12 June 2023

Dramatic Arctic sea-ice news should not be a shock: We were warned.

by David Spratt

It's almost unthinkable. The Arctic Ocean blue all over in summer, with none of the eight million square kilometres of sea-ice — a thin frozen white crust floating on the ocean surface — that covered it in summer just 40 years ago.

No wonder it made headlines this month when researchers found, as reported by The Washington Post, that "a summer in which the Arctic Ocean features almost entirely open water could be coming even sooner than expected and may become a regular event within most of our lifetimes".

The research is "Observationally-constrained projections of an ice-free Arctic even under a low emission scenario", and was published in Nature on 6 June 2023.   It projected "an ice-free Arctic in September under all [emission] scenarios considered", including low greenhouse gas emission scenarios. In other words, even if  emissions are sharply reduced, the Arctic will be ice-free at the end of the northern summer in September in coming decades.

If emissions decline only slowly or continue to rise (as looks likely), then the first ice-free summer could be in the 2030s, a decade earlier than projections reported by the IPCC. 

Observations of annual average Arctic sea ice extent
for the period 1900 to 2008. The gray shading indicates
less confidence in the data before 1953. 
Source: US Global Change Research Program 2009.
Arctic sea-ice fluxes from a greater extent in the cold, dark winter months to a smaller extent in the warmer, 24-hours-a-day light of the northern summer. The annual average extent was relatively stable in the first half of the 20th century, but started to steadily decrease from the 1950s, with a dramatic collapse in 2007 (see chart at right), which I will return to shortly.

The new research is a shock to many, because it is now clearer that the Arctic, and the global climate system it influences, will suffer almost unimaginable change.  One consequence of a sea-ice-free Arctic summer is the large amount of additional heat in the region as reflective ice is replaced by the heat-absorbing, dark ocean surface. 

Greenland is already passed a tipping point for accelerating ice-mass loss, but that will speed up in sea-ice-free conditions.  And that means not only a faster rate of sea-level rise, but an increasing flow of fresh cold water into the north Atlantic which will further slow the Atlantic Meridional Overturning Circulation (AMOC, sometimes colloquially called the Gulf Stream, which transport equatorial ocean heat up the north America coast and to Europe).

Arctic melt has pushed the Jet Stream into
a more meandering, S-shape pattern, dragging
down and stalling cold and wet conditions
over Europe

And then there is the destabilisation of the Jet Stream, which has and will increasingly result in more and more extreme sub-Arctic climate events. This cascade of system-level climate consequences was outlined in our recent report, Hotter, Higher, Faster and in this blog

In September 2007, there was a calamitous drop in the summer Arctic sea-ice extent, such that one well-respected glaciologist responded at the time in shock that the sea-ice appeared to be shrinking "100 years ahead of schedule".

At the time I was transfixed by the event, because the implications changed everything. It was clear that the sea-ice had no hope of existing in summer at 2°C of warming, which at that time was the climate policy-makers' target. The evidence was that its tipping point had already passed, and what would be safe for the Arctic and its ecosystem was warming of less than 0.5°C, a measure which has already been exceeded. The policy-making paradigm was being turned upside down, but few wanted to notice.

The Arctic need to be cooled, urgently, but who wanted that conversation? The case for cooling is even more compelling today.

This was early in my days of writing about climate, and I decided to put down on paper what was happening, what the very good science journalists at that time were saying, and of course what scientists were, and had previously, said about sea-ice and Greenland.  The result was a short report, "The Big Melt: Lessons from the Arctic summer of 2007" of 16 pages. 

Within a few days, it had been downloaded 20,000 times. This was totally unexpected. The dramatic Arctic story was capturing peoples' interest, and there was a bigger story to tell, too. 

That understanding led to a decision to write a longer report on other, under-reported, aspects of unexpected and non-linear climate change; a project which within months morphed into the book "Climate Code Red: The case for emergency action", with Philip Sutton. 

When the news stories were published last week about the new sea-ice projections, I sat down and re-read "The Big Melt" for the first time in 15 years.  What is remarkable is that the cause of the shock today was well understood a decade-and-a-half ago, if one bothered to find out or listen to scientists. 

Take this example which has proven to be right on the money: "In December 2006, data was presented to a American Geophysical Union conference suggesting that the Arctic may be free of all summer ice by as early as 2030 and likely by 2040."

This event has been coming for a long time, and the alarm bells have been ringing loudly since at least 2006-7. We ignore history at our peril.

So, below, I have republished the relevant sections from "The Big Melt"

"The Big Melt" (2007) Extract

The accelerating loss of the Arctic ice sheet 

"We are all used to talking about these impacts coming in the lifetimes of our children and grandchildren. Now we know that it's us." – Professor Martin Parry, co-chairman of the IPCC impacts working group (Adam, 2007b) 

Events in the Arctic in the northern summer of 2007 have profound consequences for climate policy, the credibility of the IPCC, the assessment of projected sea-level rises and the question as to whether we may have already passed one or more of the critical "tipping points" for dangerous anthropogenic interference. 

In its 2007 Fourth Assessment Report, the IPCC said that "Arctic sea ice is responding sensitively to global warming. While changes in winter sea ice cover are moderate, late summer sea ice is projected to disappear almost completely towards the end of the 21st century" (IPCC, 2007a: 776). 

But even before they were drafted, the 2007 IPCC projections were well behind the physical reality in the environment. In late 2005, Tore Furevik of the Geophysical Institute in Bergen had graphically demonstrated that "the recent [Arctic] sea-ice retreat is larger than in any of the (19) IPCC models" (Furevik, 2005). In December 2006, data was presented to a American Geophysical Union conference suggesting that the Arctic may be free of all summer ice by as early as 2030 and likely by 2040 (Holland, Bitz et. al., 2006) – setting up "a positive feedback loop with dramatic implications for the entire Arctic region" (Amos, 2006). 

This was affirmed by studies published in March and May 2007 (Serreze, Holland et al., 2007; Stroeve, Holland, et al., 2007) which led Penn State climatologist Richard Alley to comment that the ice sheets appear to be shrinking "100 years ahead of schedule" (Spotts, 2006). 

Image page from "The Big Melt"
Despite the warnings, experts were "shocked" at the extent of Arctic ice-sheet loss during the 2007 northern summer; Mark Serreze, an Arctic specialist at the US National Snow and Ice Data Centre (NSIDC) at Colorado University in Denver, told the Guardian: "It's amazing. It's simply fallen off a cliff and we're still losing ice" (Adam, 2007a). 

The 2007 sea ice minimum on 16 September was 4.13 million square kilometers, compared to the previous record low of 5.32 million square kilometers in 2005, representing a precipitous decline of 22 per cent in two years: "The minimum for 2007 shatters the previous five-day minimum set on September 20–21, 2005, by 1.19 million square kilometers (460,000 square miles), roughly the size of Texas and California combined, or nearly five United Kingdoms" (NSIDC, 2007). This loss of ice extent of more than 20 per cent in two years compares to the decreasing trend in ice area of 7 per cent per decade between 1979 and 2005 (Alley, 2007). The ice retreat is likely to be even bigger next summer because this winter's freeze is starting from such a huge ice deficit (Revkin, 2007). 

NSIDC research scientist Walt Meier said it was "the biggest drop from a previous record that we've ever had and it's really quite astounding... Certainly we've been on a downward trend for the last 30 years or so, but this is really accelerating the trend" (McCarthy, 2007). As well, large areas of the Arctic sea ice are now only one metre deep, which means the thickness of the ice has halved since 2001 (Bjornes, 2007) and down from a thickness of 3.5 metres in the early 1960s, and around about 2.5 metres in the late 1980s and early 1990s (Blakemore and Sandell, 2006). The decrease in both extent and thickness suggests that the summer sea ice has lost more than 80 per cent of its volume in 40 years. When the sea ice thins to around half a metre in thickness, it will be subject to even more rapid disintegration by wave and wind action. 

Serreze says we may have already reached the tipping point when there is a rapid sea ice disintegration: "The big question is whether we are already there or whether the tipping point is still 10 or 20 years in the future... my guts are telling me we may well be there now" (Connor 2007b) and "an educated guess right now would be 2030" for the transition to an ice-free Arctic summer (McCarthy, 2007). His colleague at Colorado, Ted Scambos, agrees that "that 2030 is not unreasonable... I would not rule out 2020, given non-linearity and feedbacks" (Scambos, 2007). These views are supported by Ron Lindsay of the University of Washington: "Our hypothesis is that we've reached the tipping point. For sea ice, the positive feedback is that increased summer melt means decreased winter growth and then even more melting the next summer, and so on" (Connor and McCarthy, 2006). 

Australian-of-the-Year Tim Flannery suggests that "at the trajectory set by the new rate of melt, however, there will be no Arctic icecap in the next five to 15 years" (Flannery, 2006). Dr Wieslaw Maslowski of the Naval Postgraduate School in California, whose research focuses on modelling the processes of Arctic sea ice loss, projects a blue Arctic Ocean free of sea ice by the summer of 2013 (Revkin, 2007), the main reason being that the modeled thickness and volume appear to be decreasing at a much faster rate than the satellite derived ice extent (Maslowski, 2007). Maslowski's work suggests the sea ice is significantly being thinned by the effect of warming seas beneath, not just higher air temperatures. 

"The reason so much (of the Arctic ice) went suddenly is that it is hitting a tipping point that we have been warning about for the past few years," says NASA's James Hansen; Germany's Potsdam Institute for Climate Impact Research says Arctic sea ice has "already tipped"; while Paal Prestrud of Oslo's Center for International Climate and Environmental Research says "I'd say we are reaching a tipping point or are past it for the ice. This is a strong indication that there is an amplifying mechanism here" (Doyle, 2007). 

The central point is that Arctic is now irreversibly headed to total summer sea ice loss very quickly, way beyond the expectation of the IPCC whose Arctic scenarios are no longer credible [see figure], and of most scientists' views only two-to-three years ago. It is an instance of the non-linearity in climate systems that should reinforce the need for strict adherence to the precautionary principle in assessing what is likely to constitute dangerous human interference, and how we should respond in constructing emission scenarios and policies to avoid it. 

Stability of the Greenland ice sheet 

Global warming so far has been greatest in the high latitudes of the northern hemisphere, particularly in the sub-Arctic boreal forests of Siberia and North America (ACIA 2005). Arctic temperatures will rise much more quickly than the global average: for a global warming of 2°C, the area-mean annual temperature increase over the Arctic (60-90°N) is likely to be between 3.2° and 6.6°C (0.45° to 0.75°C per decade, and possibly even as large as 1.55°C per decade) (New, 2006). 

The view that a 2°C global temperature increase will be hard to avoid is widespread: from Nicholas Stern (Stern, 2006a: 4) to the co-chair of the IPCC's impacts working group, Martin Parry (Adam, 2007b). But well before two degrees average global warming, a high momentum melting of much of the Greenland ice sheet will be underway (Hansen, 2005). Greenland's critical melt threshold is a regional temperature rise of 2.7 degrees (Gregory, Huybrechts et al, 2004), but with its temperature increase at least 2.2 times the global average (Chylek and Lohmann, 2005), that point will have been triggered at just over a one degree global rise. [Nevertheless, the 2001 IPCC report thought that neither Greenland nor Antarctica would lose significant mass by 2100.] 

The loss of the Arctic sea ice "100 years ahead of schedule" raises two questions of significance about the Greenland ice sheet: what will the effect on the timing of the Greenland tipping point be; and what will be effect on the rate of ice loss from Greenland (which if fully achieved would raise the global sea level by 5–7 metres)? 

Rising Arctic regional temperatures resulting from sea ice loss and the albedo effect (white reflective ice replaced by dark, heat-absorbing sea) are already at "the threshold beyond which glaciologists think the (Greenland) ice sheet may be doomed"; this accelerated melting "is caused by meltwater penetrating crevasses and lubricating the glaciers' flow... The ice is in effect sliding into the ocean on rivers of water," an effect not included in models of the effect of global warming on the Arctic (New Scientist, 2006). A recent study found that the Greenland ice cap "may be melting three times faster than indicated by previous measurements" and that "the mass loss is increasing with time" (Young, 2006). Greenland experienced more days of melting snow in 2006 than the island had averaged over recent decades (Saupe, 2007), the edges of the ice-sheet are melting up to 10 times more rapidly than earlier research had indicated, and the ice sheet height is falling by up to 10 metres a year (Shukman, 2007). As well, the Greenland ice cap is melting so quickly that it is triggering earthquakes as pieces of ice several cubic kilometres in size break off, with "a massive acceleration of the speed with which these glaciers are moving into the sea" (Brown, 2007). 

James Hansen notes that "Ice sheet disintegration starts slowly but multiple positive feedbacks can lead to rapid non-linear collapse" and than "equilibrium sea level rise for ~3°C warming _(25±10 m = 80 feet) implies the potential for us to lose control" because "we cannot tie a rope around a collapsing ice sheet" (Hansen 2006a, Hansen 2006b). 

At this point there is a methodological problem: climate scientists have had difficulty modelling _ice- sheet streams and dynamics (Oppenheimer & Alley, 2004). Robert Corell, a US-based Arctic scientist and member of the IPCC says of Greenland: "Nobody knows now how quickly it will melt... This is all unprecedented in the science... Until recently we didn't believe it possible, for instance, for water to permeate a glacier all the way to the bottom. But that's what's happening. As the water pools, it opens more areas of ice to melting" (Hilton, 2007). With the uncertainty and lack of verifiable projections, at an official level little is said, or what is said is dangerously conservative. This is what the 2007 IPCC report did in regard to sea-level rises, where its projection of a 18-59 cm rise by 2100 was based on models which do not "include the potential for increasing contributions from rapid dynamic processes in the Greenland and West Antarctic ice sheets, which have already had a significant effect on sea level over the past 15 years and could eventually raise sea level by many meters. Lacking such processes, models cannot fully explain observations of recent sea level rise, and accordingly, projections based on such models may seriously understate potential future increases" (Oppenheimer, O’Neill et al., 2007).

But the lack of tested projections is not to say that large parts of Greenland may not have already passed their "tipping point", just because there are not strict, verifiable models to support the assertion. The same was true of the Arctic sea ice, which was why the conservatism of the scientific method meant that there was a failure to predict the events until they were all but upon us, at which point even those scientists who had speculated as to what was about to happen were "shocked" at the sea ice loss in the northern summer of 2007. 

Thus James Hansen identifies a "scientific reticence" that "in at least some cases, hinders communication with the public about dangers of global warming... Scientific reticence may be a consequence of the scientific method. Success in science depends on objective skepticism. Caution, if not reticence, has its merits. However, in a case such as ice sheet instability and sea level rise, there is a danger in excessive caution. We may rue reticence, if it serves to lock in future disasters" (Hansen, 2007a). 

But there are useful sources other than models for thinking about the likely future rate of loss of the Greenland ice sheet, including expert elicitations and paleoclimatology. In response to the deep concerns about the 2007 IPCC Working Group 1 Summary for Policymakers, it has been proposed that the base of inputs be broadened "to give observational, paleoclimatic, or theoretical evidence of poorly understood phenomena comparable weight with evidence from numerical modeling. In areas in which modeling evidence is sparse or lacking, IPCC sometimes provides no uncertainty estimate at all. In other areas, models are used that have quantitatively similar structures, leading to artificially high confidence in projections (e.g., in the sea-level, ocean circulation, and carbon-cycle examples above). One possible improvement would be for the IPCC to fully include judgments from expert elicitations" (Oppenheimer, O’Neill et. al., 2007). 

One expert elicitation suggests: "Could the Greenland ice sheet survive if the Arctic were ice-free in summer and fall? It has been argued that not only is ice sheet survival unlikely, but its disintegration would be a wet process that can proceed rapidly. Thus an ice-free Arctic Ocean, because it may hasten melting of Greenland, may have implications for global sea level, as well as the regional environment, making Arctic climate change centrally relevant to definition of dangerous human interference" (Hansen & Sato, 2007). Off the record, Arctic climate researchers will say this is not an unreasonable view; on the record they will say there are no verifiable models which produce this result. These statements are not in contradiction. 

So, for example, Eric Rignot, a lead author of a paper (Rignot & Kanagaratnam, 2006) showing a doubling of loss from the Greenland ice sheet over a decade, was moved to comment that "These results absolutely floored us... The glaciers are sending us a signal. Greenland is probably going to contribute more and faster to sea-level rise than predicted by current models" (New Scientist, 2006). Another informed opinion comes from Robert Correll of the Arctic Climate Impact Assessment, who reports, as mentioned above, that the Greenland ice cap is melting so quickly that it is triggering earthquakes as pieces of ice several cubic kilometres in size break off, such that "scientists monitoring events this summer say the acceleration could be catastrophic in terms of sea-level rise and make predictions this February by the Intergovernmental Panel on Climate Change far too low" (Brown, 2007). 

As for paleoclimate record, global average temperatures are within 1°C of those that thawed much of Greenland's ice cap some 130,000 years ago, when the planet last enjoyed a balmy respite from continent-covering glaciers, and sea were 5–6 metres higher than today. Global warming appears to be pushing vast reservoirs of ice on Greenland and Antarctica toward a significant, long-term meltdown, and the world may have as little as a decade to take the steps to avoid this scenario (Spotts, 2006; Hansen, 2005; Hansen, Sato et al, 2006). 

To recap, it is reasonable to expect the very rapid loss of the Arctic sea ice, with a significant impact on regional temperatures due to the albedo effect. It is also reasonable to expect, as a consequence, an acceleration of the rate of loss of the Greenland ice sheet, which may already be at or near its disintegration tipping point for a large part of the ice sheet, a situation that was previously not expected for a long time. The precautionary principle suggests that we fully take into account the possibility of these outcomes, especially for their wider impact on the climate system (NASA, 2007), and on the sea-level rise the loss of the Greenland ice sheet will produce, perhaps in as little as a century or so. 

[Extract ends]