The Arctic's Rapidly Shrinking Sea Ice Cover, a review of Stroeve et al
The Arctic’s rapidly shrinking sea ice cover: a research synthesis Stroeve et al 2011: A Review
The Arctic Ice Mass is a radically different place to how it once was. It was a place that was like a continent of ice in its own right. The two land continents of North America, and Asia were connected by solid ice like a bridge at the times of maximum ice extent. It formed a barrier to sea faring traffic, but the polar bears, whales, seals, all manner of creatures had evolved to this harsh environment. Now the ice is opening up, and summers are not what they used to be.
"The sequence of extreme September sea ice extent minima over the past decade suggest acceleration in the response of the Arctic sea ice cover to external forcing, hastening the ongoing transition towards seasonally open Arctic Ocean. " -- Stroeve.
The maps are wrong. I was misled by maps all my life when I looked at the Arctic Ocean, and thought it was just an ocean. When I looked at the satellite imagery, the water is not the dominant feature of the Arctic. It is the ice in summer and winter. The amount of blue water that showed through the ice was less than 15%. This paper by Stroeve et al reduced its scope to study the forces that work on the sea ice extent minimum, which is in September. The Arctic sea ice extent from 1979 to 2011, showed downward trends in all months. This confirmed the findings of Cavalieri. Stroeve and Cavalieri both report increased seasonality, and variability, and reduced predictability, of sea ice loss. The ice extent minimum in September showed the biggest loss compared to winter, which had the smallest sea ice loss. Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report (AR4) incorporates a few different coupled global climate models, and also confirms the reports of Stroeve and Cavalieri, that the Arctic sea ice is declining.
Figure 1 was a coloured Arctic map. It showed the monthly sea ice concentration for September 2010 with the minimum sea ice extents of the years 2007, 2008, and 2009, overlaid. They were each less than the average monthly sea ice extent for September over a time series from 1979-2010.
Figure 2 was a plot of the values of the Average Monthly Arctic Sea Ice Extent for September over the time series of 1979 to 2010. It showed a negative trend in sea ice area extent meaning the sea ice had declined. They said that "ice-free summers might be realized as early as 2030." An ice free summer has been defined to mean not being actually completely ice free, but with less than 1 million square kilometres of sea ice. It is not expected that the very thick ice around the Canadian Arctic Archipelago would melt by 2030.
The Fragmented Ice Cover.
A long time ago before the invention of the satellites, the polar bears had evolved to be able to live on the Arctic Ice Mass. The ice was effectively a bridge in the winter at the maximum ice extent where the bears were able to travel from the continents onto the ice, and stay there for a period of time. Now, the multiyear floes are coming apart. This weakens the structure of the whole ice mass. The behaviour of the bears is changing as a consequence. They are becoming more cautious of the rotten ice.
"The Arctic has warmed in all seasons leading to a reduced likelihood of unusually cold conditions that could bring about temporary recovery through natural climate variability." -- Stroeve.
This climate region is extremely variable. It is the most sensitive of all the geographical regions to changes in atmospheric temperature. Stroeve reported being able to quantify changes in the spring ice age distribution as a proxy for ice thickness by using the remote sensing. Figure 2 was a feedback model diagram illustrating mutually supporting processes favouring acceleration of the September sea ice trend. These processes favoured or enhanced sea ice decline.
Then they reported how the distribution of the ice has changed. There was less of the older ice floes, in proportion to the younger ice. The ice age was derived from an algorithm reported by Fowler 2004, and the motion of the ice was also documented by Fowler.
Was there an accelerating decline in the September minimum sea ice extent?
Overall there could be shown an increased variability in sea ice decline, that it did not occur uniformly over the year. More ice declined in September than in the winter. This supported the findings of Cavalieri.
The Arctic as a system.
Then they discussed the linked processes in the Arctic. This was about how the atmosphere, and the ocean currents combined forces to influence the behaviour of the sea ice, to freeze or melt. Cyclonic winds over the Beaufort and Chukchi seas likely favoured ice divergence (Ogi and Wallace 2007) helping slow the seasonal decline in ice extent through August and September.
From the years 2007 through 2010, ice minimum of 2007 occurred as a combination of factors, of the already shrinking ice cover and an anomalous atmospheric sea level pressure event called the Summer Arctic Dipole Anomaly (DA). This was were high pressure over water and low pressure over land, combined forces to transport ice to the North Pole instead of flushing it out of the Arctic Ocean. There were many low pressure systems entering the central Arctic that caused the ice loss to slow as well as cloud patterns that favoured the retention of ice.
Figure 9 showed total cloud cover over the Arctic for summers from 2007 to 2010.
"At no other time during the satellite data record was there a total ice loss from maxiumum to minimum that exceeded 10 million square kilometres." -- Stroeve
"The obvious feature for July is above-average daily retreats [of the sea ice] in all [Arctic] sectors for the last 4 years. " -- Stroeve
Polynas were opened up. A polyna is a patch of open ocean surrounded by ice. This was observed in the remote sensing. These polynas enhanced melting. Under different conditions of freezing air, sea ice will form in a polyna. Normally as the summer solstice passes, insolation decreases as the days get shorter, and the nights longer. This usually causing the building of heat in the system to slow down and decrease, even though the system is still building to the minimum ice extent. This feedback effects the ice melt rate to slow down, but in 2008 this did not happen.
Then they talked about the "memory" of the Arctic system. This referred to the conditions that led to the system being in a particular state that were likely to influence the same state in the system. If it was cold, it will be cold again. If it was warmer, it will be warmer again. The conditions, or the configuration of the actors in the system lead to a particular system state. They described a two state climate regime of warm and cool, for further discussing the conditions in the Arctic.
Is the Arctic system slowing in recovery from perturbation? It's a bit like a person as they get older. It takes longer for them to recover from a shock or injury or illness. They defined an increase in lag-1 autocorrelation to be interpreted as slowness of recovery. There was uncertainty around the role of the ice divergence on melt during this season. Ice divergence can promote ice extent, but it also accelerate ice melt.
Figure 11 was a plot of Year vs Correlation of de-trended 20-year segments of related ice thickness and ice extent. This graph represented the lag-1 autocorrelation and showed that in a predictive model over a 100 year timeline, the system could be slowing down to a rapid transition state where the system state it is transitioning to the ice free Arctic. The summer atmospheric patterns favouring summer ice loss became more effective, and the patterns that favoured ice retention became less effective.
The oceans have changed. The way that water from the other oceans came into the Arctic Ocean have different effects to what has been documented in the past. The waters coming into the Arctic Ocean were helping to melt the ice. This was indicated by the subduction of warm water from the Atlantic under the colder Arctic water through the Fram Strait.
"Stronger near surface stratification from increasing ice melt stores the heat in the Near Suface Temperature Maximum NSTM for longer periods of time." -- Stroeve. This made the Arctic Ocean warmer for longer, and less able to retain ice.
So this paper did not actually discuss when we are likely to see our first ice free summer. It discussed how this record September sea ice minimum of 2007 had come about, and how unstable the Arctic has become. What the consequences could be for the Arctic, for any kind of recovery of the ice that is diminishing into unreliability. We can't count on it anymore. The shortness of the satellite era time series does not recount or include the stories of the Indigenous people of the ice. People live in the Arctic, the Inuit and the Aleuts, as examples. They have lived there for thousands of years, and their historical observations of ice are codified into their rituals and behaviour. We can learn about the ice from how the Polar Bears have evolved their hunting patterns.
As a thought experiment, if the Polar Bears were evolving now, what would they be doing differently? The hunting patterns that they exhibit now are unlikely to be evolving under the current Arctic conditions.
Basically, Stroeve et al confirmed the Arctic ice diminishing overall, that is in the Cavalieri conclusion, and more about the consequences for short term or long term recovery of the sea ice. They are not favourable to the recovery of the ice.
All references in this blog article can be found in Stroeve's paper.
The Arctic Ice Mass is a radically different place to how it once was. It was a place that was like a continent of ice in its own right. The two land continents of North America, and Asia were connected by solid ice like a bridge at the times of maximum ice extent. It formed a barrier to sea faring traffic, but the polar bears, whales, seals, all manner of creatures had evolved to this harsh environment. Now the ice is opening up, and summers are not what they used to be.
"The sequence of extreme September sea ice extent minima over the past decade suggest acceleration in the response of the Arctic sea ice cover to external forcing, hastening the ongoing transition towards seasonally open Arctic Ocean. " -- Stroeve.
The maps are wrong. I was misled by maps all my life when I looked at the Arctic Ocean, and thought it was just an ocean. When I looked at the satellite imagery, the water is not the dominant feature of the Arctic. It is the ice in summer and winter. The amount of blue water that showed through the ice was less than 15%. This paper by Stroeve et al reduced its scope to study the forces that work on the sea ice extent minimum, which is in September. The Arctic sea ice extent from 1979 to 2011, showed downward trends in all months. This confirmed the findings of Cavalieri. Stroeve and Cavalieri both report increased seasonality, and variability, and reduced predictability, of sea ice loss. The ice extent minimum in September showed the biggest loss compared to winter, which had the smallest sea ice loss. Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report (AR4) incorporates a few different coupled global climate models, and also confirms the reports of Stroeve and Cavalieri, that the Arctic sea ice is declining.
Figure 1 was a coloured Arctic map. It showed the monthly sea ice concentration for September 2010 with the minimum sea ice extents of the years 2007, 2008, and 2009, overlaid. They were each less than the average monthly sea ice extent for September over a time series from 1979-2010.
Figure 2 was a plot of the values of the Average Monthly Arctic Sea Ice Extent for September over the time series of 1979 to 2010. It showed a negative trend in sea ice area extent meaning the sea ice had declined. They said that "ice-free summers might be realized as early as 2030." An ice free summer has been defined to mean not being actually completely ice free, but with less than 1 million square kilometres of sea ice. It is not expected that the very thick ice around the Canadian Arctic Archipelago would melt by 2030.
The Fragmented Ice Cover.
A long time ago before the invention of the satellites, the polar bears had evolved to be able to live on the Arctic Ice Mass. The ice was effectively a bridge in the winter at the maximum ice extent where the bears were able to travel from the continents onto the ice, and stay there for a period of time. Now, the multiyear floes are coming apart. This weakens the structure of the whole ice mass. The behaviour of the bears is changing as a consequence. They are becoming more cautious of the rotten ice.
"The Arctic has warmed in all seasons leading to a reduced likelihood of unusually cold conditions that could bring about temporary recovery through natural climate variability." -- Stroeve.
This climate region is extremely variable. It is the most sensitive of all the geographical regions to changes in atmospheric temperature. Stroeve reported being able to quantify changes in the spring ice age distribution as a proxy for ice thickness by using the remote sensing. Figure 2 was a feedback model diagram illustrating mutually supporting processes favouring acceleration of the September sea ice trend. These processes favoured or enhanced sea ice decline.
Then they reported how the distribution of the ice has changed. There was less of the older ice floes, in proportion to the younger ice. The ice age was derived from an algorithm reported by Fowler 2004, and the motion of the ice was also documented by Fowler.
Was there an accelerating decline in the September minimum sea ice extent?
Overall there could be shown an increased variability in sea ice decline, that it did not occur uniformly over the year. More ice declined in September than in the winter. This supported the findings of Cavalieri.
The Arctic as a system.
Then they discussed the linked processes in the Arctic. This was about how the atmosphere, and the ocean currents combined forces to influence the behaviour of the sea ice, to freeze or melt. Cyclonic winds over the Beaufort and Chukchi seas likely favoured ice divergence (Ogi and Wallace 2007) helping slow the seasonal decline in ice extent through August and September.
From the years 2007 through 2010, ice minimum of 2007 occurred as a combination of factors, of the already shrinking ice cover and an anomalous atmospheric sea level pressure event called the Summer Arctic Dipole Anomaly (DA). This was were high pressure over water and low pressure over land, combined forces to transport ice to the North Pole instead of flushing it out of the Arctic Ocean. There were many low pressure systems entering the central Arctic that caused the ice loss to slow as well as cloud patterns that favoured the retention of ice.
Figure 9 showed total cloud cover over the Arctic for summers from 2007 to 2010.
"At no other time during the satellite data record was there a total ice loss from maxiumum to minimum that exceeded 10 million square kilometres." -- Stroeve
"The obvious feature for July is above-average daily retreats [of the sea ice] in all [Arctic] sectors for the last 4 years. " -- Stroeve
Polynas were opened up. A polyna is a patch of open ocean surrounded by ice. This was observed in the remote sensing. These polynas enhanced melting. Under different conditions of freezing air, sea ice will form in a polyna. Normally as the summer solstice passes, insolation decreases as the days get shorter, and the nights longer. This usually causing the building of heat in the system to slow down and decrease, even though the system is still building to the minimum ice extent. This feedback effects the ice melt rate to slow down, but in 2008 this did not happen.
Then they talked about the "memory" of the Arctic system. This referred to the conditions that led to the system being in a particular state that were likely to influence the same state in the system. If it was cold, it will be cold again. If it was warmer, it will be warmer again. The conditions, or the configuration of the actors in the system lead to a particular system state. They described a two state climate regime of warm and cool, for further discussing the conditions in the Arctic.
Is the Arctic system slowing in recovery from perturbation? It's a bit like a person as they get older. It takes longer for them to recover from a shock or injury or illness. They defined an increase in lag-1 autocorrelation to be interpreted as slowness of recovery. There was uncertainty around the role of the ice divergence on melt during this season. Ice divergence can promote ice extent, but it also accelerate ice melt.
Figure 11 was a plot of Year vs Correlation of de-trended 20-year segments of related ice thickness and ice extent. This graph represented the lag-1 autocorrelation and showed that in a predictive model over a 100 year timeline, the system could be slowing down to a rapid transition state where the system state it is transitioning to the ice free Arctic. The summer atmospheric patterns favouring summer ice loss became more effective, and the patterns that favoured ice retention became less effective.
The oceans have changed. The way that water from the other oceans came into the Arctic Ocean have different effects to what has been documented in the past. The waters coming into the Arctic Ocean were helping to melt the ice. This was indicated by the subduction of warm water from the Atlantic under the colder Arctic water through the Fram Strait.
"Stronger near surface stratification from increasing ice melt stores the heat in the Near Suface Temperature Maximum NSTM for longer periods of time." -- Stroeve. This made the Arctic Ocean warmer for longer, and less able to retain ice.
So this paper did not actually discuss when we are likely to see our first ice free summer. It discussed how this record September sea ice minimum of 2007 had come about, and how unstable the Arctic has become. What the consequences could be for the Arctic, for any kind of recovery of the ice that is diminishing into unreliability. We can't count on it anymore. The shortness of the satellite era time series does not recount or include the stories of the Indigenous people of the ice. People live in the Arctic, the Inuit and the Aleuts, as examples. They have lived there for thousands of years, and their historical observations of ice are codified into their rituals and behaviour. We can learn about the ice from how the Polar Bears have evolved their hunting patterns.
As a thought experiment, if the Polar Bears were evolving now, what would they be doing differently? The hunting patterns that they exhibit now are unlikely to be evolving under the current Arctic conditions.
Basically, Stroeve et al confirmed the Arctic ice diminishing overall, that is in the Cavalieri conclusion, and more about the consequences for short term or long term recovery of the sea ice. They are not favourable to the recovery of the ice.
All references in this blog article can be found in Stroeve's paper.