CR
Cryospheric Sciences

Image of the Week

Climate Change & Cryosphere – Why is the Arctic sea-ice cover retreating?

Climate Change & Cryosphere – Why is the Arctic sea-ice cover retreating?

The Arctic Ocean surface is darkening as its sea-ice cover is shrinking. The exact processes driving the ongoing sea-ice loss are far from being totally understood. In this post, we will investigate the different causes of the recent retreat of the Arctic sea-ice cover, using the most updated literature…


Arctic sea ice is disappearing

Due to its geographical position centered around the North Pole, the Arctic Ocean is relatively cold compared to other world oceans. This means that, each winter, ocean temperatures fall below the freezing point, and sea ice forms on top of the ocean surface.

The Arctic sea-ice extent reaches about 15 million km2 in March, at its maximum (see left panel in Fig. 1 and Fig. 2). In spring, the ice starts to melt and reaches its minimum extent in September, which is about three times smaller than its maximum extent (see right panel in Fig. 1 and Fig. 2).

Figure 1: Maps of mean Arctic sea-ice concentration (percentage of sea ice in a given grid cell) in March (left) and September (right), averaged over 1979-2015, from satellite observations. The red line (right panel) shows the sea-ice edge in September 2012 (record minimum) [Credit: Ocean and Sea Ice Satellite Application Facility (OSI SAF)].

Satellite observations clearly show that the Arctic sea-ice cover has been shrinking since the beginning of the satellite record in 1979 (see this post and this post for more information about sea-ice satellite observations). The sea-ice loss is about 13% per decade in September and 3% per decade in March (see this post, this post and this post for further information on recent Arctic sea-ice changes). A recent study using data from a series of different observations (ship reports, airplane surveys, analyses by national services, etc.) shows that the recent Arctic sea-ice loss, as measured by satellites, is unprecedented as far back as 1850.

Figure 2: Seasonal cycle of Arctic sea-ice extent from satellite observations. The solid blue and dashed red lines show the 2019 (ongoing) and 2012 (record minimum) values. The dark gray curve shows the average over the period 1981-2010 with the corresponding uncertainty range in light gray (+/- 2 standard deviations) [Credit: National Snow and Ice Data Center].

The year 2012 was particularly exceptional in the sense that it featured the record minimum in September since the beginning of satellite measurements (dashed red curve in Fig. 2). 2019 was on a ‘good path’ to break this record, but the sea-ice loss rate started to lower from mid-August (blue curve in Fig. 2, see also here).

 

What are the drivers of the Arctic sea-ice loss?

The recent changes in Arctic sea ice have been caused by three main factors:

  1. External forcing: the variability caused by external factors, which can either be human (e.g. anthropogenic greenhouse gas emissions) or natural (e.g. changes in solar activity, volcanic eruptions).
  2. Internal variability: the variability caused by the chaotic nature of processes at work in the climate system. It is internal variability that prevents us to make accurate weather forecasts beyond a few days.
  3. Positive feedbacks: the processes by which a change in the climate can amplify, e.g. the ice-albedo feedback. These feedbacks are described in more detail in this post.

 

External forcing

Several studies have analyzed the links between the changes in external forcing and the recent changes in Arctic sea ice in both observations and models. It has been found that the anthropogenic global warming, caused by increased greenhouse gas concentration in the atmosphere, is the main driver of the long-term sea-ice loss in the Arctic. In particular, Notz and Stroeve (2016) found that for each ton of CO2 released into the atmosphere, the Arctic loses about 3 m2 of sea ice in September, as shown in Fig. 3 below (see this post).

Figure 3: September Arctic sea-ice area against cumulative CO2 emissions since 1850 for the period 1953-2015. Grey circles and diamonds show the results from in-situ (1953-1978) and satellite (1979-2015) observations, respectively. The thick red line shows the 30-year running mean and the dotted red line represents the trend of 3 m2 in sea-ice area loss per ton of CO2 emitted [Credit: D. Notz, based on Notz & Stroeve (2016)].

Other changes in external forcing have had a more limited impact on the recent changes in Arctic sea ice. For example, the volcanic eruptions of El Chichón in 1982 and Mount Pinatubo in 1991 caused a small increase in Arctic sea-ice extent (see this study), but their impact cannot be clearly identified in individual climate models (see here). Similarly, the impact of the solar activity on the recent Arctic sea-ice changes is very small.

 

Internal variability

The sea-ice evolution is also strongly subject to internal variability. A good explanation of the concept of internal variability can be found here and in this study.

Internal variability is often estimated using climate models. Running the same model with exactly the same parameters and external forcing, but with slightly different initial conditions (for example a different sea surface temperature), is a common method to get an idea of the internal variability of the climate system. Figure 4 below shows the evolution of the Arctic sea-ice extent using the Community Earth System Model (CESM) run 40 times with different initial conditions. The spread of the ensemble represents the range of the effect of internal variability.

Figure 4: Evolution of Arctic September sea-ice extent using the Community Earth System Model Large Ensemble (CESM-LE) in the less optimistic scenario (RCP8.5). The blue curves show all the 40 model members. The red curve shows the NSIDC satellite observations [Credit: Fig. 1a of Jahn et al. (2016)].

Variations in heat transport from the Atlantic Ocean due to internal variability have caused strong reductions in sea-ice area in the Barents Sea (see this study and this study), and probably other seas located in the Atlantic sector of the Arctic Ocean (as shown here). Changes in large-scale atmosphere circulation, also associated with internal variability, have contributed to sea-ice reductions as well (see this study).

However, several studies found that internal variability was not the key cause of the recent Arctic sea-ice loss over the past 40 years (e.g. this study). Instead, internal variability acts as an amplifier of the external forcing (see this study), so it only explains a small part of the recent Arctic changes. In the climate models used in the IPCC AR5 report, the impact of internal variability is of maximum 1 million km2 (see this study).

 

Feedbacks

A last cause for the recent changes in Arctic sea ice is the presence of positive feedbacks, which can amplify ongoing changes. One of the main feedbacks acting in the Arctic is the ice-albedo feedback (see this post and this study). Since ice reflects more sunlight than water, if the sea-ice cover decreases, more heat is trapped by the surface of the Arctic Ocean, leading to more ice melting.

However, as we have seen above, there is a clear linear relationship between Arctic sea-ice extent and cumulative CO2 emissions (Fig. 3). If the ice-albedo feedback was important in explaining the recent loss of Arctic sea ice, this linear relationship would break after years of strong or weak ice loss, which is not the case.

In fact, the positive feedbacks (like the ice-albedo feedback) are partly compensated by negative feedbacks that stabilize the climate system (see this previous post for a description of feedbacks in polar regions). Thus, while these feedbacks play a key role in the short term, they cannot explain the bulk part of the sea-ice loss since 1979.

 

The future

In conclusion, the recent loss of Arctic sea ice is strongly linked to anthropogenic global warming, although the changes in atmosphere circulation and ocean heat transport, associated with internal variability, also influence the sea-ice evolution. Research continues on the topic in order to capture the exact contribution of the different causes to the Arctic sea-ice loss.

On the long term, Arctic sea ice will continue disappearing. Based on the current emission rates of greenhouse gas emissions into the atmosphere, it is probable that the Arctic Ocean will be ice free during summer before 2050 (see this post, this study and this study).

Due to its linear relationship with CO2 emissions, the Arctic sea-ice cover is a strong indicator for the pace of current climate change. Its rapid disappearance should be seen as a warning light for other impacts to come…

 

Further reading

Edited by Clara Burgard


David Docquier is a post-doctoral researcher at the Earth and Life Institute of Université catholique de Louvain (UCL) in Belgium. He works on the development of processed-based sea-ice metrics in order to improve the evaluation of global climate models (GCMs). His study is embedded within the EU Horizon 2020 PRIMAVERA project, which aims at developing a new generation of high-resolution GCMs to better represent the climate.

Image of the Week – Looking to the past for answers

Figure 1 – The lateral moraines of the Khumbu Glacier, Nepal (A+B). Taken from the true right of the glacier and the confluence with Changri Nup/Shar. A shows the original photo; B shows the annotation highlighting different moraines. Numbers assigned based on distance from glacier tongue. Dots represent where rock samples were collected from moraine crests. Yellow circle highlights walkers for scale. [Credit: Martin Kirkbride (photo), mapping and sample collection completed by Jo Hornsey]

We’re only just really starting to comprehend the state and fate of Himalayan glaciers due to a scarcity of research along the monumental mountain range. Climbers and scientists have been observing these lofty glaciers since the 1900s. However, is that looking back far enough? Glacier moraines, featuring in this Image of the Week, can reveal change extending back thousands of years.


You may look at Figure 1 and think ‘what is that?! It’s a mess!’ and you would be right to do so. The only glacier ice visible is where ice cliffs break the debris-covered surface of Khumbu Glacier (Figure 2), which begins in the Western Cwm. If you let your eyes adjust to the medley of rocks and many shades of brown, you can start to pick out lines and shapes. Some are highlighted by the sunlight whilst others take a more discerning eye. If your eyesight is very good, you can spot the people on a path in the lower right area (highlighted by the yellow circle), which give a sense of scale to this landscape. These huge mounds of rock and debris (called moraines), though appearing messy and interwoven, are vital pieces of evidence which show how much the glacier is shrinking; extending thousands of years back beyond the satellite record or human observations.

But why the mess?

The young Himalayan fold mountains produce huge amounts of debris due to the extreme weather and ongoing orogeny. The summer monsoon also provides significant amounts of intense precipitation, which erodes slopes and sediment, causing a highly mobile landscape, and a continual cacophonous supply of rocks and sediment to the Khumbu Glacier (Figure 2); creating a surface blanket known as debris cover. This debris alters how the glacier would normally melt and results in the surface of the glacier lowering through time, rather than the terminus retreating as so often seen on ‘clean’ ice glaciers. Though data collection is improving constantly, access to the Khumbu (nearly a week’s trek with several days of altitude acclimatisation) limits the range of monitoring techniques available and reduced oxygen controls your ability to collect data. Whilst there is observational data, such as satellite imagery and observations from explorers/scientists during the 1900s, it is limited in temporal and spatial resolution. This is where my research on glacier moraines comes in.

A longer time frame

I have spent the last year and ¾ (I am a PhD student; I am counting every second!) mapping the landforms which these great bodies of ice leave behind. This mainly consists of mapping lateral moraines (Figures 1 and2) as these represent the height of the glacier surface at the time it built the moraines and can be used to reconstruct a patterns of glacier evolution. These landforms have differing patterns (size, shape, preservation etc) between glacial valleys, sometimes even within the same valley, telling us that there are local and regional differences in glacier behaviour. To uncover when this all happened, I did what every person thinks I do when I tell them my PhD is based in geography; I went to collect rock samples from the glacier moraines (Figures 1 and 3).

Figure 2 – Mapped moraines. The moraines are identified by the different coloured lines (higher numbers represent older moraines). The dots represent the areas where rocks were sampled for dating. The contemporary glacier outlines were taken from the Randolf Glacier Inventory GLIMS data set. The Digital Elevation Model was taken from the High Mountain Asia 8m resolution data set (Shean, 2017). [Credit: Mapping of moraines and sample collection completed by Jo Hornsey].

Rocks can tell the time?

Well, no; they can’t. But using a technique known as Exposure Dating, we can assign an age to a rock surface if we’re sure that surface has been in that position on that landform since it was put there by the glacier. This means, if we choose rocks on the crests of the moraines (similar to the one I am stood on in Figure 3), we can interpret the age we get from that rock as the time that the glacier surface was there building that landform. If we do it for the mapped moraines of the Khumbu Glacier (Figure 2), then we can start to build a timeline of glacier recession. Thanks to studies in the 1980’s  and to a couple done in the early 2000’s (e.g. Richards, et al., 2000; Finkel, et al., 2003), we’re pretty confident that the most glacier proximal landforms were built around 500 years ago during a hemispheric wide cooling event. This event was significant enough to build the towering lateral moraines which are significantly larger than those mounds you can see bordering them in Figure 1. Importantly, this event occurred before the western industrial revolution. Therefore, by looking at moraine building events before this, we can recreate how glaciers were before humanity’s dependence on fossil fuels developed, i.e. a time where glaciers were able to reach a point of stability and build landforms. By including smaller, distal moraines as well as the mammoth slopes of those most proximal to the glacier, we can construct a chronology of the Khumbu Glacier’s behaviour over the Late Holocene (2500 years ago to present in the Himalaya), into the last point of stability, and onto the behaviour we see today.

So, then what?

I’m glad you asked. Once we have a chronology for the glacier’s behaviour, we can start to compare it to the modern-day behaviour. This can be done using direct comparisons between glacier extent and thickness or using glacier models. Models can be useful as they are able to recreate the behaviour between the moraine building phases. I will be applying my chronology to a dynamic glacier model known as iSOSIA, one that was adapted to be able to simulate the development of debris covered glaciers. The chronology will act as parameters within the model (so that it knows the moraines must be built by a certain time), and the model can then recreate the glaciers behaviour whilst it was building these striking landmarks. We can use the model to improve our understanding of how these glaciers have become debris covered, what they would be like if they weren’t, and what might happen to them as climate change continues.

What is this all for?

Having travelled to the Khumbu valley and spent days staggering around on the debris-covered glacier trying in vain to catch my breath, the sense of our impact on the world hit home quite dramatically. Whilst the Khumbu valley is a particularly busy valley, you’re still days away from any form of infrastructure. The moment you travel off the well-worn path, it’s the most incredibly peaceful landscape. Not because it’s silent; the glacier is constantly making noise as it slowly flows down the valley and debris shifts around the surface. It is because it is entirely natural. When there are no helicopters to be heard, in that moment, you could be the only person alive. In the face of an ever-expanding world, I believe it is important to protect and preserve these natural spaces, and those dependent on them.

Humans will not last forever, or even for a long time in the grand scheme of things, but if we’re not careful, our impact might. I don’t think future populations of any species deserve that.

Figure 3 – Looking back up the valley whilst I contemplate a rock. Pumori summit can be seen to the left of the photograph, and Nuptse summit can be seen on the right. Existential questions on a postcard please. [Credit: Martin Kirkbride]

Further Reading

Edited by Scott Watson


Jo Hornsey is a PhD student at the Department of Geography in Sheffield, UK. She is researching the changing extent of Himalayan glaciers over the last 2500 years with specific focus on the Little Ice Age event ~500 years ago; dating the patterns of glacier retreat in the Khumbu Valley; using this information to improve accuracy of the iSOSIA model; and applying IPCC Climate Scenarios to analyse future glacier behaviour in the Khumbu Valley. In her downtime she can be found walking, climbing, running, acting, playing Dungeons and Dragons, and invariably trying to show you pictures of her cats.
Twitter – @joshornsey
Email – jhornsey1@sheffield.ac.uk

Image of the Week – Unravelling the mystery of the 2017 Weddell Polynya

Figure 1: The Moderate Resolution Imaging Spectroradiometer (MODIS) on NASA’s Terra satellite acquired these images of the Maud Rise or Weddell polynya in the eastern Weddell Sea on September 25, 2017. The first image is natural color and the second is false color where areas of ice are in blue and clouds are in white. [Image credit: NASA Earth Observatory].

The mysterious appearance and disappearance of the Weddell Polynya, a giant hole in the ice, has long puzzled scientists. Recent work reveals that it is tightly tied to energetic storms. Read on to find out more…


The eastern side of the Weddell Sea is a region known for its low concentration of sea ice due to the presence of a seamount, an underwater plateau called the Maud Rise. The seamount influences ocean circulation by bringing warm water closer to the surface, preventing the formation of thick ice. In the early 1970s, when satellites first began snapping photos of Earth, scientists noticed a mysterious hole in Antarctica’s seasonal sea ice floating in this area. This phenomenon is known as a polynya, and for decades its occurrence went unexplained. Then in 2017, during the continent’s coldest winter months, when ice should be at its thickest, a giant 9,500-square-kilometre hole suddenly showed up in the same region (Figure 1). Two months later it had grown 740% larger, before merging with the open ocean at the beginning of the melt season.

The Weddell Polynya is a rather famous hole in the ice (see this previous post). Scientists have been investigating such features in the Southern Ocean for decades, but the true reasons for the appearance and disappearance of the Weddell Polynya were still surrounded by mystery – until now.

Why does the Weddell Polynya form?

Recently, our new study found that these mid-sea polynyas can be triggered by strong cyclonic storms. Using satellite observations and reanalysis data, we found that in some winters, atmospheric circulation moves a significant amount of heat and moisture from mid-latitudes to Antarctica, allowing large cyclones to develop over the sea ice pack. When strong cyclones – some as strong as hurricanes – form and spin over the ice pack, the strong cyclonic winds they can drag the floating sea ice in opposite directions away from the cyclone center, creating the opening.

Sea ice typically drifts in a direction turned 30° on average to the left of the atmospheric flow, with a speed amounting to 1–2% of the surface wind speed. Those rules, when applied to a cyclonic wind situation (i.e., two opposing winds around a center), imply divergence in the motion of sea ice leading to open water area within the cyclone center, as in Fig. 2. We can see how such a situation occurs in real life for the Weddell Polynya when looking at Fig. 3, where near-surface winds exceeding 20 m/s are pushing the ice in opposite directions away from the cyclone center, characterized by weak winds, and the hole in the ice underneath it.

Figure 2: Sketch summarizing the mechanisms by which the cyclone can open the polynya [Credit: Francis et al., 2019].

Why does the Weddell Polynya matter?

Once opened, the polynya works like a window through the sea ice, transferring huge amounts of energy during winter between the ocean and the atmosphere. Because of their large size, mid-sea polynyas are capable of impacting the climate regionally and globally. This includes impact on the regional atmospheric circulation, the global overturning circulation, Antarctic deep and bottom water properties, and oceanic carbon uptake. It is important for us to identify the triggers for their occurrence to improve their representation in models and their effects on climate.

What might happen in the future?

Under future warming-climate conditions, previous studies have predicted an intensification of the activity of polar cyclones and a poleward shift of the extratropical storm track. Others have shown that a poleward shift of the cyclone activity can result in a reduced sea ice extent, a situation similar to that observed in 2016 and 2017. When the sea ice extent is reduced, preferable polynya areas (i.e. areas of thinner ice, for example the Maud Rise) located in the ice pack become closer to the ice edge and hence to the cyclogenesis zone. Given the link between polynya occurrence and cyclones, polynya events may thus become more frequent under a warmer climate.

Figure 3 AMSR2‐derived sea ice concentrations on 16 September 2017 at 1200 UTC (colors) and ERA5 10‐m winds less than 20 m/s in black contours, and greater than 20 m/s in red contours.The solid yellow contour is the 15% ERA‐Interim sea ice contour, the dotted yellow contour is the 50% ERA‐Interim sea ice contour, and the dashed white contour is the 15% ice from satellite data delineating the polynya area. [Credit: Francis et al., 2019].

Further reading

Edited by Lettie Roach


Diana Francis is an atmospheric scientist at New York University Abu Dhabi, UAE. She investigates atmospheric dynamics in polar regions with focus on polar meteorology and links to changes in land and sea ice conditions. To this end, she uses regional models together with available observations and reanalyses. She tweets as @drdianafrancis.
Contact Email:  diana.francis@nyu.edu

Climate Change & Cryosphere – Caucasus Glaciers Receding

Climate Change & Cryosphere – Caucasus Glaciers Receding

The Tviberi Glacier valley is located in the Svaneti Region – a historic province of the Georgian Caucasus. Between 1884 and 2011, climate change has led to a dramatic retreat of the ice in this valley. Other glaciers in the Greater Caucasus evolved in a similar way in past decades. We investigated glaciers and their changes both in-situ and with remote sensing techniques in the 53 river basins in the southern and northern slopes of the Greater Caucasus in order to analyze glacier dynamics in combination with climate change over the last decades…


Why are glaciers important for the Caucasus region?

On the one hand, in a high mountain system such as the Greater Caucasus, glaciers are the source of rivers through snow and ice melting. They are therefore an important source of water for agricultural production, for several hydroelectric power stations, for water supply, and for recreational opportunities. Also, the Greater Caucasus glaciers have a positive impact on the economy by being a major tourist attraction. The Svaneti, Racha and Kazbegi regions in Georgia welcome thousands of visitors each year.

On the other hand, glacier hazards are relatively common in this region, leading to major casualties. On the 20th September 2002, for example, Kolka Glacier (North Ossetia) initiated a catastrophic ice-debris flow killing over 100 people, and, on the 17th May 2014, Devdoraki Glacier (Georgia) caused a rock–ice avalanche and glacial mudflow killing nine people (Tielidze and Wheate, 2018).

 

Tviberi Glacier Degradation over the last century

According to our investigation, the Tviberi was the largest glacier of the southern slope of Georgian Caucasus in the end of the 19th century with a total area of 49.0 km2. The glacier terminated at a height of 2030 m above sea level (a.s.l) in 1887 (based on topographical maps, see Fig.2a). Before the 1960s, the largest ice stream – the Kvitoldi Glacier – separated from the Tviberi, and became an independent glacier (Fig. 2b). The 1960 topographical map shows that, as a consequence, the Tviberi Glacier shrinked to an area of 24.7 km2 and the glacier tongue ended at 2140 m a.s.l. (Fig. 2b). Finally, the Landsat 2014 image shows the degradation of the Tviberi Glacier after 1960, as it decomposed into smaller simple-valley glaciers and even smaller cirque glaciers developed (Fig. 2c) (Tielidze, 2016).

Fig.2: a – Tviberi Glacier, topographical map 1887; b – topographical map 1960, 1: Tviberi Glacier, 2: Kvitlodi Glacier; c – Landsat L8 imagery 2014. [Credit: modified from Fig.2 in Tielidzle, 2016]

Latest Caucasus Glacier Inventory

In our remote-sensing survey of glacier change in the Greater Caucasus based on large-scale topographic maps and satellite imagery (Corona, Landsat and ASTER), we show that the evolution of the Tviberi Glacier reflects the evolution of the majority of glaciers in the region. The main aim of this study was to present an updated and expanded glacier inventory at three time periods (1960, 1986, 2014) covering the entire Greater Caucasus (Russia-Georgia-Azerbaijan).

According to our study, glaciers on the northern slope of the Greater Caucasus lost 0.50% of their area per year between 1960 and 2014, while the southern slope glacier area decreased by 0.61% per year. Glaciers located on Mt. Elbrus lost 0.27% of their combined area per year during the same period. Overall, the total ice area loss between 1960 and 2014 was 0.53% per year, while the number of glaciers reduced from 2349 to 2020 for the entire Greater Caucasus (Fig. 3) (Tielidze and Wheate, 2018).

 

Fig.3: Greater Caucasus glacier area decrease by slopes, sections and mountain massifs in 1960–1986, 1986–2014 and 1960–2014 [Credit: Fig.4 in Tielidze and Wheate, 2018]

We have observed strong positive linear trends in the mean annual and summer air temperatures at all selected meteorological stations for the period 1960-2014 (Fig. 4). These climate data suggest that the loss of glacier surface area across the Greater Caucasus between the 1960 and 2014 mostly reflects the influence of rising temperatures in both the northern and southern slopes of the Greater Caucasus. The highest temperature increase was recorded in the eastern Greater Caucasus where the glacier recession was highest at the same time. If the decrease in the surface area of glaciers in the eastern Greater Caucasus continues over the 21st century, many will disappear by 2100 (Tielidze and Wheate, 2018).

 

Fig.4: Mean annual air temperatures at the seven meteorological stations in the years 1960–2014. [Credit: Levan Tielidze]

Want to use these and more data?

This new glacier inventory has been submitted to the Global Land Ice Measurements from Space (GLIMS) database and can be used as a basis data set for future studies.

 

Further reading

Edited by Clara Burgard


Levan Tielidze is a senior research scientist at Institute of Geography, Tbilisi State University. He is also a PhD student of School of Geography, Environmental and Earth Sciences, and Antarctic Research Centre at Victoria University of Wellington. The field of his research is modern glaciers and glacial-geomorphological studies of the mountainous areas in the Quaternary (Late Pleistocene and Holocene). Contact Email: levan.tielidze@tsu.ge/levan.tielidze@vuw.ac.nz.

Image of the Week – The GReenland OCEan-ice interaction project (GROCE): teamwork to predict a glacier’s future

Figure 1: The GROCE project, with 11 working groups and more than 30 scientists from across Germany, aims to understand what the present-day state of the 79°N glacier in Greenland is. On a windy day in May 2019, the GROCE teams met up at the annual update meeting to present findings and discuss the next steps to understand this complex system. Photo credit: Mario Hoppman, AWI


The GROCE project, funded by the German Ministry for Education and Research (BMBF), takes an Earth-System approach to understand what processes are at play for the 79°N glacier (also known as Nioghalvfjerdsfjorden), in northeast Greenland. 79°N is a marine-terminating glacier, meaning it has a floating ice tongue (like an ice shelf) and feeds into the ocean. Approximately 8% of all the ice contained in the Greenland Ice Sheet feeds into the 79°N glacier before it reaches the ocean (Seroussi et al., 2011). Therefore, in a worst-case-scenario where it melted entirely, this would lead to 1.1 m of sea level rise (Mayer et al., 2018). In recent years, the glacier’s ice flow to the ocean has increased in speed (Khan et al., 2014) and at the same time, the atmosphere at the surface in the region has warmed by 3°C over the last 40 years (Turton et al., 2019). This means that the 79°N glacier is being affected by both the warming atmosphere and the warming ocean simultaneously and will therefore be highly sensitive to future climate changes. However, without understanding its current state through accurate monitoring, predicting what the future may hold for this glacier is difficult.

Figure 2: The 79°N glacier and its floating tongue respond to warm waters coming from the surrounding ocean, producing melt-water which circulates underneath the floating tongue. In turn, the less saline melt-water affects and changes the large-scale ocean circulation itself. Simultaneously, at the surface, a warming atmosphere leads to more surface melt-water, some of which drains to the base of the glacier. Infographic credit: Mario Hoppman/AWI/Martin Künsting.

We are 11 different working groups, all attempting to understand the 79°N glacier, each group investigating a different, but complementary, aspect of the glacier. Our investigations include: the interactions between the atmosphere and the ice, the ocean circulation around the ice tongue, melting at the surface of the ice, melting near the bedrock beneath the ice, the location of the grounding line (where the ice meets the ocean and starts to float to form the floating ice tongue), tidal processes and many more (see Figure 2 for some of these processes). The 11 working groups, which includes 34 scientists and PhD students from 8 universities and research institutes, are spread all over Germany: from FAU University of Erlangen-Nürnberg (most southerly) to IOW (Leibniz Institute for Baltic Sea Research Warnemünde) (see Figure 3 for a map of the locations of all the research groups). The project is about to enter its third and final year, which means a lot of exciting new results are emerging from the project and there will be many more to come…

Figure 3: The locations of the 8 main partner universities and research institutions involved in the GROCE project. Figure made with google my maps.

If you’re interested in learning more about the GROCE project, its members or research outcomes, you can find a lot of information on our website: www.groce.de.

Figure 4: Jenny Turton reports on the progress made in regional atmospheric modelling efforts within the last year during the progress meeting in May 2019. Photo credit: Mario Hoppman, AWI

Edited by Marie Cavitte


Jenny Turton is a post-doc researcher in the climate system research group at Friedrich-Alexander University (FAU) in Erlangen. She currently investigates the interaction between the atmosphere and the cryosphere. More specifically, her current research focuses on the link between atmospheric processes and the glacier surface of 79°N glacier in northeast Greenland.

 

Image of the Week – The Lost Meteorites of Antarctica…

Image of the Week – The Lost Meteorites of Antarctica…

 When most people think of Antarctica, meteorites aren’t the first things that come to mind. Perhaps they imagine the huge ice shelves, the desolate interior, or perhaps penguin colonies near one of the scientific bases — but usually not meteorites. So why is our project looking for meteorites in Antarctica, and besides, aren’t they all lost until they are found?


Let’s start with the Antarctic part. Surprisingly, Antarctica is a great place to hunt for meteorites, with two-thirds of all known meteorites found there (see also this previous post). Despite the difficult searching conditions, the dark meteorites show up well on the bright white surface of the ice sheet. Also, due to the cold conditions, the weathering of meteorites is slower than in warmer regions, such as hot deserts. However, Antarctica is vast and most meteorites are small (a few cm in size) so finding them is still difficult. One thing that helps in this case is the presence of mountain ranges and nunataks near the perimeter of the continent. These slow the ice flow and cause stagnation points — blue ice areas (see Fig.1) — where meteorites can collect and accumulate on the surface usually at elevations of 1500m to 2500m or so.

To understand the “lost” part, you need to know that not all meteorites are created equally – some are stony, somewhat like the rocks found in the Earth’s crust, whilst others are more metallic, having a high iron content perhaps close to 90%. The latter tell a meteoriticist (yep, this is a real job definition!) about the cores of planets and the early formation of the solar system. In the rest of the world these rarer and more interesting iron meteorites usually make up about 5.2% of all samples found — but on the ice sheet, they only represent about 0.6%, an order of magnitude less.

This under-representation is a little odd and unexpected. We don’t expect any particular bias due to delivery to Earth, and if you’re out hunting for meteorites on your skidoo and spot a dark rock that’s potentially a meteorite, you’re still going to pick it up: so no bias in collection.

So where are the missing iron meteorites?

About 3 years ago a team from Manchester were looking at this problem and came up with a hypothesis that might explain it: as sunlight enters the blue ice and gets scattered below the surface it gets absorbed by meteorites and heats them up. This happens more on their upper surface, and due to the higher conductivity of iron meteorites in comparison to their stony counterparts, the heat is transferred to the ice below. Given the right conditions during the Austral summer, this process can provide enough energy so that iron meteorites melt the ice underneath them and sink a few centimetres below the surface whilst the stony ones stay on top. Though we put this on a mathematical footing by way of a model and confirmed the mechanism by some laboratory experiments, the only way to be sure that there is a layer of iron meteorites hiding below the surface is to go out and find them.

Fig. 2: Meteorite close-up: partially embedded in the blue ice surface [Credit: Katherine Joy/Lost Meteorites of Antarctica].

The current project

If you’re going to try and find iron meteorites hidden below the ice surface in Antarctica, then you’re going to need a new way of doing things. Ordinarily, searches are carried out by systematically searching an area of blue ice on skidoo and looking for dark rocks that might be potential meteorites (see Fig. 2). Obviously this won’t work for samples below the surface, so we had to come up with a new method that allows a good sized area to be covered (even in a relatively productive area we estimate the density of irons is <1 km-2) and can cope with the conditions up on the Antarctic plateau. Given that the key discriminating characteristic of the subsurface meteorites is their metallic content, it makes sense to use a system based on metal detector technology.

Our system is somewhat different to what you might use for hunting for archaeological coins or have to walk through at the airport, and instead the detector coils are embedded into an array of large polymer panels (the same material the British Antarctic Survey use for transporting fuel drums on). It’s entirely bespoke, including the pulse and detection electronics, data acquisition and analysis, and importantly, it’s designed to be able to deal with the conditions we expect down South.

Fig 3: The prototype detection system being tested at Sky-Blu Field Station during the 2018/19 season. Panels are on the left of the image, control electronics are beneath the photovoltaic panel in the centre, and an indicator box on the skidoo shows when something metallic is detected [Credit: Geoff Evatt/Lost Meteorites of Antarctica].

Even so, given the logistical challenges of remote working in Antarctica (the area we want to search is ~700 km from the nearest base), we thought it prudent to do some tests of the prototype equipment closer to home. To that end, we’ve had two field trips to the UK Arctic Research Station in Ny Ålesund on Svalbard for the initial testing of the detection equipment.

In addition, we need to figure out exactly where to search. We therefore need to confirm an area has surface meteorites before we can hope to find the subsurface layer. That was the point of last Austral summer’s expeditions: while one team kept close to the Sky-Blu base for the first Antarctic test of full detector array, the other team went out to the Recovery Glacier region for a visual search of surface meteorites. Thankfully both were a success, with our meteorite hunting team bringing back a total of 36 surface meteorites.

Fig 4: The meteorite search team’s home for a few weeks. 700 km from Halley and poor weather meant return to civilisation was delayed until the Twin Otter could fly in [Credit: Katherine Joy/Lost Meteorites of Antarctica].

A bit more planning and we’re lined up for next year’s expedition where it all comes together and we find out if the hidden layer is there – or there’s something else at work…

You can find out more about the project and what we’re up to on the project blog: https://ukantarcticmeteorites.com/blog/

Further reading

Edited by Clara Burgard


Andy Smedley trained as an atmospheric scientist measuring and modelling how sunlight interacts with the atmosphere. Recently his research interests have expanded to include sunlight’s interaction with, and impacts on, the cryosphere. He is currently working on the Leverhulme Trust funded “Lost Meteorites of Antarctica” project at the University of Manchester where he deals with the logistics of Antarctic field expeditions, mapping and analysis to select the field sites, and trying to better understand how solar radiation interacts with blue ice and light absorbing particles – including meteorites.

Image of the Week – Who let the (sun)dogs out?

Figure 1a: Atmospheric formations on the interior Antarctic plateau near Dome Fuji. Photo credit: B. Van Liefferinge

How peaceful it is to contemplate the sky … This is especially true of polar northern or southern skies where the low temperatures can engender unique light phenomena. We often tend call them all, wrongly, sundogs, but in fact, many more phenomena exist. To list a few, you can observe a parhelic circle, a 22° halo, a pair of sun dogs, a lower tangent arc, a 46° halo, a circumzenithal arc, a parry arc, … This year, I had the chance to observe several of these phenomena during my fieldwork on the Antarctic plateau. I am no cloud specialist or meteorologist, but I would like to give you some explanations to better understand this sky art you might see one day.


Figure 1b: Atmospheric formations on the interior Antarctic plateau near Dome Fuji. Photo credit: B. Van Liefferinge

Everything starts with the ice crystals inside the clouds. Ice crystals can be approximated as hexagonal prisms, and two shapes can be found naturally: plates and columns (Fig.2), mainly dependent on air temperature and humidity. Clouds formed from these ice crystals are relatively thin and therefore sun rays can pass through them easily. In cold regions (polar or not), the ice crystals can also be found at ground level (in that case, they are called diamond dust).

Figure 2: atmospheric halos in the Antarctic plateau, 27 ‎November ‎2018. Plate crystal (left) and column crystal (right), modified from Walter Tape, 1994. Photo credit: B. Van Liefferinge

The observed light patterns in the sky are caused by the refraction of sun rays on the ice crystals. As they hit the crystals, the sun rays are diverted from their trajectory. Depending on the orientation of the crystals in the cloud, and the type of crystal present in the cloud, this will affect the sun ray paths differently. The resulting halos or arcs seen in the sky will result from the combination of all the different ray paths through all the clouds’ crystals together. The sun ray paths through the crystals can become quite complex, as illustrated in Fig. 3.

Figure 3: Example of a complex ray path through a plate crystal which contributes to forming a left parhelic circle (left), example of a complex ray path through a hexagonal crystal which contributed to forming a tricker arc (right), (Walter Tape, 1994). Photo credit: B. Van Liefferinge

As an example, let’s try to understand how the 22° halo forms (see Fig.1 a and b), one of the most common halos observed. The 22° halo is formed in clouds that have randomly oriented column crystals (although this is still under debate). When a sun ray hits an ice crystal, the most common path it will take is to refract through the face of the hexagon it hits (e.g. face 1 on Fig.4) and refract back out of a face opposite (face 3 on Fig.4). The angle between face 1 and face 3 is of 60°. A light ray passing through two faces of an ice crystal inclined at 60° from each other is deflected through angles from 22° up to 50°. The deflection at 22° is the most probable and therefore creates the brightest circle in the sky (the 22° halo!). The other deflections above 22° are less common but occur nonetheless and form the fade disk (Fig. 4). No light can be refracted through smaller angles then 22° (this is a result of the air-to-ice index of refraction). This is why you see a “darker sky” inside the halo. Now, add to this that we have been considering white light in general. But in fact, visible white light is composed of visible red through blue rays, which do not deviate by the same amount (21.54° for red light to 22.37° for blue light). This explains why the 22° halo looks like a rainbow.

Figure 4: Sun ray path through an ice crystal (left), resulting 22° halo (right) [Credit: B. Van Liefferinge].

Now let’s go back to our famous sundog. If you understood the 22° halo, it will be a piece of cloud! Sundogs follow the same rule as the 22° halo: sun rays passing through two ice crystal faces inclined at 60° to each other are deflected through a minimum angle of 22°. The difference here is that for a sundog to form, the ice crystals must all be aligned along the horizontal direction. This is the case when the ice crystals are all plate crystals (Fig.5), which, as a result of their shape, tend to align horizontally in clouds. As sun rays traverse them, they are deflected into two specific spots either side of the sun, instead of along a circle when crystals are randomly oriented (like for the 22° halo).

Figure 5: Sundog formation [Credit: Atmospheric Optics website]

And finally, I’ll let you enjoy a light show, filmed recently over Svalbard on a perfect day for atmospheric formations (video courtesy of Ashley Morris)…

 

A few more fun facts:

Edited by Marie Cavitte


Brice Van Liefferinge is a trained geographer, glaciologist and modeller. With a background in geography at the Université libre de Bruxelles (ULB, Belgium), he pursued his interest in Earth sciences during his PhD looking at the thermal regime of the Antarctic Ice Sheet and working on the Beyond Epica Oldest Ice project. He is now working on the Oldest Ice Dome Fuji project with Dr. Kenny Matsuoka at the Norwegian Polar Institute (NPI, Tromsø, Norway) for which he just came back from 3 months of fieldwork at Dome Fuji, Antarctica.

 

Image of the Week – Kicking the ice’s butt(ressing)

Risk map for Antarctic ice shelves shows critical ice shelf regions, where local thinning increases the ice flow from the continent into the ocean [Credit: modified from Reese et al., 2018]

Changes in the ice shelves surrounding the Antarctic continent are responsible for most of its current contribution to sea-level rise. Although they are already afloat and do not contribute to sea level directly, ice shelves play a key role through the buttressing effect. But which ice shelf regions are most important for this?


The role of ice-shelf buttressing

Schematic ice-sheet-shelf system: buttressing arises when an ice shelf is laterally confined in an embayment or locally grounds at pinning points [Credit: Ronja Reese & Maria Zeitz]

In architecture, the term “buttress” is used to describe pillars that support and stabilize buildings, for example ancient churches or dams. In analogy to this, buttressing of ice shelves can stabilize the grounded ice sheet (see this blog article about the marine ice sheet instability). It can be understood as a backstress that the ice shelf exerts on the grounding line – the line that separates the grounded ice from the floating ice shelves. When an ice shelf thins or disintegrates, this stress can be reduced, then the ice flow upstream is less restrained and can increase.

This effect has been widely observed in Antarctica: the thinning of ice shelves in the Amundsen Sea is driven by the ocean and linked to ice loss there (see this blog article) and after the spectacular disintegration of Larsen A and B ice shelves the adjacent ice streams accelerated.

Which ice shelf regions are important?

Risk maps show how important each ice-shelf location is: if an ice shelf thins in this location, how much does the flux across the grounding line increase? We estimated this immediate increase using the numerical ice-flow model Úa. At first glance, one can see that all ice shelves have regions that influence upstream ice flow, and thus, provide buttressing. The highest responses occur near grounding lines of fast-flowing ice streams. Equally strong responses are found in the vicinity of ice rises or ice rumples – where the ice shelf re-grounds locally and is subject to basal drag. On the other hand, “passive” regions with negligible flux response are located towards the calving front, but also in spots close to the grounding line. Flux response signals can sometimes travel quite far – for example a perturbation near Ross Island accelerates the ice flow in almost the entire Ross Ice Shelf and reaches ice streams more than 900km away (not visible in the figure).

Risk maps for Antarctic ice shelves, as presented here, help us to get a better understanding of the critical ice shelf regions – if you are interested to read more, please see for example Gagliardini, 2018 and Reese et al., 2018.

Edited by Scott Watson and Sophie Berger


Ronja Reese is a postdoctoral researcher at the Potsdam Institute for Climate Impact Research, Germany, in the ice dynamics working group. She investigates ice dynamics in Antarctic with a focus on ice-ocean interactions and ice-shelf buttressing. She created the risk map together with Ricarda Winkelmann, Hilmar Gudmundsson and Anders Levermann. Contact Email: ronja.reese@pik-potsdam.de

.

Image of the Week – Fifty shades of May (Glacier)

Image of the Week – Fifty shades of May (Glacier)

With over 198 000 glaciers in the world, you can always find a glacier that fits your mood or a given occasion. So why not for example celebrate the first Image of the Week of May with a picture of the aptly named May Glacier?


May Glacier is in fact not named after the month, but after Mr May, an officer onboard the Flying Fish during her expedition to the East Antarctic coast in the 1840s. Apart from that, there is not much to say about this 9 x 11 km glacier located in East Antarctica around 130°E, except that it is really hard to find a picture with the keywords “may” and “glacier”… So I let you enjoy the Image of the Week, which combines three satellite images of May Glacier (at the centre) and its surroundings exactly a month ago, when the skies were clear and the sea ice pretty:

See you in a month to talk about the June(au) ice field!

Image of the Week – Life in blooming melting snow

Melting snowfields in a forested catchment of glacial lake in Šumava (mid-April), the Czech Republic [Credit: Lenka Procházková]

The new snow melting season has just started in the mountains of Europe and will last, in many alpine places, until the end of June. Weather in the middle of April is changeable. In the last few days sub-zero air temperatures have prevailed in the mountains during the day. In a frame of an international research project, me (Charles University) and Daniel Remias (Applied University Upper Austria), are both packing warm winter clothes as well as all the research equipment necessary for a new field mission: the aims are to find blooming spots of snow algae and to collect it for analyses. Upon our arrival in Šumava, a surprising but wonderful sunny day welcomes the expedition and we regret not taking the sun cream with us. While we are walking on still-compact partly frozen snowfields, our heads feel that they are exposed to hot summer.


Snow blooms – what do they look like?

Red snow colouration at nearly all ice-covered parts of a high-alpine glacial lake (mid-June), High Tatras (Slovakia). Detailed view of red snow after harvest [Credit: Daniel Remias and Lenka Procházková, see study Procházková et al. 2018a]

Snow blooms – see the figure above – can be found in polar and alpine regions worldwide. Availability of liquid water is a key factor for the development of a snow algae population. In our experience, only wet and slowly melting snowfields are suitable.  This colourful phenomenon can appear in different colour shades, as green, yellow, pink, orange or blood-red (Procházková et al., 2018a). Snow blooms are currently a focus of an increasing number of studies because of their significant effects on albedo reduction and subsequent acceleration of snow and ice melting.

Why are they colourful?

A few representatives of microalgae forming blooming snow – a coloured frame of each of these species corresponds with a colour of blooming snowfields [Credit: Lenka Procházková and Daniel Remias]

The macroscopic blooms are caused by microalgae of a cell size ranging from ~5 µm up to ~100 µm. During the melting season, cells live in a water film microhabitat surrounding large snow grains. The main genera that form these blooms are Chloromonas, Sanguina and Chlainomonas, each associated with a specific bloom color (see the figure above).  A massive population development of golden algae can also occur.

When in the season do blooms occur?

Typical seasonal life cycle of a snow alga (Chloromonas nivalis), based on observation over many seasons in European Alps [Figure modified with permission from Sattler et al. 2010]

I would like to reveal a few secrets of snow algae.
The first strategy represents their seasonal life cycle. At the beginning the season in late April, one can hardly see any snow colouration. Snow algae from the previous seasons are lying at the interface between snow and soil in a resistant stage (called cyst). Snow is starting to melt slowly, and the cysts recognize the availability of liquid water and germinate. Flagellates are released and migrate upwards to the sub-surface layers, where they mate. With proceeding melting the cysts are accumulated and exposed at the snow surface. After total snowmelt these resistant stages should survive over summer in soil or at bare rock, where they can be subject to long-distance transport by wind.

The red colour of snow is caused by astaxanthin

A cross-section of a typical snow algal cyst, Chloromonas nivalis-like species, with abundant lipid bodies (“L”) with astaxanthin and plastids (“P”) [transmission electron microscope, credit: Lenka Procházková]

The next strategy of snow algae is an accumulation of the red pigment astaxanthin during their maturation, which has many benefits to life of these microorganisms. For example, astaxanthin is a powerful antioxidant, and its synthesis is not limited by the supply of nitrogen.
Another big advantage of astaxanthin is its protective action against excessive visible and harmful ultraviolet irradiation which are characteristic for snow surfaces in alpine and polar regions. This “sunscreen” effect of astaxanthin – which has maximum absorbance in the visible light region and also a significant capability of UV protection – is supported by the algae’s clever intracellular arrangement (shown in the figure above), namely that sensitive compartments of the algae, like chloroplast or nucleus, are located in the central part, whereas lipid bodies, which accumulate the astaxanthin, are in the periphery.

Our mission

Sequence-related sampling in Lower Tauern, Austria. Checking of a qualitative composition of a sampled spot using light microscope already in field. [Credit: Linda Nedbalová]

Do you wonder why we explore the physiology and biodiversity of snow algae? Because these extremophilic organisms cope with high ultraviolet radiation, repeated freeze-thaw cycles, desiccation, mechanical abrasion, limited nutrients and short season and are well adapted to it! Because of their ability to adapt to these extreme conditions, pigments of snow algae (as the astaxanthin presented above) are even used as biomarkers to detect life on Mars! Moreover, these microalgae are essential primary producers in such an extreme ecosystem, where phototrophic life is restricted to a few specialised organisms. For instance, they provide a basic ecosystem for snow bacteria, fungi and insects. Snow algae communities play an important role in supraglacial and periglacial snow food webs and supply nutrients that will be delivered throughout the glacial ecosystem.

 

Further reading

                               Edited by Jenny Turton


Lenka Procházková is a PhD Student at the Charles University, Prague, the Czech Republic. She investigates biodiversity and ecophysiology of snow algae. Her favourite algal group is in her focus in a lab as well as in field samplings in the European Alps, High Tatras, Krkonoše, Šumava and Svalbard. Contact Email:  lenkacerven@gmail.com