Skip to content

Climate Change Clarity, IPCC Comprehensibility, and More on Antarctic Deglaciation


The Stephen and Tony Show – finally some straight (if misguided) talk from Harper as he meets with new friend Abbott

A number of factors have conspired to reduce my attention to this blog over the past six weeks, but the pause has given me a chance to reflect.  What topics deserve most of my attention?  Canadian politics around tar sands, pipelines, climate science, and Canadian CO2 emissions are as lively as ever.  Why, just the other night I saw Stephen Harper finally come clean, bluntly and unambiguously, about his attitude to climate change.  In a joint media event with his new best boyfriend, Tony Abbott, Prime Minister of Australia, Harper stated, more or less, ‘when it comes to a choice between the economy and the climate, the economy always comes first’.

combo Harper Abbott photos

On the left, Stephen Harper enunciating his view on the primary of the economy while Australian PM, Tony Abbott, looks on approvingly; on the right, earlier in the day, Tony Abbott practicing the Stephen Harper, ‘how-big-a-lie-am-I-telling’ gesture at the Canadian War Museum.  Photos © Fred Chartrand/Canadian Press (left) and Andrew Meares/Sydney Morning Herald (right).

That he said this did not come as much of a surprise.  His government has consistently obfuscated, prevaricated, and delayed any actions on Canadian emissions, while pretending that we were making progress towards meeting the commitments his government voluntarily made at the Copenhagen Climate Conference in 2009.  Simultaneously, he has embarked on an unprecedented muzzling of environmental science across the country, combined with an effort to label anybody who questions current policy as a traitor or worse.  Meanwhile, the Harper government has done everything possible (except have all Cabinet members work their vacation days in the tar sands, digging the stuff up and hauling it out) in an effort to help the multinational energy corporations triple production as rapidly as possible.  And this ruthlessly pro-resources-sector policy has been sold to us, in an endless stream of advertisements that we pay for, as a responsible focus on jobs and the economy.  All his words last night did was confirm what was already widely surmised – Harper, and presumably his trusted advisors, see the relation between environment and economy as a simple either/or choice, and they have chosen economy.

Harper should have been here in Muskoka in mid-May to attend the Muskoka Summit where six experts who actually knew what they were talking about made it abundantly clear that a) it’s not a simple either/or choice, b) there are more ways to build an economy in Canada than focusing on digging up resources to peddle overseas, and c) the wrong kind of economy, like the one Harper is building for Canada, cannot succeed in the long term – the planet will always win in the end, although the win may not be pretty.  Coincidentally, the pundits are pretty certain that the visit to Ottawa by Tony Abbott was mainly done to enable some planning before the coming Brisbane G20 meeting.  These two fossil-fuel fans are plotting to ensure that Obama or some other leader does not try and do anything sensible like introduce a carbon market.  The leaders of two middling nations, united by their British Empire heritage and their fondness for fossil fuels, are plotting to ensure no progress on climate change is made.

See.  There is plenty to blog about, dealing only with Canadian politics!  But I do not want this blog to be about Canadian politics; I want it to be about the environmental crisis, the evidence for that crisis, and particularly the new evidence that keeps piling up.  So I am going to try and pick topics that allow me to use my own experience and expertise to evaluate the steady stream of scientific reports that add to our knowledge of how we are doing.

Comprehensible Communication

The June issue of Nature Climate Change contains an interesting article by David Budescu of Fordham University, and three colleagues, titled “The interpretation of IPCC probabilistic statements around the world”.  It deals not with climate science, but with the difficulty of reporting that science to the public.  From the outset, the Intergovernmental Panel on Climate Change (IPCC) has attempted to prepare its documents so that they can be read and understood by non-specialists.  When talking about any scientific topic there is always uncertainty, and scientists use statistical techniques to assess the level of uncertainty (or imprecision of results) and express it in numerical terms.  For example, the probability statement, P<0.05, means to a scientist that there is only one chance in twenty (5%) that results observed could have arisen by chance variation alone.

(Quick aside; uncertainty in science does not mean ignorance or mistakes.  Uncertainty exists in every measurement anyone makes, every estimate anyone attempts to make, every conclusion based on an evaluation of factual knowledge.  Scientists know and understand this fact of life, and they attempt to quantify the degree of uncertainty in the information they publish.  This is in stark contrast to climate denialists, evangelicals, and people who still believe strongly that the world is flat, that razor blades resharpen themselves if stored under a pyramid, or that Elvis Pressley still lives.)

Expressions like ‘the 90th percentile’, ‘the median income’, and ‘a variance of 2%’ all have explicit, mathematical meanings, although the precision of these statements may be less clear to the non-specialist, and a statement such as ‘the mean yield was 0.5mg ± 0.17SD’ is quite precise yet almost unintelligible to the average reader.  IPCC’s solution to the need to report the precision of particular estimates was to create a list of specific phrases, each of which would refer to a particular numerical probability statement.  Thus, for P<0.05 IPCC used ‘extremely unlikely’, while for P>0.90 ‘very likely’ was used.  A table listing the numerical estimates and the equivalent phrases has been included in each IPCC report.

IPCC Table of probs NCC June 2014

Table from Budescu’s article showing the meaning assigned to specific terms by IPCC

Budescu’s article reports on the meanings that readers attach to four of these phrases when they are presented in sentences about climate.  The four phrases tested were ‘very unlikely’, ‘unlikely’, ‘likely’, and ‘very likely’ (equivalent to <10%, <33%, >66% and >90% respectively).  What happened was that respondents tended to interpret each of the phrases as meaning a probability somewhat closer to 50% than was intended.  In other words, extremely improbable and very nearly certain results became homogenized with others to a degree, suggesting the scientists had ideas but were not really sure about any of them.  This was the case despite the fact that respondents had the table linking phrases to probabilities available throughout the test.

IPCC interpretation by readers June 2014 nclimate2194-f1

Figure 1 from David Budescu’s article in Nature Climate Change compares the estimates (percentage scale on left axis) for the quantitative meaning of four phrases presented in sentences to subjects who either had a table of equivalencies (as shown at the bottom of the figure) available (darker bars), or had the phrases immediately adjacent to the numerical equivalent within the sentences (pale gray bars).  For each case, the thin vertical line spans 90% of the range of percentages chosen by respondents, the bar spans the most correct 50%, the mean response is denoted by a + and the median response by a horizontal line.  Figure © Nature Climate Change.

Budescu and colleagues then repeated the test to a separate set of people without the table of equivalencies, but with the sentences modified so that the words were followed by the numerical equivalent in parentheses – “…it was very unlikely (P<10%) that…”.  Under these circumstances the respondents performed better, although they still did not treat the phrases correctly.  These results were more or less the same across some 24 countries and 17 languages!

Budescu’s team also collected survey data from participants to establish whether they ‘believed in’ climate change or not, and whether they supported the idea of taking action to reduce GHG releases.  They found that, in general, ‘believers’ showed less bias than ‘deniers’ in how they interpreted the probability statements – that is, they tended to interpret statements more correctly.  Budescu recommends that IPCC rethink how it expresses quantitative information in its reports.

I think there is a wider issue here, however.  Even with the numerical equivalent embedded in the sentences, subjects biased their interpretation to make the results less extreme.  Language is flexible, and readers take in new information while having in their memories old information and beliefs.  Those memories and beliefs seem to be influencing their understanding of the new information.  And we, or most of us, anyway, still intuitively expect the world to be a reliable, constant, safe place.  The biased interpretation of the new information happens because, subconsciously, we’d prefer not to learn just how serious some of the conclusions about climate change are. And this makes me more convinced than ever that we have got to tell the story of climate change clearly and explicitly and often, hammering away at a seemingly innate effort to not grasp the significance of it.

Marshall McLuhan tthjBea

Marshall McLuhan also said, “If I had not believed it, I never would have seen it”.
Perhaps appropriate here.

Antarctic Deglaciation

I’ll close today by discussing some more of the recent information I’ve come across about the melting of ice.  Either it’s just because I have started looking for them, or there has been a recent explosion of studies on this topic, but it seems to me that there are a lot of new studies discussing the melting of ice in the Arctic and Antarctic, and the consequences for sea level rise or future rates of warming.

The Arctic and Antarctic are polar opposites, and not just because they lie at opposite ends of the Earth.  With the exception of Greenland, the bulk of Arctic ice is sea ice, while the bulk of Antarctic ice is on land.  Melting of floating sea ice will do nothing to alter sea level, although it will hasten warming of the ocean (because open water is darker in color and less likely to reflect light than ice), and will also lower salinity, perhaps upsetting the global current patterns, referred to as the great ocean conveyor, that drive circulation and turning over of the oceans.  These changes can have immense implications for global climate.  The melting of Antarctic ice (and the melting of Greenland’s glacial cap) will directly impact sea level to a potentially enormous degree, while the melting of floating sea ice around Antarctica will cause similar changes to salinity and reflectivity of southern ocean waters as will happen in the Arctic.  What is important to understand is that the continental ice mass of Antarctica is so immense that much of it extends beyond the Antarctic land mass, but rests solidly upon submerged shelves rather than floating.  In other words, what might appear to be sea ice around Antarctica is, in shoreward portions, resting on solid rock rather than floating.

A couple of months ago, I discussed a couple of papers on polar melting, one dealing with Greenland and possible effects on sea level, and the other dealing with happenings in the Southern Ocean that might be slowing the Ocean Conveyor while trapping heat in the ocean that would otherwise be warming our climate.  Today I am focusing on several articles that have appeared quite recently, all dealing with the way in which glaciers in Antarctica melt.  The short answer appears to be: they ARE melting, the melting is going to continue for a long time, the melting may include sudden episodes followed by long pauses, and sea level will ultimately rise far more than the 50cm or so that some people are talking about with concern.

Snow falls on Antarctica adding mass to the glaciers that cover its surface.  The glaciers move outwards towards and beyond the coast.  Sea water, far warmer than the air, melts and undercuts the advancing glaciers which calve off icebergs.  The icebergs get transported by currents around Antarctica and eventually away to the north where they melt.  The scale of all this is immense.  The icebergs calved off amount to 1,300 to 2000 Gt (Gigatonnes) of ice per year.  That’s equivalent to between 520 and 800 million Olympic-sized swimming pools full of water per year once those icebergs melt.  At least half of these icebergs are 18 km or more in maximum dimension.  Most of the icebergs are transported out across the Weddell Sea along a route known as iceberg alley until they reach the warmer waters of the Scotia Sea.

Weber et al 5 June 2014 nature13397-f1

Iceberg Alley, the route taken by icebergs calved from Antarctica, and the two core sites (MD07-3133 & 3134) that provided the data for the study by Weber et al.  Figure © M. Weber, Nature

It is from the Scotia Sea that Dr. Michael Weber, University of Cologne, Germany, and a multinational group of 11 co-authors assembled the data for their recent article in Nature.  Rather than current or future rates of melting, they were interested in determining the pattern of melting in the past.  To do this they relied upon records of IBRD – iceberg-rafted debris – beneath the Scotia Sea.  All icebergs collect sand, pebbles and rocks as they move across the land.  As they melt, this debris rains down to the sea bed to become covered in marine sediments.  Using two sea-floor cores collected from sites about 300 km apart in the Scotia Sea, they examined the frequency of small particles of IBRD.  Using several procedures they were also able to date precisely the sediments along the length of the cores.  Places along the cores where abundance of IBRD was high were taken to be times in the past at which production of icebergs peaked.  Their results reveal a pattern since the peak of the last Pleistocene glaciation of eight episodes of pronounced iceberg production, a pattern they interpret as eight pulses of deglaciation interspersed with periods when production of icebergs slowed or even ceased.

Their rather complicated Figure 2 reveals considerable similarity in the variation in amount of IBRD through time (black spectra c and d), which resolves into 8 clearly defined pulses when they are combined (spectra e and f).  The earliest to these episodes of iceberg calving (which Weber and colleagues refer to as Antarctic Ice-sheet Discharges, AID) occurred between 20 and 19 thousand years ago, the most recent between 10 and 9 thousand years ago.  Individual episodes reveal a rapid initial rise in the abundance of IBRD, a duration of 100 to 1000 years or so, and a relatively rapid drop in abundance of IBRD at the end.  The timing of the episodes correlates with a number of known climatic or sea level events.

Weber et al 5 June 2014 nature13397-f2

Figure 2 from Weber et al study.  Rows c and d represent the variation in abundance of IBRD at different distances along the core – distance is expressed as age (shallowest/youngest at right) on the x axis.  The eight vertical tan bars labeled AID 1 to 8 represent the eight episodes of pronounced iceberg production during this period when the melting of the Pleistocene glaciers took place.  By 6000 years ago (just past the end of the axis in the figure) sea level had almost reached present levels and has remained remarkably stable until recently.  Figure © M. Weber, Nature.

The reasons why melting of glaciers should be so episodic are less certain.  One possibility is that the melting back of a grounded glacier (one resting on solid rock ledges below sea level) can lead to a sudden release if that rock ledge slopes to deeper water closer to shore – suddenly the front of the glacier is floating and outward movement and iceberg calving can both speed up.  Changes in climate or oceanography could also bring about a sudden change in rate of deglaciation.  The important conclusion to draw is that episodic deglaciation does occur, and these episodes of more rapid deglaciation are relatively short, start quickly, and can result in rapid changes in sea level.  For example, during the 350 years of what is called melt-water pulse 1A (MWP 1A), 14.65 to 14.3 thousand years ago, mean sea level rose 14 meters, and melting in Antarctica accounted for half this rise.  MWP 1A corresponds to AID6, the largest episode in the record of IBRD from the Scotia Sea.  At 4.5cm per year, that is a rate of sea level rise more than 10 times the present rate (3.1mm per year), and that rate continued for 300 years.  The past is not a firm predictor of the future, but neither is the future guaranteed to not repeat the past.  We should recognize that changing the climate of the planet has already led to increasing melt of glaciers, and it could lead in future to much more rapid rates of melting than we are seeing now.

Ian Joughin and colleagues at University of Washington published an article in Science in mid-April which illustrates one type of circumstance that could give rise to an episode of rapid Antarctic melting such as observed by Weber’s team.  They focused attention on the Thwaites glacier which is more or less in the center of the West Antarctic ice sheet.  It is currently melting, losing about 52 Gt of mass per year as warm ocean waters undercut its grounded base calving off glaciers; it has retreated 14 km in the last 19 years.


The seaward edge of the Thwaites glacier, photographed 16 October 2013.
Photo courtesy NASA Earth Observatory, taken by Jim Yungel

Joughin and colleagues confirmed that the base of the Thwaites glacier is currently firmly grounded on a subsea ledge 600 m deep.  However, this ledge slopes downward towards a deeper basin lying between it and the Antarctic mainland.  Thus, as it melts back it is gounding on ever deeper rock, resulting in less impediment to further outward movement and accelerated melting.  There will come a time when the ledge is too deep for the glacier to rest on it.  At that point the front edge of the glacier will be floating and the rate of movement outwards will increase substantially, to about 5 km per year.  The scientists used a modeling approach to estimate the likely future behavior of this glacier under different rates of initial melting.

Joughin and colleagues found that current rates of melting strongly suggest that the process of destabilizing this glacier has already started, and while they are talking in terms of 200 years or so, they are talking about the inevitability of total collapse on this coast.  Once again we see a general pattern:  environmental changes do not necessarily proceed gradually and at constant rates.  The Thwaites glacier has started to melt.  Its rate of melt is going to accelerate until it disappears entirely as a feature of the Antarctic landscape.  And the rest of the West Antarctic ice sheet is thought to be following along not far behind.  I conclude that the 20 cm change in sea level we have seen over the last 100 years is nothing compared to the change that is coming – a change that we cannot now stop.  It raises the question of whether we should bother building shore defenses at great expense – we will have to keep extending them if we want our great coastal cities to last more than a couple of hundred years.

Lest you are becoming complacent, thinking that the Thwaites glacier is unusual, or that the melting of Antarctica at the end of the Pleistocene is somehow not relevant to today, Matthias Mengel and Anders Leverman of the Potsdam Institute for Climate Impact Research, published an article in the June 2014 issue of Nature Climate Change, dealing with melting of the East Antarctic ice sheet.   Just as the Thwaites glacier, and most of the West Antarctica ice sheet is grounded on a shoreward-deepening shelf, the same topography is true of much of East Antarctica.  But East Antarctica holds more than five times the ice that is present in West Antarctica.  Mengel and Levermann focus their attention on the Wilkes Basin, a large deep basin with a shallower sill, which underlies the ice sheet.  Their simulations suggest that melting back of these glaciers behind that sill will destabilize the system, emptying out ice from the entire basin and substantially reducing the mass of ice in East Antarctica.  Mengel and Levermann speak in terms of a few centuries, but also use words like ‘inevitable’ and ‘collapse’.  And they mention global sea level rise of three to four meters caused by this event alone.

These three articles should give us pause, especially given that they follow a number of others.  In the Working Group 1 component of its Fifth Assessment brought out late in 2013, the IPCC stated that the average rate of ice loss from the Antarctic ice sheet appears to have increased from between 37 and 97 Gt.yr–1 during the period 1992–2001 to between 72 and 221 Gt.yr–1 during 2002–2011.  The IPCC projected rise in sea level (relative to sea level during 1986 to 2005) by the end of this century under business as usual (RCP8.5) will be between 0.52 and 0.98 meters.  I suspect that the actual numbers, both the current Antarctic ice loss and the sea level rise by century end will be towards or beyond the upper bound of these estimates – every new study seems to point to more significant changes than we had anticipated.  While sea level rise seems likely to happen slowly by human time frames, we are setting in place an inexorable process that will continue long after 2100.  The economic costs as we adapt our civilization to these changes will be substantial, and it did not have to be this way.  Sea level rise following the Pleistocene slowed by 6 thousand years ago and came essentially to a halt by 4 thousand years ago.  It would not have started up again without our tampering with the composition of the atmosphere.


Male Island, capital of the Maldives Islands, in the Indian Ocean.  It won’t only be nations like this that suffer economically as sea level rises.  Most of the world’s big cities are coastal too.
Photo by Peter Essick/Aurora Photos, © National Geographic