CR
Cryospheric Sciences

Sea-ice

Image of the Week — Seasonal and regional considerations for Arctic sea ice changes

Monthly trends in sea ice extent for the Northern Hemisphere’s regional seas, 1979–2016. [Credit: adapted from Onarheim et al (2018), Fig. 7]

The Arctic sea ice is disappearing. There is no debate anymore. The problem is, we have so far been unable to model this disappearance correctly. And without correct simulations, we cannot project when the Arctic will become ice free. In this blog post, we explain why we want to know this in the first place, and present a fresh early-online release paper by Ingrid Onarheim and colleagues in Bergen, Norway, which highlights (one of) the reason(s) why our modelling attempts have failed so far… 


Why do we want to know when the Arctic will become ice free anyway? 

As we already mentioned on this blog, whether you see the disappearance of the Arctic sea ice as an opportunity or a catastrophe honestly depends on your scientific and economic interests.  

It is an opportunity because the Arctic Ocean will finally be accessible to, for example: 

  • tourism; 
  • fisheries; 
  • fast and safe transport of goods between Europe and Asia; 
  • scientific exploration. 

All those activities would no longer need to rely on heavy ice breakers, hence becoming more economically viable. In fact, the Arctic industry has already started: in summer 2016, the 1700-passenger Crystal Serenity became the first large cruise ship to safely navigate the North-West passage, from Alaska to New York. Then in summer 2017, the Christophe de Margerie became the first tanker to sail through the North-East passage, carrying liquefied gas from Norway to South Korea without an ice breaker escort, while the Eduard Toll became the first tanker to do so in winter just two months ago. 

On the other hand, the disappearance of the Arctic sea ice could be catastrophic as having more ships in the area increases the risk of an accident. But not only. The loss of Arctic sea ice has societal and ecological impacts, causing coastal erosion, disappearance of a traditional way of life, and threatening the whole Arctic food chain that we do not fully understand yet. Not to mention all of the risks on the other components of the climate system. (See our list of further readings at the end of this post for excellent reviews on this topic). 

Either way, we need to plan for the disappearance of the sea ice, and hence need to know when it will disappear. 

Arctic sea ice decrease varies with region and season 

In a nutshell, the new paper published by Onarheim and colleagues says that talking about “the Arctic sea ice extent” is an over simplification. They instead separated the Arctic into its 13 distinct basins, and calculated the trends in sea ice extent for each basin and each month of the year. They found a totally different behaviour between the peripheral seas (in blue on this image of the week) and the Arctic proper, i.e. north of Fram and Bering Straits (in red). As is shown by all the little boxes on the image, the peripheral seas have experienced their largest long term sea ice loss in winter, whereas those in the Arctic proper have been losing their ice in summer only. In practice, what is happening to the Arctic proper is that the melt season starts earlier (note how the distribution is not symmetric, with largest values on the top half of the image).  

Talking about Arctic sea ice extent is an over simplification

Moreover, Onarheim and colleagues performed a simple linear extrapolation of the observed trends shown on this image, and found that the Arctic proper may become ice-free in summer from the 2020s. As they point out, some seas of the Arctic proper have in fact already been ice free in recent summers. The trends are less strong in the peripheral seas, and the authors write that they will probably have sea ice in winter until at least the 2050s. 

So, although Arctic navigation should become possible fairly soon, in summer, you may need to choose a different holiday destination for the next 30 winters. 

Melting summer ice. [Credit: Mikhail Varentsov (distributed via imaggeo.egu.eu)]

But why should WE consider the regions separately? 

The same way that you would not plan for the risk of winter flood in New York based on yearly average of the whole US, you should not base your plan for winter navigation from Arkhangelsk to South Korea on the yearly Arctic-wide average of sea-ice behaviour. 

Scientifically, this paper is exciting because different trends at different locations and seasons will also have different consequences on the rest of the climate system. If you have less sea ice in autumn or winter, you will lose more heat from the ocean to the atmosphere, and hence impact both components’ heat and humidity budget. If you have less sea ice in spring, you may trigger an earlier algae bloom. 

As often, this paper highlights that the Earth system behaves in a more complex fashion that it first appears. Just like global warming does not prevent the occurrence of unpleasantly cold days, the disappearance of Arctic sea ice is not as simple as ice cubes melting in your beverage on a sunny day.  

Reference/Further reading

Bhatt, U. S., et al. (2014), Implications of Arctic sea ice decline for the Earth system. Ann. Rev. Environ. Res., 39, 57-89 

Meier, W. N., et al. (2014), Arctic sea ice in transformation: A review of recent observed changes and impacts on biology and human activity. Reviews of Geophysics, 52(3), 185-217. 

Onarheim, I., et al. (2018), Seasonal and regional manifestation of Arctic sea ice loss. Journal of Climate, EOR.  

Post, E., et al. (2013), Ecological consequences of sea-ice decline. Science, 341, 519-524 

Edited by Sophie Berger

Image of the Week – The colors of sea ice

Image of the Week – The colors of sea ice

The Oscars 2018 might be over, but we have something for you that is just as cool or even cooler (often cooler than -20°C)! Our Image of the Week shows thin sections of sea ice photographed under polarized light, highlighting individual ice crystals in different colors, and is taken from a short video that we made. Read more about what this picture shows and watch the movie about how we got these colorful pictures…


Sea ice can vary in salinity

Sea ice forms differently than fresh water ice due to its salt content. When sea water begins to freeze, the ice crystals aren’t able to incorporate salt into their structure and hence reject salt into the surrounding water. This increases the density of the remaining sea water which sinks (see this previous post). Some salty water gets trapped between the crystals though. This water will also slowly freeze, always rejecting the salts into the remaining water. The saltier the water, the lower its freezing point. This means the remnant very salty water, which we call brine, remains liquid even at temperatures below -20oC!

Sea ice crystals can vary in shape

The first layer of sea ice is typically granular – the crystals are small and round, with a diameter around one centimeter. This is because this layer is formed in open seas, where the crystals which go on to form this layer are spun and broken up by surface waves. This granular structure includes lots of ‘pockets’ of trapped brine. Under this surface ice layer, which is typically 10-30 cm thick, ice starts growing in more sheltered conditions. Such sea ice is columnar. The crystals are flat and elongated – like layers in a vertical cake. The brine is trapped between these layers in brine channels. When ice is relatively warm, for example shortly after freezing or before it starts melting, such channels are wide and can be connected. Brine can then escape from them at the lower end into the ocean. The channels also allow small, hardy microscopic plants and animals to travel through the ice. Often air bubbles are trapped in them too.

Sea ice can vary in how it looks too!

The size and form of sea ice crystals – sea ice texture – impacts various properties of the sea ice including its salt content, density and suitability as a habitat. It also influences the optical properties of ice, however. While pure water ice is transparent (see this previous post), sea ice appears milky. That is because of brine channels and bubbles between the crystals.

When looking at large regions of sea ice from space by sensors mounted on satellites, sea ice texture will be important too. Visible light has a short wavelength and this means it only penetrates into the top millimeter of ice. Images collected in the visible light range (see this previous post) will show features dominated by the surface properties of the ice. In comparison, microwaves have a longer wavelength and can penetrate deeper into the ice. Hence imagery of the sea ice cover collected in the microwave spectrum of light (see this previous post) will display features influenced by the internal structure of the sea ice in addition to the surface features.

 

The video below shows how the sea ice samples are analyzed for texture and how we got the colorful pictures for our Image of the Week…

 

Further reading

Edited by Adam Bateson and Clara Burgard


Polona Itkin is a Post-doctoral Researcher at the Norwegian Polar Institute, Tromsø. She investigates the sea ice dynamics of the Arctic Ocean and its connection to the sea ice thickness. In her work she combines the information from in-site observations, remote sensing and numerical modeling. Polona is part of the social media project ‘oceanseaiceNPI’ – a group of scientists that communicates their knowledge through social media channels: Instagram.com/OceanSeaIceNPI, Twitter.com/OceanSeaIceNPI, Facebook.com/OceanSeaIceNPI, contact Email: polona.itkin@npolar.no

Image of the Week – A Hole-y Occurrence, the reappearance of the Weddell Polynya

Image of the Week – A Hole-y Occurrence, the reappearance of the Weddell Polynya

During both the austral winters of 2016 and 2017, a famous feature of the Antarctic sea-ice cover was observed once again, 40 years after its first observed occurrence: the Weddell Polynya! The sea-ice cover exhibited a huge hole (of around 2600 km2 up to 80,000 km2 at its peak!), as shown on our Image of the Week. What makes this event so unique and special?


Why does the Weddell Polynya form?

The Weddell Polynya is an open ocean polynya (a large hole in the sea ice, see this previous post), observed in the Weddell Sea (see Fig.2). It was first observed in the 1970s but then did not form for a very long time, until 2016 and 2017…

 

Fig. 2: Map of the sea ice distribution around Antarctica on 25th of September 2017, derived from satellite data. The red circle marks the actual Weddell Polynya [Credit: Modified from meereisportal.de]

In the Southern Ocean, warm saline water masses underlie cold, fresh surface water masses. The upper cold fresh layer acts like a lid, insulating the warmer deep waters from the cold atmosphere. While coastal polynyas (see this previous post) are caused by coastal winds, open ocean polynyas are more mysteriously formed as it is not as clear what causes the warm deep water to be mixed upwards. In the case of the Weddell polynya, it forms above an underwater mountain range, the Maud Rise. This ridge is an obstacle to the water flow and can therefore enhance vertical mixing of the deeper warm saline water masses. The warm water that reaches the surface melts any overlying sea ice, and large amounts of heat is lost from the ocean surface to the atmosphere (see Fig. 3).

 

Fig. 3: Schematic of polynya formation. The Weddell polynya is an open ocean polynya [Credit: National Snow and Ice Data Center].

 

Why do we care about the Weddell Polynya?

Overturning and mixing of the water column in the Weddell Polynya forms cold, dense Antarctic Bottom Water, releasing heat stored in the ocean to the atmosphere in the process. Antarctic Bottom Water is formed in the Southern Ocean (predominantly in the Ross and Weddell Seas) and flows northwards, forming the lower branch of the overturning circulation which transports heat from the equator to the poles (see Fig. 4). Antarctic Bottom Water also carries oxygen to the rest of the Earth’s deep oceans. The absence of the Weddell polynya could reduce the formation rate of Antarctic Bottom water, which could weaken the lower branch of the overturning circulation.

Fig.4: Schematic of the overturning (thermohaline) circulation. Deep water formation sites are marked by yellow ovals. Modified from: Rahmstorf, 2002 [©Springer Nature. Used with permission.]

How often does the Weddell Polynya form?

The last time the Weddell Polynya was observed was during the austral winters of 1974 to 1976 (see Fig. 5). It was then absent for nearly 40 years (!) up until austral winter 2016. In a modelling study, de Lavergne et al. 2014 suggested that the Weddell Polynya used to be more common before anthropogenic CO2 emissions started rising at a fast pace. The increased surface freshwater input from melting glaciers and ice sheets, and increased precipitation (as climate change increases the hydrological cycle) have freshened the surface ocean. This freshwater acts again as a lid on top of the warm deeper waters, preventing open ocean convection, reducing the production of Antarctic Bottom Water.

Fig. 5: Color-coded sea ice concentration maps derived from passive microwave satellite data in the Weddell Sea region from the 1970s. The Weddell Polynya is the extensive area of open water (in blue) [Credit: Gordon et al., 2007, ©American Meteorological Society. Used with permission.].

The reappearance of the Weddell Polynya over the past two winters despite the increased surface freshwater input suggests that other natural sources of variability may be currently masking this predicted trend towards less open ocean deep convection. Latif et al. 2013 put forward a theory describing centennial scale variability of Weddell Sea open ocean deep convection, as seen in climate models. In this theory, there are two modes of operation, one where there is no open ocean convection and the Weddell Polynya is not present. In this situation, sea surface temperatures are cold and the deep ocean is warm, and there is relatively large amount of sea ice. The heat at depth increases with time, as it is insulated by the sea ice and freshwater lid. Then, eventually, the deep water becomes warm enough that the stratification is decreased sufficiently so that open water convection begins again, forming the Weddell Polynya. This process continues until the heat reservoir depletes and surface freshwater forcing switches off the deep convection. Models show that the timescale of this variability is set by the stratification, and models with stronger stratification tend to vary on longer timescale, as the heat needs to build up more in order to overcome the stratification.

 

In the end, the Weddell Polynya is still surrounded by some mystery… Only the next decades will bring us more insight into the true reasons for the appearance and disappearance of the Weddell Polynya…

 

Further reading

Edited by Clara Burgard


Rebecca Frew is a PhD student at the University of Reading (UK). She investigates the importance of feedbacks between the sea ice, atmosphere and ocean for the Antarctic sea ice cover using a hierarchy of climate models. In particular, she is looking at the how the importance of different feedbacks may vary between different regions of the Southern Ocean.
Contact: r.frew@pgr.reading.ac.uk

Image of the Week – The Gap, the Bridge, and the Game-changer

The Gap, the Bridge, and the Game-changer, together with many of the passive microwave satellite missions relevant for sea ice concentration mapping for the period 1980s to 2030s [Credit: T. Lavergne].

The Gap, the Bridge, and the Game-changer are three series of satellites. They carry instruments that measure the microwave radiation emitted by the Earth (called passive microwave instruments), while flying 800 km above our heads at 7,5 km/s. Since the late 1970s, most sea ice properties (concentration, extent, area, velocity, age and more!) have been measured with such passive microwave instruments.
So who are the Gap, the Bridge, and the Game-changer? Their story is what this Image of the Week is about…


The Gap

Since 1978, the U.S. equipped 11 satellites with passive microwave instruments to observe global sea ice. These instruments are called SMMR, SSM/I and SSMIS. Their measurements have produced a continuous, almost 40 year long climate data record of sea ice (see how satellite observations are converted into sea ice properties in this previous post). However, as described late last year in a Nature article, the remaining three of these instruments are ageing, already beyond their expected lifetime, and with no planned continuation from the U.S (see SSMIS F16-18 on our Image of the Week).

Europe will be operating a series of similar instruments (the MicroWave Imagers, MWI) on their 2nd Generation Polar System from 2023. A (looming future) gap is feared if the last U.S. instruments fail before the European ones are fully operating.

The decline of summer sea ice extent in the Arctic is an iconic indicator of climate change and U.S. satellites have enabled and sustained its monitoring for all these years (see this earlier post). More than a news magnet, the satellite time series is a back-bone for our understanding of the evolution of global sea ice. It is a key asset for developing and evaluating our climate models. The possibility of a data gap understandably caught the attention of the scientific community and the general public. This (looming future) «Gap» is the first character in our story.

The Bridge

The «Bridge» is known under the code name Feng Yun 3 (FY3) MWRI and is Chinese. The FY3 programme, operated by the Chinese Meteorological Administration (CMA), is a series of satellites with passive microwave instruments very similar to the ones on the American and European satellites. FY3D -the 4th satellite in the FY3 series- was successfully launched in late 2017, bridging the data gap that was feared to happen, even if the remaining U.S. SSMIS satellites would fail next month.

Over the past few months, scientists at the EUMETSAT OSI SAF (the European Organization for the Exploitation of Meteorological Satellites – Ocean and Sea Ice Satellite Application Facility) have been investigating the quality of FY3 passive microwave data. They adapted their algorithms to retrieve sea ice concentration from raw satellite measurements, so that they yield very similar accuracy to the sea ice concentration data they obtain from the SSMIS. An example sea ice concentration map using the OSI SAF algorithm on raw FY3 data is shown below. Such maps can extend the climate data record released in early 2017, should the last SSMIS fail.

Sea Ice Concentration maps for February 6th 2018 (left: Northern Hemisphere, right: Southern Hemisphere). These are computed by the OSI SAF algorithms applied on raw FY3 MWRI data [Credit: A. Sørensen].

Access to the FY3 data was facilitated by bi-lateral agreements between EUMETSAT and CMA. National and international space agencies coordinate their activities in a variety of forums such as CEOS (Comittee on Earth Observation Satellites), CGMS (Coordination Group for Meteorological Satellites) or WMO PSTG (the World Meteorological Organization Polar Space Task Group) to cite a few. This global-scale coordination goes mostly unnoticed to the public and the scientific community. It is, however, a great aid for our ability to continuously monitor and predict the global environment.

You might think that, now that the Gap is Bridged, I have nothing more to tell you about passive microwave satellites for sea ice observations? Well, think again. There is a third character to our story: the «game-changer».

The Game-changer

Without further teasing you, our «game-changer» is CIMR. CIMR stands for the «Copernicus Imaging Microwave Radiometer». It might get selected for joining the family of Copernicus satellites some time in the late 2020s.

Before I tell you what makes CIMR so special, we need a short introduction on what passive microwave instruments are, why we like them for observing sea ice, and how they work:

T. Lavergne (2018) Passive Microwave Remote Sensing of Sea Ice : a crash-course in just four list items, Int. J. of Short Lists

  1. The best satellite instruments for measuring sea ice use the microwave part of the electromagnetic spectrum (from ~1 to ~100 GHz). This type of radiation does not depend on Sun light, and is not blocked by clouds.

  2. Passive microwave instruments record a tiny amount of radiation naturally emitted at the surface of the Earth and in the atmosphere. Aboard the satellite, the radiation is reflected by an antenna towards a recording instrument: the radiometer.

  3. Radiometers can measure at several frequencies. Once the images are back at the processing centers on Earth, algorithms are applied to compute geophysical products such as sea ice concentration.

  4. Radiometers with low frequencies (e.g. 6 GHz) yield best accuracy for sea ice concentration products. The bigger the antenna, the better the final resolution of the product.

One of a kind, the CIMR will focus on the low frequencies (6, 10, and 18 GHz), and fly an antenna big enough to ensure much better resolution than any of the passive microwave instruments we ever used before. This requires the antenna of CIMR to be substantially larger than that of SSMIS (60cm diameter), MWI (75cm) or even AMSR2 (2.1m)! The AMSR-E instrument and its followers were game-changers 15 years ago, and still offer the best resolution today… but future operational models and polar applications will require better sea ice products all too soon.

An exciting time opens for satellite-based observations of polar sea ice, as the pre-studies for CIMR are started by the European Space Agency this spring! Will industry take-up the challenge and build a big enough antenna for CIMR? Will CIMR be selected as EU’s future polar Copernicus mission? If “yes” to both, Europe will have a game-changer: high-resolution all-weather daily global accurate mapping of sea ice concentration.

I will definitely follow the developments with CIMR! Maybe I’ll tell you how it went in a future blog post? 🙂

Note: Were there too many acronyms in this blog? Well, we are sorry about that. Those satellite-people just LOVE their acronyms! A good resource for searching what satellite acronyms mean is the “Space capability” page from the World Meteorological Organization: https://www.wmo-sat.info/oscar/spacecapabilities (enter the acronym in the Quick Search, top-right for the page).

Further reading

Edited by David Docquier and Clara Burgard


Thomas Lavergne is a research scientist at the Norwegian Meteorological Institute. His main interest is in improving algorithms to improve sea ice satellite products, and help towards a better understanding between observation and model communities. He recently worked with EUMETSAT OSI SAF and ESA CCI to produce Climate Data Records for Sea Ice Concentration. He tweets as @lavergnetho.

Image of the Week – Arctic changes in a warming climate

Image of the Week – Arctic changes in a warming climate

The Arctic is changing rapidly and nothing indicates a slowdown of these changes in the current context. The Snow, Water, Ice and Permafrost in the Arctic (SWIPA) report published by the Arctic Monitoring and Assessment Program (AMAP) describes the present situation and the future evolution of the Arctic, the local and global implications, and mitigation and adaptation measures. The report is based on research conducted between 2010 and 2016 by an international group of over 90 scientists, experts, and members of Arctic indigenous communities. As such, the SWIPA report is an IPCC-like assessment focussing on the Arctic. Our Image of the Week summarizes the main changes currently happening in the Arctic regions.


What is happening to Arctic climate currently?

The SWIPA report confirms that the Arctic is warming much faster than the rest of the world, i.e. more than twice the global average for the past 50 years (Fig. 2). For example, Arctic surface air temperature in January 2016 was 5°C higher than the average over 1981-2010. This Arctic amplification is due to a variety of climate feedbacks, which amplify the current warming beyond the effects caused by increasing greenhouse gas concentrations alone (see the SWIPA report, Pithan & Mauritsen (2014) and this previous post for further information).

Fig.2: Anomaly of Arctic and global annual surface air temperatures relative to 1981-2010 [Credit: Fig. 2.2 of AMAP (2017), revised from NOAA (2015)].

This fast Arctic warming has led to the decline of the ice cover over both the Arctic Ocean (sea ice) and land (Greenland Ice Sheet and Arctic glaciers).

For sea ice, not only the extent has dramatically decreased over the past decades (see Stroeve et al. 2012 and Fig. 3), but also the thickness (see Lindsay & Schweiger, 2015). Most Arctic sea ice is now first-year ice, which means that it grows in autumn-winter and melts completely during the following spring-summer. In contrast, the multiyear sea-ice cover, which is ice that has survived several summers, is rapidly disappearing.

Fig. 3: Arctic sea-ice extent in March and September from the National Snow and Ice Data Center (NSIDC) and the Ocean and Sea Ice Satellite Application Facility (OSI SAF) [Credit: Fig. 5.1 of AMAP (2017)].

In terms of land ice, the ice loss from the Greenland Ice Sheet and Arctic glaciers has been accelerating in the recent decades, contributing a third of the observed global sea-level rise. Another third comes from ocean thermal expansion, and the remainder comes from the Antarctic Ice Sheet, other glaciers around the world, and terrestrial storage (Fig. 4, see also this previous post and Chapter 13 of the last IPCC report).

Fig. 4: Global sea-level rise contribution from the Arctic components (left bar), Antarctic Ice Sheet and other glaciers (middle-left bar), terrestrial storage (middle-right bar) and ocean thermal expansion (right bar) [Credit: Fig. 9.3 of AMAP (2017)].

Besides contributing to rising sea levels, land-ice loss releases freshwater into the Arctic Ocean. Compared with the 1980-2000 average, the freshwater volume in the upper layers of the Arctic Ocean has increased by more than 11%. This could potentially affect the ocean circulation in the North Atlantic through changes in salinity (see this previous post).

Other changes currently occurring in the Arctic include the decreasing snow cover, thawing permafrost, and ecosystem modifications (e.g. occurrence of algal blooms, species migrations, changing vegetation, and coastal erosion). You can have a look at the main Arctic changes in our Image of the Week.

 

Where are we going?

The SWIPA report highlights that the warming trends in the Arctic will continue, even if drastic greenhouse gas emission cuts are achieved in the near future. For example, mean Arctic autumn and winter temperatures will increase by about 4°C in 2040 compared to the average over 1981-2005 according to model projections (Fig. 5, right panel). This corresponds to twice the increase in projected temperature for the Northern Hemisphere (Fig. 5, left panel).

Fig. 5: Autumn-winter (NDJFM) temperature changes for the Northern Hemisphere (left) and the Arctic only (right) based on 36 global climate models, relative to 1981-2005, for two emission scenarios [Credit: Fig. 2.15 of AMAP (2017)].

This Arctic amplification leads to four main impacts:

  1. The Arctic Ocean could be ice-free in summer by the late 2030s based on extrapolated observation data. This is much earlier than projected by global climate models.

  2. Permafrost extent is projected to decrease substantially during the 21st Century. This would release large amounts of methane in the atmosphere, which is a much more powerful greenhouse gas than carbon dioxide.

  3. Mean precipitation and daily precipitation extremes will increase in a warming Arctic.

  4. Global sea level will continue to rise due to melting from ice sheets and glaciers, ocean thermal expansion, and changes in terrestrial storage. However, uncertainties remain regarding the magnitude of the changes, which is linked to the different emission scenarios and the type of model used.

What are the implications?

A potential economic benefit to the loss of Arctic sea ice, especially in summer, is the creation of new shipping routes and access to untapped oil and gas resources. However, besides this short-term positive aspect of Arctic changes, many socio-economic and environmental drawbacks exist.

The number of hazards has been rising due to Arctic changes, including coastal flooding and erosion, damage to buildings, risks of avalanches and floods from rapid Arctic glacier melting, wildfires, and landslides related to thawing permafrost. Furthermore, Arctic changes (especially sea-ice loss) may also impact the climate at mid-latitudes, although many uncertainties exist regarding these possible links (see Cohen et al., 2014).

What can we do?

The SWIPA report identifies four action steps:

  1. Mitigating climate change by decreasing greenhouse gas emissions. Implementing the Paris Agreement would allow stabilizing the Arctic temperatures at 5-9°C above the 1986-2005 average in the latter half of this century. This would also reduce the associated changes identified on our Image of the Week. However, it is recognized that even if we implement the Paris Agreement, the Arctic environment of 2100 would be substantially different than that of today.

  2. Adapting to impacts caused by Arctic changes.

  3. Advancing our understanding of Arctic changes through international collaboration, exchange of knowledge between scientists and the general public, and engagement with stakeholders.

  4. Raising public awareness by sharing information about Arctic changes.

Further reading

Edited by Scott Watson and 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 – Understanding Antarctic Sea Ice Expansion

Fig. 1: Average monthly Antarctic sea ice extent time series in black, with the small increasing trend in blue. [Credit: NSIDC]

Sea ice is an extremely sensitive indicator of climate change. Arctic sea ice has been dubbed ‘the canary in the coal mine’, due to the observed steady decline in the summer sea ice extent in response to global warming over recent decades (see this and this previous posts). However, the story has not been mirrored at the other pole. As shown in our image of the week (blue line in Fig. 1), Antarctic sea ice has actually been expanding slightly overall!


The net expansion is the result of opposing regional trends

The small increasing trend in Antarctic sea-ice extent is the sum of opposing regional trends (click here for definitions of area, concentration and extent). Sea ice in the Weddell and Ross seas has expanded whereas in the Amundsen and Bellingshausen (A-B) seas the sea-ice cover has diminished (Holland 2014). The size of these trends varies with the seasons (Fig. 2). There are no significant trends in ice concentration – the fraction of a chosen area/grid box that is sea ice covered — if you look at (Southern hemisphere) winter values, however we do see trends when looking at a time series of summer values. The differences in trends between seasons suggests interactions with atmosphere and ocean (feedbacks) that amplify (in the spring) and dampen (in the autumn) changes in the ice cover, creating this seasonality. Some of this variability can be explained by changes in the winds (Holland and Kwok, 2012). But the complexity of the trends can’t be explained by one single change in forcing (e.g. winds, snowfall or temperature) or a single process (e.g. ice albedo feedback acting in the spring/summer).

 

Fig. 2: Seasonal trend in ice concentration. Maximum trends are seen in summer. Large increases are seen in the Weddell and Ross seas, and decreases in the Amundsen and Bellingshausen (A-B) seas. [Credit: Fig 2 from Holland (2014). , reprinted with permission by Wiley and Sons].

Why hasn’t Antarctic sea ice extent been decreasing?

There is no clear consensus on this. In short, we don’t really know… It is not as intuitive as the ‘warmer climate results in less ice’ narrative for the Arctic. We only have a time series of Antarctic sea ice extent from 1979 (the start of satellite observations). We therefore can’t be sure what role natural variability is having on decadal and longer timescales, i.e. if this is just natural ups and downs or an “unusual” trend related to climate change. Another difficulty is that we don’t have a reliable time series of sea ice volume as we have difficulties in getting reliable sea ice thickness measurements, because of the thick snow covering on sea ice in the Southern Ocean. For example, it could be that the ice is becoming thinner although the sea-ice area has increased.

There are important processes and/or feedbacks between sea ice and ocean or between sea ice and atmosphere that we are missing from our models

Currently, global climate models are poor at reproducing the observed Antarctic sea ice changes (Turner et al. 2013). Models simulate a decrease in the overall sea ice extent, instead of the observed increase. They also fail to reproduce the correct spatial variations, as shown in Fig. 2. This makes it very hard to make predictions about future changes in Antarctic sea ice from model results, and implies that there are important processes and/or feedbacks between sea ice and ocean or between sea ice and atmosphere that we are missing from our models, and therefore our understanding of the Southern Ocean climate system is incomplete.

 

However, there are some suggestions as to processes that could explain some of the observed Antarctic sea ice variability. The largely fall into two main categories: natural variability and anthropogenic changes.

 

1.Natural Variability

Natural variability refers to the repeating oscillations and patterns we see in the climate system. Some of these repeating patterns can be correlated with increases/decreases in Antarctic sea ice. In particular El Nino Southern Oscillation (ENSO) and the Southern Annular Mode (SAM) have been linked to Antarctic sea ice changes. The SAM is a measure of the difference in pressure between 40°S and 65°S, a positive SAM indicates a stronger difference in pressure, driving stronger westerly winds around Antarctica, increasing the thermal isolation of Antarctica. Stronger westerlies are associated with cooler sea surface temperatures and expansion of the sea ice cover on short  timescales (seasons to years).

The SAM has been in a mostly positive phase since the mid-1990s, so is believed may have something to do with some of the small increase in sea ice extent we have seen. However, variability on longer time scales (decades or longer) could also explain some of the small increase, but this is tricky to assess without a longer observational time series.

 

2. Anthropogenic Changes

The main two human-induced changes on the Antarctic climate system are the ozone hole and increased melting of the Antarctic ice sheet.

  • Ozone hole
    The ozone hole causes the westerly winds to strengthen, making the sea ice cover expand. However it is more complicated than this, as the impact on the sea ice may depend on what timescale we look at. Over longer timescales (years to decades) the initial response may be outweighed by an increase in ocean upwelling (due to the stronger winds). This brings warm water from below the cold surface layer up to the surface, melting the sea ice from below, eventually resulting in a net sea ice area decrease in response to the ozone hole. See Ferreira et al. (2015) for details.
  • Increased melting of the Antarctic ice sheet
    This could also play a role in the observed sea ice expansion, by increasing the ocean stratification. This results in a cooler and fresher surface layer, favouring the growth of sea ice (Bintanja et al. 2015).

 

It is very tricky to distinguish what is natural variability, what is human induced, or a complicated combination of two.

 

It is very tricky to distinguish what is natural variability, what is human induced, or a complicated combination of two. This means we don’t really know whether the observed large decrease in Antarctic sea ice extent seen in 2016/2017 (read more about it here) is just an anomaly or the start of a decreasing trend. So, in summary Antarctic sea ice is confusing, and we still can’t claim to completely understand observed variability. But this makes it interesting and means there is still a wealth of secrets left to be discovered about Antarctic sea ice!

 

Further reading

 

Edited by Clara Burgard et Sophie Berger


Rebecca Frew is a PhD student at the University of Reading (UK). She investigates the importance of feedbacks between the sea ice, atmosphere and ocean for the Antarctic sea ice cover using a hierarchy of climate models. In particular, she is looking at the how the importance of different feedbacks may vary between different regions of the Southern Ocean.
Contact: r.frew@pgr.reading.ac.uk

Image of the Week – Does size really matter? A story of ice floes and power laws

Figure 1: Sea ice extent in 2014 during the melting season. The pink lines mark the inner and outer extent of the marginal ice zone. The data comes from the CPOM setup of the CICE sea-ice model run with 9 years spin up from 2005. [Credit: Adam Bateson]

The retreating Arctic sea ice is one of the most well-known facets of Climate Change. Images of polar bears desperately swimming through polar seas searching for somewhere to rest and feed resonate strongly with the public. Beyond these headlines however, the Arctic Ocean is displaying a rapid transition from having mostly permanent ice cover to a more seasonal cover.


The Marginal Ice Zone

As both atmospheric and ocean average temperatures increase over the 21st century, the region of the Arctic considered either marginal or seasonal i.e. regions where sea ice is present for at least some of the year but with periods of either no or incomplete sea ice cover, is projected to increase significantly. Our image of the Week (Fig. 1) shows how the Marginal Ice Zone (defined here as regions with 15 % – 80 % ice coverage) evolves through the melting season. This means that the thermodynamic (i.e. melting, freezing) and dynamic (i.e. mechanical) processes which dominate the marginal ice zone are likely to become more important in influencing how the sea ice evolves in future.

Floe size matters

A key parameter to describe the behaviour of this region is the size of the individual ice floes – sheets of floating sea ice – which form the sea ice cover (see also a previous post on this topic). Floe size impacts melt rate, floe mechanical response, atmosphere-ocean momentum exchange and wave-ice interactions (Fig. 2). The sea ice component of climate models usually assumes all floes have the same, constant size; this assumption removes the ability of sea ice models to represent the complexity of the marginal ice zone. As a result processes which influence floe size such as wave induced break up of floes and lateral melting can’t be represented adequately in current climate models.

Figure 2: Video shows significant wave height in 2014 (darker blue colours indicate bigger waves; note also that a white colour indicates no waves, not necessarily sea ice cover). The purple/pink lines mark the inner and outer extent of the Marginal Ice Zone respectively. The data come from the CPOM setup of the CICE sea-ice model run with 9 years spin up from 2005. [Credit: Adam Bateson]

How can we represent different floe sizes in models?

Given the changing Arctic environment, representing floe size as a variable quantity is likely to be important for future accuracy of sea ice modelling. Currently sea ice models tend to divide the Polar Regions into grid cells, with properties defined as an average across the grid cell. However floe sizes can vary significantly over sub kilometre scales. There are four alternative approaches to representing such a non-uniform distribution of floe sizes within a grid cell:

  1. Define floes individually within the model and allow each floe to evolve independently.
  2. Use a categorical floe size distribution i.e. assign floes to size categories of 1 – 10 m, 10 – 20 m etc. (e.g. Horvat et. al, 2015).
  3. Impose a floe size distribution on each grid cell which evolves over time driven by relevant processes such as lateral melting or floe break-up (e.g. Williams et. al, 2013 a & b).
  4. A single floe size for each grid cell is diagnosed from the fractional ice coverage.

 

Option 1 would be the ideal approach from a Physics perspective. It assumes nothing about what form the floe size distribution may take and allows us to properly assess the impact of different processes for floes of different sizes. This approach is computationally expensive however, which means it will take longer for models to run. Option 4 would the simplest option and wouldn’t have negligible impacts on model run times, however it wouldn’t be possible to include in the model any processes which influence floe size. Option 2 and option 3 represent intermediates between these two extremes. In particular option 3 would be a preferred compromise if floes can be represented as a coherent distribution which evolves over time at the grid length scale.

We should now look at whether observations support the use of such a distribution.

The power law distribution

Figure 3: The cumulative number density for floe size can be represented by a power law. Note that both scales are log scales, and that C in the equation is a constant. The blue and green lines show the distribution for smaller and larger exponents respectively. Note the cumulative number density for a given floe diameter, x, is the fraction of floes size x and larger. [Credit: Adam Bateson]

The floe size distribution is most commonly fitted to a power law (Fig. 3). Power law systems have the property of self-similarity, a term attributed to a system which looks the same over different scales (e.g. the metre or kilometre scale). Power law distributions are relatively easily to investigate mathematically, and can easily be incorporated into a model without significant computational expense.

Do observations support the use of a power law?

Many individual experiments to assess the floe size distribution have shown a good fit to the power law. However, a large range of values for the power law exponent have been reported with observations ranging from 0.9 to 4. Other papers have proposed two power laws over different size ranges, with smaller exponents used for the smaller floe range. Herman et. al (2010) proposed that a distribution with a variable exponent would produce a better fit than a power law. There are further questions we need to consider as well. Over what scale are power laws a valid approximation? What determines the exponent of the power law and can it be assumed that this is constant? Is the power law only valid over a certain range of floe sizes and if so what determines this range?

These questions are not trivial, and the available observations are not sufficient to answer them. However, there is still value in testing different distributions and approaches within models. This can provide information about how sensitive the sea ice cover is to different distributions and which processes in particular are important to accurately model winter ice growth and summer ice loss in the marginal ice zone.

References/Further Reading

Edited by Sophie Berger


Adam Bateson is a PhD student at the University of Reading (United Kingdom), working with Danny Feltham. His project involves investigating the fragmentation and melting of the Arctic seasonal sea-ice cover, specifically improving the representation of relevant processes within sea-ice models. In particular he is looking at lateral melting and wave induced fragmentation of sea-ice as drivers of break up, as well as the role of the ocean mixed layer as either an amplifier or dampener to the impacts of particular processes. Contact: a.w.bateson@pgr.reading.ac.uk or @a_w_bateson on twitter.

Image of the Week — Climate change and disappearing ice

The first week of the Climate Change summit in Bonn (COP 23  for those in the know) has been marked by Syria’s decision to sign the Paris Accord, the international agreement that aims at tackling climate change. This decision means that the United States would become the only country outside the agreement if it were to complete the withdrawal process vowed by President Trump.

In this context, it has become a tradition for this blog to use the  United Nations climate talks as an excuse to remind us all of some basic facts about climate change and its effect on the part we are most interested in here: the cryosphere! This year we have decided to showcase a few compelling animations, as we say “a picture is sometimes worth a thousand words”…


Arctic sea ice volume

Daily Arctic sea ice volume is estimated by the PIOMAS reconstruction from 1979-present [Credit: Ed Hawkins]

The volume of Arctic sea ice has declined over the last 4 decades and reached a record low in September 2012. Shrinking sea ice has major consequences on the climate system: by decreasing the albedo of the Arctic surface, by affecting the global ocean circulation, etc.

More information about Arctic sea ice:

Land ice losses in Antarctica and Greenland

Change in land ice mass since 2002 (Right: Greenland, Left: Antarctica). Data is measured by NASA’s Gravity Recovery and Climate Experiment (GRACE) satellites. [Credit: Zack Labe]

Both the Antarctic and Greenland ice sheets have been losing ice since 2002, contributing to global sea-level rise (see previous post about sea level) .  An ice loss of 100 Gt raises the  sea level by ~0.28 mm (see explanations  here).

More information about ice loss from the ice sheets:

 

The cause: CO2 emissions and global warming

Finally we could not close this post without showing  how the concentration of carbon dioxide have evolved  over the same period and how this has led to global warming.

CO₂ concentration and global mean temperature 1958 – present. [Credit:Kevin Pluck]

More information about CO2 and temperature change

  • Global Temperature | NASA: Climate Change and Global Warming
  • Carbon dioxide | NASA: Climate Change and Global Warming

More visualisation resources

Visualisation resources | Climate lab

 

Edited by Clara Burgard

Image of the Week – Sea-ice dynamics for beginners

Image of the Week – Sea-ice dynamics for beginners

When I ask school children or people who only know about sea ice from remote references in the newspapers: ‘How thick do you think is the Arctic sea ice?’, I often get surprising answers: ’10 meters? No, it must be thicker – 100 meters!’. It seems like sea ice, often depicted as a uniform white cover around the North Pole and as a key element in accelerated warming of the Polar Regions, imposes a majestic image. Unfortunately, sea ice is much more fragile.


Growing in the current

Actually, sea ice is on average just about 2 m thick. It used to be thicker, up to 3 m, but such ice needs several winters to grow and is quite rare in the modern-day Arctic as winters are warmer than they used to be (see this previous post). Currently, more than half of the Arctic sea-ice area melts away completely during summer and grows back during the next winter. Such a thin layer of frozen water floating on the ocean is not strong enough to resist the forces of the wind, which pushes it around in ocean surface currents. In order for the ice to move, it has to deform and breaks into ice floes (read more in this previous post). Some ice floes move apart (divergence) in leads and polynyas (see this previous post), while others are pressed together (convergence) in pressure ridges, where blocks of ice pile up against and on top of each other (see our Image of the Week).

Ice grows from the ocean surface layer by water freezing. This is called thermodynamical growth. Thermodynamical growth produces most of the ice forming in the time from freeze-up in fall until the ice becomes about 1 to 2 m thick in mid-winter. At that point, sea ice approaches equilibrium thickness, i.e. the sea ice is thick enough to insulate the cold atmosphere from the relatively warm ocean. But because sea ice deforms, it can continue growing during the rest of the winter too. Pressure ridges sails can stick several meters out of the icy landscape, while their much larger and bulkier keels are hidden below the surface. Ridges can store large volume of sea ice – about a third (Hansen et al, 2013)! At the same time, new ice can grow in leads where open water is exposed to the atmosphere.

The following video is a collection of movies showing consequences and acts of sea-ice deformation. The first part is taken from R/V Lance – the ice-strengthened research ship of Norwegian Polar Institute, while she is navigating along a lead in late winter. Observe how much space is taken by pressure ridges! The second part of the movie shows a pressure ridge growing. Listen to the sound of deforming ice!

Another positive feedback

In winter, temperatures are so low that all the fractures, leads and pressure ridges freeze back – they heal. In summer, however, these damages are the first to appear again. Dark water with low albedo (read more about the albedo feedback in this post) is exposed and the ice melts faster in such regions. Because the Arctic sea ice became relatively thin over the recent decades, it also became less resistant to the forces of the wind. Such thin ice breaks more easily (e.g. Itkin et al, 2017). This means that, as more of such damaged ice is present in summer, the ice cover melts faster. So, here is an additional positive feedback for the Arctic ice under climate change: thinner ice melts faster also because it has become weaker and therefore breaks up easier.

Further reading

Edited by Clara Burgard


Polona Itkin is a Post-doctoral Researcher at the Norwegian Polar Institute, Tromsø. She investigates the sea ice dynamics of the Arctic Ocean and its connection to the sea ice thickness. In her work she combines the information from in-site observations, remote sensing and numerical modeling. Polona is part of the social media project ‘oceanseaiceNPI’ – a group of scientists that communicates their knowledge through social media channels:

Instagram.com/OceanSeaIceNPI, Twitter.com/OceanSeaIceNPI, Facebook.com/OceanSeaIceNPI, contact Email: polona.itkin@npolar.no

Image of the Week – Of comparing oranges and apples in the sea-ice context

Image of the Week – Of comparing oranges and apples in the sea-ice context

In the last fifty years, models and observations have enabled us to better understand sea-ice processes. On the one hand, global climate models have been developed, accounting for the sea-ice component in the climate system. On the other hand, satellite instruments have been developed to monitor the “real” sea-ice evolution. These satellite observations are often used to evaluate climate models. However, lately, doubts have arisen to whether comparing model output to observations is the most reasonable method. Are we sometimes comparing oranges to apples? To discuss this matter, sea-ice modelers and sea-ice observers met over three days in Hamburg earlier this month at the Workshop on improved satellite retrievals of sea-ice concentration and sea-ice thickness for climate applications.


How do we measure past and current seaice changes?

The great advantage of satellites over in-situ measurements is that they measure changes in sea-ice concentration (fraction of ocean covered by sea ice in a given area) and sea-ice thickness in a continuous way with an almost complete spatial coverage of the polar regions (see this previous post) and with a high temporal resolution.
Of course, satellites do not directly measure sea-ice concentration. Rather, they measure the brightness temperature (left part of Fig. 2), which is a measure of the radiation emitted by the Earth’s surface and atmosphere, with passive microwave sensors (such as SSM/I). From this brightness temperature, it is possible to compute sea-ice concentration (see this article for further information). Microwave sensors are used because they can “see” the surface through clouds and during polar night, which is not possible with visible sensors (such as MODIS).

In a similar way, satellites do not directly measure sea-ice thickness. Rather, they measure sea-ice freeboard (right part of Fig. 2), i.e. the height of the ice above the sea surface, with different kinds of sensors (laser altimetry [e.g. ICESat], radar altimetry [CryoSat]). Sea-ice thickness is then retrieved through appropriate algorithms (see e.g. this article).

 

Fig. 2: Satellite measurement techniques that lead to the observed brightness temperatures and sea-ice freeboard [Credit: C. Burgard].

 

Are the satellite retrievals accurate?

Each method of deriving sea-ice concentration or thickness from satellite measurements has its own uncertainties. For example, algorithms to retrieve sea-ice concentration use several assumptions about the state of the atmosphere and surface emissivity. Also, melt ponds and thin ice show up as lower concentration regions. Similarly, different assumptions about snow depth on ice and about sea-ice density impact the retrieved sea-ice thickness. Therefore, the sea-ice variables retrieved from satellite observations may deviate from their actual “real” state.

 

How do we project future sea-ice changes?

In order to project the future sea-ice evolution, different climate models are used (see this article for example). These climate models are usually evaluated against satellite observations in order to assess their performance. While all models present biases compared to observations, it cannot always be concluded that this is necessarily a problem of the models as observations also have uncertainties as previously said. Therefore, the main discussions at the workshop in Hamburg (Fig. 3) were about reducing uncertainties in the comparison between observations and models.

 

How can we better compare satellite observations and models?

The discussion in Hamburg was very lively as modelers and observers exchanged about how they actually reach the results they provide, explaining in detail their models and algorithms. It became rapidly clear that comparing observed sea-ice concentration to modeled sea-ice concentration might be like comparing apples to oranges under certain circumstances.

Thomas Lavergne, a researcher at the Norwegian Meteorological Institute, gave a presentation related to this discussion by presenting the picture that is our Image of the Week. The classical method up to now has always been to transform the measured satellite signal into a “satellite product”, a quantity that is directly computed by models (direction from right to left in the Image of the Week), so that we can compare this quantity to the corresponding model variable. As already mentioned, this can lead to assumptions and introduce errors into the observations, while one would expect observations to be the best representation of the “real” world.

Another possible approach, already well accepted in the community of weather forecasting, is to transform the model variables all the way into simulated brightness temperatures and compare these to satellite data (direction from left to right in the Image of the Week). The algorithms that transform model variables into simulated satellite quantities are observation operators. Although an active field of research, the observation operators for sea ice are not ready, and the comparison of sea-ice simulations to satellite observations will for the foreseeable future rely on satellite “products”.

At the workshop, Lavergne advocated for a middle-ground solution, where satellite products “take a step back” and climate models “take a step forward” using tailored observation operators. This would reduce the need for assumptions in the satellite products but still be manageable for modelers, and would most likely offer the best consensus for the two communities. This way, observed and modeled quantities can still be compared with each other and uncertainties introduced into the comparison can be reduced.

Fig. 3: Happy sea-ice observers and sea-ice modelers in Hamburg [Credit: Julika Doerffer, CEN, Universität Hamburg]

 

Perspectives

The workshop aim was to bring together sea-ice observers and modelers. While no real consensus on the proposed approach was found, the reflection has been launched and probably deserves some more attention in the future in order to better compare sea-ice models and satellite observations. This might move the debate from an apple-to-oranges comparison to a pear-to-pear one. This will hopefully improve sea-ice models and satellite observations and improve future projections of sea-ice evolution.

 

Further reading

With contributions by Thomas Lavergne and Clara Burgard

Edited by Clara Burgard and Sophie Berger


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.