NSIDC: Arctic and Antarctic Sea Ice News for November 2016 -- lowest extent on record for November by large margin

Average Arctic sea ice extent for November set a record low, reflecting unusually high air temperatures, winds from the south, and a warm ocean. Since October, Arctic ice extent has been more than two standard deviations lower than the long-term average. Antarctic sea ice extent quickly declined in November, also setting a record low for the month and tracking more than two standard deviations below average during the entire month. For the globe as a whole, sea ice cover was exceptionally low.

Overview of conditions

Figure 1. Arctic sea ice extent for November 2016 was 9.08 million square kilometers (3.51 million square miles). The magenta line shows the 1981 to 2010 median extent for the month. The black cross indicates the geographic North Pole. Sea Ice Index data. About the data. Credit: National Snow and Ice Data Center
High-resolution image

In November 2016, Arctic sea ice extent averaged 9.08 million square kilometers (3.51 million square miles), the lowest November in the satellite record. This is 800,000 square kilometers (309,000 square miles) below November 2006, the previous lowest November, and 1.95 million square kilometers (753,000 square miles) below the 1981 to 2010 long-term average for November. For the month, ice extent was 3.2 standard deviations below the long-term average, a larger departure than observed in September 2012 when the Arctic summer minimum extent hit a record low.

At this time of year, air temperatures near the surface of the Arctic Ocean are generally well below freezing, but this year has seen exceptional warmth. The overall rate of ice growth this November was 88,000 square kilometers (34,000 square miles) per day, a bit faster than the long-term average of 69,600 square kilometers (26,900 square miles) per day. However, for a brief period in the middle the month, total extent actually decreased by 50,000 square kilometers, or 19,300 square miles—an almost unprecedented occurrence for November over the period of satellite observations. A less pronounced and brief retreat of 14,000 square kilometers (5,400 square miles) occurred in 2013.

Ice growth during November as a whole occurred primarily within the Beaufort, Chukchi and East Siberian Seas, as well as within Baffin Bay. Ice extent slightly retreated in the Barents Sea for the month. Compared to the previous record low for the month set in 2006, sea ice was less extensive in the Kara, Barents, East Greenland, and Chukchi Seas, and more extensive in Baffin Bay this year.

Conditions in context

Figure 2a. The graph above shows daily Arctic sea ice extent as of December 5, 2016, along with daily ice extent data for four previous years. 2016 is shown in blue, 2015 in green, 2014 in orange, 2013 in brown, and 2012 in purple. The 1981 to 2010 average is in dark gray. The gray area around the average line shows the two standard deviation range of the data. Sea Ice Index data. 
Credit: National Snow and Ice Data Center. High-resolution image

Figure 2b. This plot shows air temperature difference from average in the Arctic for November 2016. Air temperatures at the 925 hPa (approximately 2,500 feet) level in the atmosphere were above the 1981 to 2010 average over the entire Arctic Ocean and, locally up to 10 C (18 F) above average near the North Pole. This is in sharp contrast to northern Eurasia, where temperatures were up to 4-8 C (7-14 F) below average. Credit: NSIDC courtesy NOAA/ESRL Physical Sciences Division. High-resolution image

Continuing the warm Arctic pattern seen in October, November air temperatures were far above average over the Arctic Ocean and Canada. Air temperatures at the 925 hPa level (about 2,500 feet above sea level) were above the 1981 to 2010 average over the entire Arctic Ocean and, locally up to 10 degrees Celsius (18 degrees Fahrenheit) above average near the North Pole. This is in sharp contrast to northern Eurasia, where temperatures were as much as 4 to 8 degrees Celsius (7 to 14 degrees Fahrenheit) below average (Figure 2b). Record snow events were reported in Sweden and across Siberia early in the month.

In autumn and winter, the typical cyclone path is from Iceland, across the Norwegian Sea and into the Barents Sea. This November, an unusual jet stream pattern set up, and storms instead tended to enter the Arctic Ocean through Fram Strait (between Svalbard and Greenland). This set up a pattern of southerly wind in Fram Strait, the Eurasian Arctic and the Barents Sea and accounts for some of the unusual warmth over the Arctic Ocean. The wind pattern also helped push the ice northwards and helps to explain why sea ice in the Barents Sea retreated during November.

Sea surface temperatures in the Barents and Kara Seas remained unusually high, which also helped prevent ice formation. These high sea surface temperatures are a result of warm Atlantic water circulating onto the Arctic continental shelf seas.

November 2016 compared to previous years

Figure 3. Monthly November ice extent for 1979 to 2016 shows a decline of 5.0% per decade. Credit: National Snow and Ice Data Center. High-resolution image

Through 2016, the linear rate of decline for November is 55,400 square kilometers (21,400 square miles) per year, or 5.0% per decade.

Warm Arctic delays ice formation in Svalbard’s fjords

Figure 4a. This plot shows ocean temperature differences from average by depth (y-axis, in decibars; a decibar is approximately one meter) along a transect (x-axis, in kilometers) from the outer continental shelf to the inner parts of Isfjorden, the largest fjord in the Svalbard archipelago, for mid November 2016. (Areas in black show the undersea topography.) Atlantic Water is as warm as 5 C (41 F) and the surface layer still about 2 C (36 F). The surface layer would normally have cooled to the salinity adjusted freezing point at (-1.8 C, 29 F) at this time of year, enabling sea ice formation. Credit: University Centre in Svalbard. High-resolution image

Figure 4b. The West Spitsbergen Current consists of three branches (red arrows) that transport warm and salty Atlantic Water northward: the Return Atlantic Current (westernmost branch), the Yermak Branch and the Svalbard Branch. The Spitsbergen Trough Current (purple) transports Atlantic Water from the Svalbard Branch into the troughs indenting the shelf along Svalbard. Since 2006, changes in atmospheric circulation have resulted in more warm Atlantic Water reaching these fjords. The blue and red circles on the figure indicate locations where hydrographic data were collected. Credit: University Centre in Svalbard (UNIS). High-resolution image

Figure 4c. An inky-black polar night—but no cooling. The moon is the only source of light in the Arctic now, and here shines over open water in Isfjorden, the largest fjord in the Svalbard archipelago, in mid-November 2016. Credit: Lars H. Smedsrud. High-resolution image

In the Svalbard archipelago, sea ice usually begins to form in the inner parts of the fjords in early November. This November, however, no sea ice was observed. Throughout autumn, the wind pattern transported warm and moist air to Svalbard, leading to exceptionally high air temperatures and precipitation, which fell as rain.

Atmospheric and oceanic conditions in the fjord system were assessed by students from the University Centre in Svalbard. They noted an unusually warm ocean surface layer about 4 degrees Celsius (7 degrees Fahrenheit) above the salinity-adjusted freezing point (Figure 4a). Coinciding with exceptionally high air temperatures over Svalbard during autumn, the water has hardly cooled at all, and it is possible that no sea ice will form this winter.

The above-average ocean temperatures arose in part from changes in ocean currents that bring warm and salty Atlantic Water into the fjords. As the warm Gulf Stream moves east, it becomes the branching North Atlantic Drift. One small branch is named the West Spitsbergen Current (Figure 4b). This current flows along the continental shelf on the west coast of Svalbard and is one mechanism for transporting heat towards the fjords. Since 2006, changes in atmospheric circulation have resulted in more Atlantic water reaching these fjords, reducing sea ice production in some and stopping ice formation entirely in others.

Antarctic sea ice continues to track well below average

Figure 5a. Monthly November Antarctic sea ice extent for 1979 to 2016 shows an increase of 0.36% per decade. Credit: National Snow and Ice Data Center. High-resolution image

Figure 5b. This plot shows air temperature difference from average in the Antarctic for October 27 to November 17, 2016. Air temperatures at the 925 hPa level (approximately 2,500 feet) during the period of rapid sea ice decline in Antarctica (October 27 through November 17) were 2-4 C (4-7 F) above average near the sea ice edge. Credit: NSIDC courtesy NOAA/ESRL Physical Sciences Division. High-resolution image

Figure 5c. This map of sea ice concentration difference from average for November 2016 shows very low ice extent in three areas of the ice edge (near the Antarctic Peninsula, near the western Ross Sea and Wilkes Land, and near Enderby Land) as well as extensive areas of lower-than-average concentration within the interior ice pack in the Weddell Sea, Amundsen Sea, and near the Amery Ice Shelf. Sea Ice Index data. Credit: National Snow and Ice Data Center. High-resolution image

This year, Antarctic sea ice reached its annual maximum extent on August 31, much earlier than average, and has since been declining at a fairly rapid pace, tracking more than two standard deviations below the 1981-2010 average. This led to a new record low for the month of November over the period of satellite observations (Figure 5a). Average extent in November was 14.54 million square kilometers (5.61 million square miles). This was 1.0 million square kilometers (386,000 square miles) below the previous record low of 15.54 million square kilometers (6.00 million square miles) set in 1986 and 1.81 million square kilometers (699,000 square miles) below the 1981 to 2010 average.

For the month, Antarctic ice extent was 5.7 standard deviations below the long-term average. This departure from average was more than twice as large as the previous record departure from average, set in November 1986.

Ice extent is lower than average on both sides of the continent, particularly within the Indian Ocean and the western Ross Sea, but also to a lesser extent in the Weddell Sea and west of the Antarctic Peninsula in the eastern Bellingshausen Sea. Moreover, several very large polynyas (areas of open water within the pack) have opened in the eastern Weddell and along the Amundsen Sea and Ross Sea coast.

Air temperatures at the 925 mbar level were 2 -4 C (4-7 F) above average near the sea ice edge during late October and early November, corresponding to the period of rapid sea ice decline (Figure 5b).

The entire austral autumn and winter (since March 2016) was characterized by generally strong west to east winds blowing around the continent. This was associated with a positive phase of the Southern Annular Mode, or SAM. This pattern tends to push the ice eastward, but the Coriolis force acting in the ice adds a component of northward drift. During austral spring (September, October and November), the SAM index switched from strongly positive (+4 in mid-September, a record) to negative (-2.8 in mid-November). When the westerly wind pattern broke down in November, winds in several areas of Antarctica started to blow from the north. Over a broad area near Wilkes Land, the ice edge was pushed toward the continent. Areas with southward winds were also located between Dronning Maud Land and Enderby Land, and near the Antarctic Peninsula. This created three regions where ice extent quickly became much less extensive than usual (Figure 5c), reflected in the rapid decline in extent for the Antarctic as a whole. Interspersed with the areas of compressed sea ice and winds from the north, areas of south winds produced large open water areas near the coast, creating the polynyas.

Arctic sea ice loss linked to rising anthropogenic CO2 emissions

Figure 6. This plot shows the relationship between September sea ice extent (1953-2015) and cumulative CO2 emissions since 1850. Grey diamonds represent the individual satellite data values; circles represent pre-satellite era values; the solid red line shows the 30-year running average. The dotted red line indicates the linear relationship of 3 square meters per metric ton of CO2. Credit: J. Stroeve, National Snow and Ice Data Center High-resolution image

A new study published in the journal Science links Arctic sea ice loss to cumulative CO2 emissions in the atmosphere through a simple linear relationship (Figure 6). Researchers conducting the study, including NSIDC scientist Julienne Stroeve, examined this linear relationship based on observations from the satellite and pre-satellite era since 1953, and in climate models. The observed relationship is equivalent to a loss of 3 square meters (9.9 square feet) for every metric ton of CO2 added to the atmosphere, compared the average from all the climate models of 1.75 square meters (5.8 square feet). This smaller value, or lower sensitivity, from the models is consistent with findings that the models tend to be generally conservative relative to observations in regard to how fast the Arctic has been losing its summer ice cover. The observed rate of ice loss per metric ton of CO2 allows individuals to more easily grasp their contribution to Arctic sea ice loss.

Global sea ice far below average

Figure 7. This time series of daily global sea ice extent (Arctic plus Antarctic, month and first day of month on the x-axis) shows global extent tracking below the 1981-2010 average. Sea Ice Index data. Credit:W. Meier, NASA Cryospheric Sciences, GSFC. High-resolution image

As a result of both Arctic and Antarctic sea ice currently tracking at record low levels, global ice extent near November’s end stood at 7.3 standard deviations below average (Figure 7). However, the processes governing the evolution of sea ice in both hemispheres is a result of different atmospheric and oceanic processes and geographies and it unlikely that record low conditions in the two hemispheres are connected. Also, it is not especially instructive to assess a global sea ice extent because the seasons are opposite in the two hemispheres. In November the Arctic is in its ice growth season while Antarctic is losing ice. Antarctic sea ice as a whole has slightly increased over the past four decades (but with the last two austral winters having average and below average extent, respectively). The slight overall increase in Antarctic ice over the satellite record can be broadly linked to wind patterns that have helped to expand the ice cover towards the north (towards the equator).

NASA Operation IceBridge completes its 2016 Antarctic campaign

Figure 8. This photograph from Operation IceBridge shows broken floes of sea ice floating in the Weddell Sea. A large area of open water can be seen on the horizon. Credit: J. Beitler/National Snow and Ice Data Center. High-resolution image

In October, four NSIDC personnel accompanied the NASA Operation IceBridge campaign on its airborne surveys over Antarctica. The campaign completed a total of 24 flights over the continent in October and November, covering sea ice, land ice, ice shelves, and glaciers as Antarctica headed into its austral summer. Missions surveyed sea ice in the Weddell and Bellinghausen Seas with instruments that measure both sea ice extent and thickness. These measurements add to a time series of data that measures changes in sea ice and helps researchers assess the future trajectory of the ice pack and its impact on the climate. Visual observations from the flights confirmed that areas in the Bellingshausen Sea that are typically covered in sea ice were open water this year.

One of this year’s missions flew over a massive rift in the Antarctic Peninsula’s Larsen C Ice Shelf. Ice shelves are the floating parts of ice streams and glaciers, and they buttress the grounded ice behind them; when ice shelves collapse, the ice behind accelerates toward the ocean, where it then adds to sea level rise. Larsen C neighbors a smaller ice shelf that disintegrated in 2002 after developing a rift similar to the one now growing in Larsen C.

The IceBridge scientists measured the Larsen C fracture to be about 70 miles long, more than 300 feet wide and about a third of a mile deep. The crack completely cuts through the ice shelf but it does not go all the way across it. Once it does, it will produce an iceberg roughly the size of the state of Delaware.

The mission of Operation IceBridge is to collect data on changing polar land and sea ice and maintain continuity of measurements between NASA’s Ice, Cloud and Land Elevation Satellite (ICESat) missions. The original ICESat mission ended in 2009, and its successor, ICESat-2, is scheduled for launch in 2018. Operation IceBridge, which began in 2009, is currently funded until 2019. The planned overlap with ICESat-2 will help scientists validate the satellite’s measurements.

Further reading

Nilsen, F., Skogseth, R., Vaardal-Lunde, J., and Inall, M. 2016. A simple shelf circulation model: Intrusion of Atlantic Water on the West Spitsbergen Shelf. J. Physical Oceanography, 46, 1209-1230. doi:10.1175/JPO-D-15-0058.1

Notz, D. and J. Stroeve. 2016. Observed Arctic sea-ice loss directly follows anthropogenic CO2 emission. Science, 11 Nov 2016: Vol. 354, Issue 6313, pp. 747-750. doi:10.1126/science.aag2345.

Parkinson, C. 2014. Global sea ice coverage from satellite data: Annual cycle and 35-year trends. Journal of Climate, December 2014. doi:10.1175/JCLI-D-14-00605.1.

References

Fetterer, F., K. Knowles, W. Meier, and M. Savoie. 2016, updated daily. Sea Ice Index, Version 2. Boulder, Colorado USA. NSIDC: National Snow and Ice Data Center. doi:10.7265/N5736NV7.

http://nsidc.org/arcticseaicenews/

tamino: Deniers: Is This Really the Hill You Want to Die On? “No global warming for 15 years” ?

by tamino, "Open Mind" blog, September 10, 2016

Of all the lies we constantly hear from deniers, one of the most common (and with them, most popular) is some variation of “no global warming for 15 years.” It started many years ago with “no global warming since 1998,” has since been revised to “no global warming since 1997” and “no global warming since 2001” and “no global warming since 2002,” has run through “no global warming for 18 years” and “no global warming for 20 years,” but now seems to have settled on “no global warming for 15 years.”

The reason it keeps changing is that whenever they decide on what to bellow about, it quickly becomes obvious it’s not true. None of their claims is true. That’s no problem for them. Doubt is their product.

They used to base such claims on Earth’s surface temperature data. But a glance at how temperature at Earth’s surface has changed will quickly dispel that notion.

It’s ridiculously obvious that in addition to the trend in temperature — the changes that persist — there are also fluctuations, changes that don’t persist. The red line in the above graph is an estimate of the trend; if we subtract it from the data we have an estimate of the fluctuations, which I’ll show here by adding a blue line to represent the year-to-year fluctuations.

To make it look like there’s no warming, they have to isolate a brief span of time, one so short that the trend doesn’t have enough time to make itself obvious, so the fluctuations are all you really see. And they have to pick a time span when the fluctuations tend downward, to cancel out the appearance of the upward trend. You might try the span from 2001 through 2014, for instance:

A close-up shows that this time span is one during which the fluctuations tended to go opposite to the trend, essentially hiding the trend:

It’s what fluctuations do. They go sometimes up, sometimes down, and when they go up then go down it can look like a downward trend, cancelling out the upward trend. But fluctuations don’t persist. That’s why those who show you a tiny time span, don’t want you to see what led up to it — or what came after.

The same claim could have been made for the period from 1980 through 1994:

A close-up shows the same:

Alas for the poor climate deniers, the last two years both set new hottest-year records and 2016 is on its way to doing it again. The result is that the record of surface temperature makes it embarrassing to tell people there’s been “no warming since…”

Another embarrassment for them is that during the very time they like to crow about “no warming,” we’ve seen such a massive decrease in Arctic sea ice:

Then there’s the amount of heat in the ocean, rather than at the surface where we live:

There’s also the level of the sea itself, which has risen because global warming has melted so much land ice and because the sea expands as it warms:

There’s also the incredible rate of the loss of Greenland ice:

And the shrinking of the world’s glaciers. And the migration of species to higher latitudes and higher elevations. And the earlier arrival of spring. And earlier snowmelt. And more wildfire burn. And earlier breakup of the ice on frozen lakes. There are so many signs, only those in denial can deny it — and those signs have been clear during the “last 15 years” just as they were before.

But there’s still a little bit of data one can twist to look like maybe, if you hide most of it and squint while looking at it in just the right way, you can make yourself believe there’s been “no global warming for 15 years.” It’s the satellite data for the temperature in the lower troposphere from the University of Alabama at Huntsville.

That’s why, when politicians whine about “no global warming” they only show you that data. Not surface temperature. Not ocean heat content or sea level. Not Greenland melting or glacier disappearance or any of the other things that show how hot it’s getting. They only show you what they can manipulate to make their case … and generally don’t even show you all of it, they have to leave out what might provide context.

But they have a problem. A big one. It’s getting more and more obvious that their “no warming for XX years” narrative is bullshit. They picked that dialogue, and it’s going to end up killing them. The day will come, soon, when not even Donald Trump will be able to weasel his way out of the reality of man-made climate change.

Deniers: you picked this hill. For you … it’s not a good choice.

---

Tamino's blog is made possible by readers like you; join others by donating at Peaseblossom’s Closet.

https://tamino.wordpress.com/2016/09/10/deniers-is-this-really-the-hill-you-want-to-die-on/

Peter Wadhams: ‘Next year or the year after, the Arctic will be free of ice’

Scientist Peter Wadhams believes the summer ice cover at the north pole is about to disappear, triggering even more rapid global warming


Peter Wadhams in the Arctic in 2007: ‘We may able to raise the Thames barrier in Britain but in Bangladesh, people will be drowned.’

by Robin McKie, The Guardian, August 21, 2016

Peter Wadhams has spent his career in the Arctic, making more than 50 trips there, some in submarines under the polar ice. He is credited with being one of the first scientists to show that the thick icecap that once covered the Arctic ocean was beginning to thin and shrink. He was director of the Scott Polar Institute in Cambridge from 1987 to 1992 and professor of ocean physics at Cambridge from 2001 to 2015. His book, A Farewell to Ice, tells the story of his unravelling of this alarming trend and describes what the consequences for our planet will be if Arctic ice continues to disappear at its current rate.

You have said on several occasions that summer Arctic sea ice would disappear by the middle of this decade. It hasn’t. Are you being alarmist?
No. There is a clear trend down to zero for summer cover. However, each year chance events can give a boost to ice cover or take some away. The overall trend is a very strong downward one, however. Most people expect this year will see a record low in the Arctic’s summer sea-ice cover. Next year or the year after that, I think it will be free of ice in summer and by that I mean the central Arctic will be ice-free. You will be able to cross over the North Pole by ship. There will still be about a million square kilometres of ice in the Arctic in summer, but it will be packed into various nooks and crannies along the Northwest Passage and along bits of the Canadian coastline. Ice-free means the central basin of the Arctic will be ice-free, and I think that that is going to happen in summer 2017 or 2018.

Why should we be concerned about an Arctic that is free of ice in summer?
People tend to think of an ice-free Arctic in summer in terms of it merely being a symbol of global change. Things happen, they say. In fact, the impact will be profound and will effect the whole planet and its population. One key effect will be albedo feedback. Sea ice reflects about 50% of the solar radiation it receives back into space. By contrast, water reflects less than 10%. So if you replace ice with water, which is darker, much more solar heat will be absorbed by the ocean, and the planet will heat up even more rapidly than it is doing at present.

Sea ice also acts as an air-conditioning system. Winds coming over the sea to land masses such as Siberia and Greenland will no longer be cooled as they pass over ice, and these places will be heated even further. These effects could add 50% to the impact of global warming that is produced by rising carbon emissions.

What will be the effects of this accelerating increase in temperatures?
The air over Greenland will get warmer, and more and more of its ice will melt. It is already losing about 300 cubic kilometres of ice a year. Antarctica is adding to the melt as well. Sea-level rises will accelerate as a result. The most recent prediction of the Intergovernmental Panel on Climate Change (IPCC) is that seas will rise by 60 to 90 centimetres this century. I think a rise of one to two metres is far more likely. Indeed, it is probably the best we can hope for.

That may not sound a lot, but it is really very serious. It will increase enormously the frequency of storm surges all over the world. We may be able to raise the Thames barrier in Britain, but in Bangladesh, it just means more and more people will be drowned.

Global warming is generally associated with increased fossil-fuel burning and consequent rises in levels of atmospheric carbon dioxide. But is that the only climate problem we face?
No, it is not. We also have the issue of methane. Russian scientists who have investigated waters off their coast have detected more and more plumes of methane bubbling up from the seabed. The reason this is happening is closely connected with the warming of the planet and the shrinking of the Arctic icecaps.

Until around 2005, even in summer, you still had sea ice near the coast. Then it started to disappear, so that for three or four months a year warm water reached the shallow waters around the shores where there had been permafrost ground since the last ice age. It has started to melt with dangerous consequences. Underneath the permafrost there are sediments full of methane hydrates. When the permafrost goes, you release the pressure on top of these hydrates and the methane comes out of solution.

Can we monitor this methane just as we can monitor carbon dioxide?
Yes, we can measure methane over large areas using satellites. These have shown that methane levels that had been fairly flat for most of the last century have started to rise and are accelerating, often with little outliers on the graph. There is a scientist called Jason Box who works in Denmark for the Greenland Survey, and he calls these outliers dragon’s breath. They are not some sort of measurement caused by dodgy instruments. They are real pulses of methane coming from offshore flumes.

Facebook

An image from the NOAA/Nasa Suomi NPP satellite taken on May 30, 2016, highlights the Arctic ice retreat off the north-west coast of Alaska. The average Arctic sea ice extent for May 2016 set a new record low since satellite observations began. Photograph: Suomi NPP/NASA/NOAA

How intense is methane as a heater of the atmosphere compared with carbon dioxide?
It is 23 times more powerful. However, methane dissipates much more quickly than carbon dioxide. It gets oxidised so that it only lingers in the atmosphere for about 7 or 8 years. By contrast, carbon dioxide hangs around in the climate system for about 100 years before it ends up in the sea and is absorbed by creatures that die and litter the seabed. At least that is what scientists thought. Today, there are quite a number of researchers who think carbon dioxide could last 1,000 years in the atmosphere.

So in the long run carbon dioxide is still going to be worse than methane in terms of heating the planet because a single methane pulse will have a disastrous effect, but if there is nothing to follow it on then it will go away. But with carbon dioxide there is a ratchet effect. All the carbon dioxide we release by burning fossil fuels just builds up in the atmosphere. We are having to live with last century’s carbon dioxide. What that says is simple: there is no such thing as a safe emission rate of carbon dioxide. That is why I am despondent about us ever being able to cut carbon emissions.

If we cannot halt the emissions of carbon dioxide, what can we do?
In the end, the only hope we have is to find a way to remove carbon dioxide from the atmosphere once it has got there. Even the IPCC has admitted that we will have to find a way to extract carbon dioxide from the air. The trouble is that they just don’t know how we can do that. The most favoured scheme is known as BECCS: bio-energy with carbon capture and storage. Essentially, you plant trees and bushes over vast swaths of ground. These grow, absorbing carbon dioxide in the process. Then you burn the wood to run power plants while trapping, liquefying and storing the carbon dioxide that is released.

It sounds straightforward. Will it work?
I am a bit suspicious of this technology. BECCS will need so much land to be effective. Calculations suggest it would need 40% to 50% of the arable land of the planet to make it work on the scale we will need and that would not leave enough land to grow crops to feed the world or to provide homes for a viable population of wild animals and plants. Other techniques, such as crushing and spreading olivine rocks, which absorb carbon dioxide, on beaches, will simply not scale up. They won’t work, so we will have to find some other way to remove carbon dioxide from the atmosphere directly.

As far as I can see, it will have to take the form of some sort of device into which you pump air at one end and you get air without carbon dioxide coming out the other end. It can be done, I am sure, but at the moment we do not have such a device. However, without something like that, I cannot see how we are going to deal with the carbon dioxide that is getting into the atmosphere. We are going to have to rely on a technology that has not yet been developed. That is a measure of the troubles that lie ahead for us. I think humanity can do it, but I would feel much better if I saw governments investing in such technology.

Farewell to Ice is published by Allen Lane (£20) on 1 September. Click here to order a copy for £16.40

https://www.theguardian.com/environment/2016/aug/21/arctic-will-be-ice-free-in-summer-next-year