’ – essentially ‘breaking the dam’. In the aftermath of Larsen B’s collapse, its tributary glaciers were seen to accelerate, thereby delivering more of their ice into the Weddell Sea. It is this aftermath that we are particularly concerned about, since it’s the accelerated tributaries that promote accelerated sea-level rise. Ice shelf collapse has little immediate impact on sea-level: since it is already floating, the shelf displaces all the water that it ever will. But, in moving more ice from the land to the sea, we risk increased sea levels and, with them, the associated socio-economic consequences.

How can we improve our predictions?

Figure 3: Computational model of the changed stress state, Δτuu, of Larsen C following the calving of A68 (output from BISICLES model, from Stephen Cornford, Swansea University). The stress change is keenly felt at the calving front, but also propagates further upstream [Credit: Stephen Cornford]

A key limitation in our ability to predict the evolution of Larsen C is a lack of observational evidence of how ice shelf stresses evolve in the short-term aftermath of a major calving event. These calving events are rare: we simply haven’t had much opportunity to investigate them, so while our computer predictions are based on valid physics (e.g., Fig. 3) it would be valuable to have actual observations to constrain them. Powerful satellite methods are available for tracking the behaviour of the shelf but these provide only the surface response; Larsen C is around 200 m thick at its calving front so there is plenty of ice that is hidden away from the satellite ‘eye in the sky’, but that is still adapting to the new stress regime. So how can we “see” into the ice?

To address this, we’ve recently been awarded an “Urgency Grant” – Response to the A68 Calving Event (RA68CE) – from NERC to send a fieldcrew to the Larsen C ice shelf, involving researchers from Leeds, Swansea and Aberystwyth, together with the British Geological and British Antarctic Surveys.

Figure 4: Emma Pearce and Dr Jim White preparing seismic equipment – intrepid geophysicists ready to wrap-up warm for field deployment on Larsen C! [Credit: Adam Booth]

The field team – Jim White and Emma Pearce (Fig. 4) – will undertake seismic and radar surveys at two main sites (Fig. 3) to assess the new stress regime around the Larsen C calving front. One of these sites is being reoccupied after seismic surveying in 2008-9, during the Swansea-led SOLIS project, allowing us to make a long-term comparison. These, and two other sites, will also be instrumented with EMLID REACH GPS sensors, to track small-scale ice movements than can’t be captured in the satellite data. The field observations will be supplied to a team of glacial modellers at Swansea University, to allow them to improve future predictions (e.g. Fig. 3), while their remote sensing team continues to monitor the evolving stress state at surface.

It’s truly exciting to be coordinating the first deployment, post A68, on Larsen C. Our data should provide a unique missing piece from the predictive jigsaw of Larsen C’s evolution, ultimately improving our understanding of the causes and effects of large-scale iceberg calving – both for Larsen C and beyond!


For ice-hot news from the field, follow Emma Pearce on twitter: @emm_pearce


Edited by Emma Smith

Further Reading

  • More information on Larsen C at the project MIDAS website
  • Learn more about ice shelf evolution with the Ice Flows game – eduction by stealth! Also check out the EGU Cryoblog post about it!
  • Borstad et al., 2017; Fracture propagation and stability of ice shelves governed by ice shelf heterogeneity; Geophysical Research Letters, 44, 4186-4194.
  • Wuite et al., 2015; Evolution of surface velocities and ice discharge of Larsen B outlet glaciers from 1995 to 2013. The Cryosphere, 9, 957-969.
  • Cornford et al., 2013; Adaptive mesh, finite volume modelling of marine ice sheets; Journal of Computational Physics, 232, 1, 529-549.

Adam Booth is a lecturer in Exploration Geophysics at the University of Leeds, UK. He is the PI on the NERC-funded project “Ice shelf response to large iceberg calving” (NE/R012334/1). After obtaining his PhD from the University of Leeds in 2008, he held postdoctoral positions at Swansea University and Imperial College London, in which he worked with diverse research applications of near-surface geophysics. He tweets as: @Geophysics_Adam

Ice-Hot News : The “Oldest Ice” quest has begun

Ice-Hot News : The “Oldest Ice” quest has begun

This is it! The new European horizon 2020 project on Oldest Ice has been launched and the teams are already heading out to the field, but what does “Old Ice” really mean? Where can we find it and why should we even care? This is what we (Marie, Olivier and Brice) will try to explain somewhat.

Why do we care about old ice, ice cores and past climate?

Figure 1: Drilling an ice core [Credit: Brice Van Liefferinge]

Figure 1: Drilling an ice core [Credit: Brice Van Liefferinge]

Unravelling past climate and how it responded to changes in environmental conditions (e.g. radiative forcing) is crucial for our understanding of the current climate and for predicting how climate will likely change in the future.

Ice cores contain unique and quantitative information on the past climate (e.g. atmospheric gas concentration). The caveat is that at the moment, we can “only” go back up to 800,000 years at EPICA Dome C ice core (Parrenin et al, 2007).

Nonetheless, marine records tell us that during the Mid-Pleistocene there was a major climate transition (0.8-1.2 million years ago): a change in the frequency of glacial-interglacial cycles in the Northern Hemisphere. Instead of an ice age every 40,000 year, the climate changed to what is termed a “100,000 year world”. Unfortunately, the time resolution of marine records are too coarse to provide details on the mechanisms behind such climate changes. We must therefore rely on ice cores to obtain a high enough temporal resolution. Furthermore, the ice traps air bubbles and can therefore provide a record of the atmospheric composition that can be used to directly measure the paleo atmosphere through the transition.

The new European project ‘Oldest ice’ was set up for this very objective: crack the Mid-Pleistocene Transition climate. It brings together engineers, experimentalists and modellers from 14 Universities around the world.

In this post, we will focus on the first mission of the project: locating areas with million year old ice in Antarctica. The next steps will be to:

  • develop the drilling technology,

  • improve our geophysical knowledge of the identified site,

  • and finally, reach the “holy grail”: recover ice from the very base of the ice sheet with a target age of 1.5 Million years.

The whole project is anticipated to last 10 years!

The new European project ‘Oldest ice’ was set up for this very objective: crack the Mid-Pleistocene Transition climate

The first mission: “Where to find million year old ice?”

Oldest Ice (ice more than 1 mio. years old) can only be recovered in Antarctica, but where exactly? This question has to be answered in a two-step approach:

  1. On a large scale, we must first narrow down places in Antarctica where Oldest Ice might be found. To do that, we rely on models.

  2. Then, we can focus our analysis on those regions by gathering field data in the form of airborne radar surveys. Further ground-based work is currently taking place.

On a larger scale, Oldest Ice in Antarctica requires:

  1. Thick ice and cold bed. We need thick ice to reconstruct past climate variations with sufficient temporal resolution (e.g. is there enough ice to measure air bubbles or other climate markers). However, the thicker the ice, the higher the basal temperature. If the bottom of the ice is too warm, the ice at the base will start to melt, potentially destroying the Oldest Ice of the ice sheet.
    Finding a suitable drill site hence requires a good trade-off between thickness and cold-bed conditions.

  2. Slow-moving ice. This is found mainly at the centre of the ice sheet. Imagine this: if ice were to flow at as little as 1 m per year over a period of 1.5 Million years, it would have travelled 1,500 km over that time interval! However, there is a catch: slow-moving areas are also low-accumulation areas, and low accumulation means warmer ice. This is because the ice is cooled by the addition of cold snow at the surface that then gets transformed to ice and then travels downwards. Indeed, the greater the accumulation, the deeper the “cold snow” can penetrate into the ice sheet!

  3. Undisturbed ice. In order to obtain an interpretable climate record, the ice recovered from the drill needs to be stratigraphically ordered, i.e. no mixing of the ice can have occurred so that we can assume that time increases with depth when we measure ice composition down the core. Variations in the height of the bedrock can induce such ice mixing.

(more information can be found in Van Liefferinge and Pattyn (2013))

Figure 2. Potential locations of cold bed (basal temperatures 2000 m), slow motion (horizontal flow speeds <2m/yr) The colour bar represents the basal temperature. The two insets focus on Dome C and Dome F, two areas highly likely to store million year old ice. [Credit: Brice Van Lieffering, updated from Van Liefferinge, B. and Pattyn, 2013]

Figure 2. Potential locations of cold bed (basal temperatures 2000 m), slow motion (horizontal flow speeds <2m/yr) The colour bar represents the basal temperature. The two insets focus on Dome C and Dome F, two areas highly likely to store million year old ice. [Credit: Brice Van Lieffering, updated from Van Liefferinge, B. and Pattyn, 2013]

While boundary conditions such as ice thickness and accumulation rates are relatively well constrained, the major uncertainty remains in determining thermal conditions at the ice base. The thermal conditions depend on the geothermal heat flow (the flux of “energy” provided by the Earth which conducts heat into the crust) underneath the ice sheet. But to measure the geothermal heat flow, you need to reach the bed.

We need to find the ideal drilling location which would satisfy all these conditions – a bit of a “Goldilocks’ choice”: thick ice but not too much, low accumulation but not too low, low geothermal heat flow but high enough to not get folded basal ice. To do this we use several models: a simple one which calculates the minimum geothermal heat flow needed to reach the pressure melting point that we can then compare to data sets, and a more complex one resolving in three dimensions the temperature field with thermomechanical coupling (i.e. linking the ice-flow component to the heat-flow component). The combination of modelling approaches shows that the most likely oldest ice sites are situated near the ice divide areas (close to existing deep drilling sites, but in areas of smaller ice thickness) (see Figure 2).

Give it a go: Try to find million year old ice yourself using this Matlab© tool:

The combination of modelling approaches shows that the most likely oldest ice sites are situated near the ice divide areas

On finer scales: we need deep radiostratigraphy and age modelling

Radar profiles

Figure 3. Radargram from the new OIA radar survey (Young et al., in review) with isochrones interpreted in red [Credit: Marie Cavitte]

Figure 3. Radargram from the new Oldest Ice A radar survey (Young et al., in review) with isochrones interpreted in red [Credit: Marie Cavitte]

Radargrams (see figure 3) are powerful tools to observe the internal structure of the ice: variations in density, acidity and ice fabric all can create conductivity contrasts, which result in radar visual stratigraphy. Below the firn column (the compacting snow, up to 100 m thick), most returns are related to acidity variations, corresponding to successive depositional events (i.e. snowfall). Radar stratigraphy in this case can be considered isochronal, i.e. every visible line (see figure 3) were formed at the same moment, (Siegert et al., 1999). Such radar isochrones can then be traced for kilometres throughout the ice sheet where radar data has been acquired. When radar lines intersect an ice core site, the radar stratigraphy can then be dated by matching the isochrone-depths to the ice core depths at the site and then transferring the age-depth timescale.

This allows to date entire sub-regions. However, the very bottom of the ice column is often difficult to interpret: radar isochrones cannot always be continuously followed from the ice core.

Radargrams are powerful tools to observe the internal structure of the ice

The newly acquired Oldest Ice A radar survey (Young et al., in review) over the Dome C region (see figure 2 for location) gives very rich stratigraphic information and the proximity of the EPICA Dome C ice core has allowed the dating of the isochrones. The ice sheet in this area could only be dated to ~360,000 years (Cavitte et al., 2016) and not further back in time because deeper isochrones are tricky to tie to the ice core, and other times, there is no clear signal (deep scattering ice, visible near the bedrock, at the bottom of Figure 3). As such, we need an age model to try to describe the age-depth relation below the deepest dated isochrones.

Modelling the ice

Figure 4. More precise analysis of the Dome C Oldest Ice target, with the lines representing the Oldest Ice A airborne survey collected in winter 2015/16 (Young et al., in review). The colours represent the modelled age of the ice 60 meters above the bedrock, in thousands of years. We can see that this whole region has a lot of potential for recovering million year old ice. [Credit: Olivier Passalacqua]

Figure 4. More precise analysis of the Dome C Oldest Ice target, with the lines representing the Oldest Ice A airborne survey collected in winter 2015/16 (Young et al., in review). The colours represent the modelled age of the ice 60 meters above the bedrock, in thousands of years. We can see that this whole region has a lot of potential for recovering million year old ice. [Credit: Olivier Passalacqua]

The age of the ice primarily depends on its vertical velocity, so we can use a simple 1D model to describe the motion of the ice in the vertical direction. We run the model for an ensemble of vertical velocity profiles and basal melt rates, and consider the distribution of the basal ages (i.e. model ages) given by the profiles that reproduce the observations the best (i.e. isochrones ages).

We need an age model to try to describe the age-depth relation below the deepest dated isochrones

After running the model, it appears that many areas of the Oldest Ice A survey region host very old ice (see red and yellow dots on figure 4 which represent ages > 1 million years). A high enough bottom age gradient, provided by the dated isochrones, is required to ensure sufficiently old ice as a drilling target. Following initial calculations, it will probably be a better choice to drill on the flank of the bedrock relief than on its top.

So in the end, where do we find the oldest ice?

We have to find areas which provide a good compromise between thick ice (for the a good temporal resolution in the ice core) but not too thick (to avoid basal melting). The best sites will be the ones close to the surface ridge (to ensure limited displacement of the ice), standing above the surrounding subglacial lakes, and for which a lot of undated isochrones below the last dated isochrone are visible.

To find out more about Beyond EPICA and keep track of progress visit the project  website and follow @OldestIce on twitter!

Edited by Sophie Berger

Brice Van Liefferinge is a PhD student and a teaching assistant at the Laboratoire de Glaciology, Université libre de Bruxelles, Belgium. His research focuses on the basal conditions of the Antarctic ice sheet. He tweets as @bvlieffe.

Marie Cavitte is a PhD student at the Institute for Geophysics at the University of Texas at Austin, Texas. Her research focuses on understanding radar internal stratigraphy and using it as a means to constrain the temporal stability of the East Antarctic Ice Sheet interior.

Olivier Passalacqua is a PhD student at the Laboratoire de Glaciologie et Géophysique de l’Environnement, Grenoble, France.

Members of the consortium

  • Alfred Wegener Institute, Helmholtz Centre for Polar and Marine Research (AWI, Germany), Coordination
  • Institut Polaire Français Paul Émile Victor (IPEV, France)
  • Agenzia nazionale per le nuove tecnologie, l’energia e lo sviluppo economico sostenibile (ENEA, Italy
  • Centre National de la Recherche Scientifique (CNRS, France)
  • Natural Environment Research Council – British Antarctic Survey (NERC-BAS, Great Britain)
  • Universiteit Utrecht – Institute for Marine and Atmospheric Research (UU-IMAU, Netherlands)
  • Norwegian Polar Institute (NPI, Norway)
  • Stockholms Universitet (SU, Sweden)
  • Universität Bern (UBERN, Switzerland)
  • Università di Bologna (UNIBO, Italy)
  • University of Cambridge (UCAM, Great Britain)
  • Kobenhavns Universitet (UCPH, Denmark)
  • Université Libre de Bruxelles (ULB, Belgium)
  • Lunds Universitet (ULUND, Sweden)

Non-Europan partners

  • Institute for Geophysics, University of Texas at Austin (UTIG, US)
  • Australian Antarctic Division (AAD, Australia)