Dec 2018

The Beginning of the End for Arctic Sea Ice?


The retreat of Arctic sea ice is one of the clearest indicators that we have that climate change is happening, and in recent years scientists have identified a direct link between rising carbon dioxide levels in the Earth’s atmosphere driven largely by human activity, and falling levels of Arctic sea ice. After news earlier this year that the strongest, thickest sea ice in the Arctic had begun to break up, it is reasonable to ask whether we are a now witnessing the beginning of the end for Arctic sea ice in summertime.  

What happened in the Last Ice Area in 2018?

The so-called ‘Last Ice Area’ (a term coined by WWF and used by Canadian Prime Minister Justin Trudeau in his 2016 Arctic Commitments) is an area of the Arctic above northern Canada and Greenland where scientific projections show summer sea ice will last the longest. One reason for this is that the prevailing ocean and wind currents in the Arctic drive the sea ice against the coast of northern Canada and Greenland, compacting it and making it more resilient to higher ambient temperatures. Local Inuit Elders call this region “Similijuaq”, meaning “place of the big ice”. The area is globally significant as it may end up as the last refuge for ice-dependent species – such as the polar bear – as the world warms.

‘The Last Ice Area’ (source: WWF)

In August, news broke that the Last Ice Area was beginning to break up, following a spike of above-average air temperatures (during an otherwise colder than usual summer). This weakened the fast sea ice along Greenland’s north coast. Strong winds driving in from the west were then able to push the ice away from the coast very quickly creating a body of open water off Cape Morris Jesup, the northernmost point in Greenland (and home to the northernmost weather station in the world). The phenomenon has been witnessed in summertime before but nowhere near as fast or as extensively.

One of the reasons for the scale of the event this summer was likely because of a similar phenomenon occurring back in February, which Dr Ruth Mottram described as “truly shocking” because that had never been recorded before. For an unusually long period
(between 17 and 21 February) air temperatures spiked above freezing reaching as high as 30° above average. While the UK suffered from the ‘Beast from the East’, at the other end of the North Atlantic ‘see-saw’, warm winds were pushing up over the north of Greenland. The ice that froze after the February melt would have been much thinner than usual, precipitating the August break-up.  

2018 in context

Despite the shocking nature of the sea ice break up in the Last Ice Area, Professor Mark Brandon emphasised the need to look at what had happened in context. Looking at 2018 in isolation actually tells us very little about what the trends are in Arctic sea ice – and indeed whether 2018 really did mark the beginning of the end for summertime sea ice.

Fortunately, sea ice extent across the whole Arctic is relatively easy to measure. Satellites have been recording Arctic sea ice extent since 1979, meaning scientists now have nearly 40 years of data to consider. The standard approach to studying a changing climate is to consider a 30-year time frame. Arctic sea ice extent in 2018 has largely tracked below the mean for the period 1981-2010. One of the most significant changes over this period was to the difference between summer and winter sea ice extent. While sea ice extent in both winter and summer has decreased, the rates are different by season and summertime melt has become more extensive over the data record.  

However, Professor Brandon argues that relying on the 30-year mean does not make much sense because the extent has been decreasing over virtually the entire record. A mean over this record can obscure how the geography of the Arctic sea ice has changed, and continues to change. To show how the sea ice has changed more clearly one can compare how the average sea ice extent on any particular day at the start of the satellite record for the period 1989-193 compares with now. With such a comparison, what comes across strongly is that the biggest changes to the Arctic sea ice over the satellite record have been seen in the Barents and Kara Seas, the Chukchi Sea and Hudson Bay (an important polar bear habitat). It also captures events like the ones witnessed in 2018 when, for a few days in February, and a few days in August, the sea ice in the Last Ice Area diminished substantially.

On the left is the sea ice extent on 2 December 2018. On the right the difference between the sea ice extent on 2 December 2018 and the average sea ice extent from 1989-2018 for the same day. Red colours indicated a decrease in sea ice extent, blue colours an increase (Courtesy of Mark Brandon).

Even so, the fact that this event happened, while unprecedented, does not change the fundamental conclusion that scientists have been drawing for the past two decades that the Arctic sea ice is likely in terminal decline, with reliably sea-ice free summers expected from the 2030s onwards. While 2018 has made headlines, the beginning of the end for Arctic sea ice in summertime likely started decades ago.

The Arctic in 3D

When examining the state of the Arctic sea ice, scientists have typically relied on the satellite record of sea ice extent produced since 1979. Only recently have scientists started using satellites to measure sea ice thickness. Radar satellites, like the European Space Agency’s CryoSat-2 (operational since 2010), are used to measure the elevation of the sea ice and the ocean so that scientists can find the difference and estimate thickness. Creating a 3D view of the Arctic sea ice is beginning to give scientists a much clearer idea of how much sea ice there is in the Arctic, and how quickly it is changing.  

As Dr Rachel Tilling explained, one of the main reasons why a 3D view of the Arctic sea ice is so valuable is because the sea ice is essentially an insulation blanket that sits on top of the Arctic Ocean. Around the world, a constant exchange (of heat, moisture, salinity etc.) is taking place between the oceans and the atmosphere. The presence of sea ice mediates this exchange in the Arctic and so its thickness really does matter to prevailing ocean and atmospheric conditions. The satellite data gathered over the past eight years is not enough to identify trends, but it has allowed scientists to start to generate a picture of what might be happening. Combining the data with older numerical models (based on sea ice extent) has led some to estimate that since 1980, an average of -290km3 of sea ice has been lost every year, equivalent to two and a half thousand double-decker buses a minute.  

Among the most interesting things in the Arctic sea ice thickness data gathered so far are the outliers, such as the one seen at the end of the summer of 2013, when the sea ice volume was 33% larger than average over the CryoSat-2 period (something that could not have been spotted in measurements of Arctic sea ice extent alone). Scientists looking into the potential cause of this surprising discovery found a strong correlation between the length of the melt season and the volume of ice melt. In 2013, the melt season lasted a week less than usual, but that was enough to produce a significant difference in the volume of sea ice left at the end of the summer. Dr Tilling thinks that similar events could happen in the future, particularly as weather conditions in the Arctic become more variable. In other words, from year to year we might see significant increases or significant decreases in the Arctic sea ice volume that deviate far from what is considered average, and which are not likely to be evident from the monitoring of sea ice extent alone. On longer-term timescales, as temperatures in the Arctic continue to rise, sea ice volume will decline further.  

Dr Tilling argued that the value of the sea ice volume data from satellites cannot be overstated. This ‘thickness’ data is increasingly replacing the ‘extent’ data in sea ice models, which often factor into climate models, producing improvements in the predictions made by organisations including the US Navy and the Met Office. Multiple satellites are now in place, but these are expensive missions to run and generally require extensive international cooperation to make them possible.  

On the ice in the High Arctic

In the summer of 2018, Dr Helen Czerski took part in a major international scientific expedition to the High Arctic, aboard the Swedish icebreaker, Oden. During the trip, she saw first-hand what the ice conditions were like at the North Pole. This was important, she argued, because we spend so much time thinking about what the Arctic is going to look like in the future that we often neglect to look at how it looks today and why it matters in the here and now.  

The icebreaker that Dr Czerski was travelling on moored up against an ice floe near the North Pole to give scientists an incredibly rare opportunity to go onto the ice and take “on-the-ground” measurements of what was going on. As the photograph below shows, near the North Pole in the summer, there were big gaps in the ice, melt ponds all over, and the whole area was constantly shifting and changing. However, this type of field assessment is incredibly difficult to do on a regular basis because accessing the High Arctic is so difficult and expensive. More work is needed to understand the links between the details of ground measurements (for example, snow data and ice shape) and the satellite data. The need for a detailed understanding of small-scale processes at the ice-ocean boundary means that field studies are still essential.


The summer of 2018 was not a tipping point for Arctic sea ice. Rather, it seems that tipping point occurred decades ago. The melt events witnessed in the Last Ice Area in February and August were dramatic but on their own do not signal a state change in the Arctic. However, recorded changes in sea ice extent over recent decades show that some parts of the Arctic – notably the Barents, Kara and Chukchi Seas, and Hudson Bay, are changing significantly, suggesting the end of summer Arctic sea ice is in sight, perhaps within as a little as a decade from now. A 3D view of the Arctic, derived from satellite measurements of sea ice thickness appears to confirm these trends, but have also highlighted how small changes in the prevailing weather and climate conditions can have large consequences for the amount of sea ice in the Arctic in any given year. Scientific expeditions like the one conducted aboard the Oden in the summer of 2018 are essential for assessing conditions on-the-ground in order to help verify what the satellites are recording.