Greenland Whodunit

“The next 5–10 years will reveal whether or not [the Greenland Ice Sheet melting of] 2012 was a one off/rare event resulting from the natural variability of the North Atlantic Oscillation or part of an emerging pattern of new extreme high melt years.”

WHO: Edward Hanna, Department of Geography, University of Sheffield, Sheffield, UK
 Xavier Fettweis, Laboratory of Climatology, Department of Geography, University of Liège, Belgium
Sebastian H. Mernild, Climate, Ocean and Sea Ice Modelling Group, Computational Physics and Methods, Los Alamos National Laboratory, USA, Glaciology and Climate Change Laboratory, Center for Scientific Studies/Centre de Estudios Cientificos (CECs), Valdivia, Chile
John Cappelen, Danish Meteorological Institute, Data and Climate, Copenhagen, Denmark
Mads H. Ribergaard, Centre for Ocean and Ice, Danish Meteorological Institute, Copenhagen, Denmark
Christopher A. Shuman, Joint Center for Earth Systems Technology, University of Maryland, Baltimore, USA,  Cryospheric Sciences Laboratory, NASA Goddard Space Flight Center, Greenbelt, USA
Konrad Steffen, Swiss Federal Research Institute WSL, Birmensdorf, Switzerland, Institute for Atmosphere and Climate, Swiss Federal Institute of Technology, Zürich, Switzerland, Architecture, Civil and Environmental Engineering, École Polytechnique Fédéral de Lausanne, Switzerland
 Len Wood, School of Marine Science and Engineering, University of Plymouth, Plymouth, UK
Thomas L. Mote, Department of Geography, University of Georgia, Athens, USA

WHAT: Trying to work out the cause of the unprecedented melting of the Greenland Ice Sheet in July 2012

WHEN: 14 June 2013

WHERE: International Journal of Climatology (Int. J. Climatol.) Vol. 33 Iss. 8, June 2013

TITLE: Atmospheric and oceanic climate forcing of the exceptional Greenland ice sheet surface melt in summer 2012 (subs req.)

Science can sometimes be like being a detective (although I would argue it’s cooler) – you’ve got to look at a problem and try and work out how it happened. These researchers set out to do exactly that to try and work out how the hell it was that 98.6% of the ice sheet on Greenland started melting last summer.

Greenland – July 8th on the left only half melting. July 12th on the right, almost all melting (Image: Nicolo E. DiGirolamo, SSAI/NASA GSFC, and Jesse Allen, NASA Earth Observatory)

Greenland – July 8th on the left only half melting. July 12th on the right, almost all melting (Image: Nicolo E. DiGirolamo, SSAI/NASA GSFC, and Jesse Allen, NASA Earth Observatory)

For a bit of context, Greenland is the kind of place where the average July temperature in the middle of summer is 2oC and the average summer temperature at the summit of the ice sheet is -13.5oC. Brrr – practically beach weather! So there’s got to be something weird going on for the ice sheet to start melting like that. Who are the suspects?

Atmospheric air conditions
Suspect number one is the atmospheric air conditions. The summer of 2012 was influenced strongly by ‘dominant anti-cyclonic conditions’ which is where warm southerly air moves north and results in warmer and drier conditions. There was also a highly negative North Atlantic Oscillation (NAO) which created high temperatures at high altitudes around 4km above sea level, which could explain the melting on the summit. The researchers also pointed out that the drier conditions meant less cloud cover and more sunny days, contributing to speedier melting.

There were issues with the polar jet stream that summer, where it got ‘blocked’ and stuck over Greenland for a while. The researchers used the Greenland Blocking Index (GBI), which while not trading on the NYSE, does tell you about wind height anomalies at certain geopotential heights (yeah, lots of meteorological words in this paper!). All of this makes the atmosphere look pretty guilty.

Jet stream getting funky – temperature anomaly patterns at 600 hectopascals pressure, aka 4000m above sea level with a big red blob over Greenland (from paper)

Jet stream getting funky – temperature anomaly patterns at 600 hectopascals pressure, aka 4000m above sea level with a big red blob over Greenland (from paper)

Sea surface temperatures
Suspect number two is sea surface temperatures. If it was warmer in the ocean – that could have created conditions where the ice sheet melted faster right? The researchers ran a simulation of the conditions around Greenland for the summer of 2012 and then played around with different temperature levels for sea surface, as well as levels of salinity. It didn’t make more than 1% difference, so they don’t think it was sea surface. Also, ocean temperatures change more slowly than air temperatures (that’s why the ocean is still so cold even in the middle of summer!) and when they looked at the data for sea surface temperature, it was actually a bit cooler in 2012 than it was in 2011. Not guilty sea surface temperatures.

Sea surface temperatures (top) and salinity (bottom) both decreasing (from paper)

Sea surface temperatures (top) and salinity (bottom) both decreasing (from paper)

Cloud patterns
Suspect number three is cloud cover patterns, which the researchers said could be a contributor to the ice sheet melting. However, they don’t have a detailed enough data set to work out how much clouds could have contributed to the melt. Not guilty for now clouds – due to lack of evidence.

Which leaves suspect number one – atmospheric air conditions. Guilty! Or, as the paper says ‘our present results strongly suggest that the main forcing of the extreme Greenland Ice Sheet surface melt in July 2012 was atmospheric, linked with changes in the summer NAO, GBI and polar jet stream’.

Now comes the scary part – it’s the atmosphere that we’ve been conducting an accidental experiment on over the last 200 years by burning fossil fuels. As the researchers point out, the North Atlantic Oscillation has a natural variability and patterns, so we could all cross our fingers and hope that the Greenland melting was a once off anomaly. Given the work that Dr Jennifer Francis has been doing at Rutgers into polar ice melt and how that slows the jet stream and causes it to meander; this may not be a good bet. Combine this with the fact that this level of melting is well beyond ‘the most pessimistic future projections’ and it’s getting scarier. This kind of melting was not supposed to occur until 2100, or 2050 in the worst case scenarios.

Interestingly, this could also link through to some of the work Jason Box is doing with his DarkSnow project in Greenland looking at how soot from fires and industry are affecting the melting of Greenland. The paper concludes that the next 5-10 years will show us whether it was a one off or the beginning of a new normal. Unless we stop burning carbon, it will only be the start of a terrifying new normal.

Advertisements

Pandora’s Permafrost Freezer

What we know about permafrost melt is less than what we don’t know about it. So how do we determine the permafrost contribution to climate change?

WHO: E. A. G. Schuur, S. M. Natali, C. Schädel, University of Florida, Gainesville, FL, USA
B. W. Abbott, F. S. Chapin III, G. Grosse, J. B. Jones, C. L. Ping, V. E. Romanovsky, K. M. Walter Anthony University of Alaska Fairbanks, Fairbanks, AK, USA
W. B. Bowden, University of Vermont, Burlington, VT, USA
V. Brovkin, T. Kleinen, Max Planck Institute for Meteorology, Hamburg, Germany
P. Camill, Bowdoin College, Brunswick, ME, USA
J. G. Canadell, Global Carbon Project CSIRO Marine and Atmospheric Research, Canberra, Australia
J. P. Chanton, Florida State University, Tallahassee, FL, USA
T. R. Christensen, Lund University, Lund, Sweden
P. Ciais, LSCE, CEA-CNRS-UVSQ, Gif-sur-Yvette, France
B. T. Crosby, Idaho State University, Pocatello, ID, USA
C. I. Czimczik, University of California, Irvine, CA, USA
J. Harden, US Geological Survey, Menlo Park, CA, USA
D. J. Hayes, M. P.Waldrop, Oak Ridge National Laboratory, Oak Ridge, TN, USA
G. Hugelius, P. Kuhry, A. B. K. Sannel, Stockholm University, Stockholm, Sweden
J. D. Jastrow, Argonne National Laboratory, Argonne, IL, USA
C. D. Koven, W. J. Riley, Z. M. Subin, Lawrence Berkeley National Lab, Berkeley, CA, USA
G. Krinner, CNRS/UJF-Grenoble 1, LGGE, Grenoble, France
D. M. Lawrence, National Center for Atmospheric Research, Boulder, CO, USA
A. D. McGuire, U.S. Geological Survey, Alaska Cooperative Fish and Wildlife Research Unit, University of Alaska, Fairbanks, AK, USA
J. A. O’Donnell, Arctic Network, National Park Service, Fairbanks, AK, USA
A. Rinke, Alfred Wegener Institute, Potsdam, Germany
K. Schaefer, National Snow and Ice Data Center, Cooperative Institute for Research in Environmental Sciences, University of Colorado, Boulder, CO, USA
J. Sky, University of Oxford, Oxford, UK
C. Tarnocai, AgriFoods, Ottawa, ON, Canada
M. R. Turetsky, University of Guelph, Guelph, ON, Canada
K. P. Wickland, U.S. Geological Survey, Boulder, CO, USA
C. J. Wilson, Los Alamos National Laboratory, Los Alamos, NM, USA
 S. A. Zimov, North-East Scientific Station, Cherskii, Siberia

WHAT: Interviewing and averaging the best estimates by world experts on how much permafrost in the Arctic is likely to melt and how much that will contribute to climate change.

WHEN: 26 March 2013

WHERE: Climactic Change, Vol. 117, Issue 1-2, March 2013

TITLE: Expert assessment of vulnerability of permafrost carbon to climate change (open access!)

We are all told that you should never judge a book by its cover, however I’ll freely admit that I chose to read this paper because the headline in Nature Climate Change was ‘Pandora’s Freezer’ and I just love a clever play on words.

So what’s the deal with permafrost and climate change? Permafrost is the solid, permanently frozen dirt/mud/sludge in the Arctic that often looks like cliffs of chocolate mousse when it’s melting. The fact that it’s melting is the problem, because when it melts, the carbon gets disturbed and moved around and released into the atmosphere.

Releasing ancient carbon into the atmosphere is what humans have been doing at an ever greater rate since we worked out that fossilised carbon makes a really efficient energy source, so when the Arctic starts doing that as well, it’s adding to the limited remaining carbon budget our atmosphere has left. Which means melting permafrost has consequences for how much time humanity has left to wean ourselves off our destructive fossil fuel addiction.

Cliffs of chocolate mousse (photo: Mike Beauregard, flickr)

Cliffs of chocolate mousse (photo: Mike Beauregard, flickr)

 How much time do we have? How much carbon is in those cliffs of chocolate mousse? We’re not sure. And that’s a big problem. Estimates in recent research think there could be as much as 1,700 billion tonnes of carbon stored in permafrost in the Arctic, which is much higher than earlier estimates from research in the 1990s.

To give that very large number some context, 1,700 billion tonnes can also be called 1,700 Gigatonnes, which should ring a bell for anyone who read Bill McKibben’s Rolling Stone global warming math article. The article stated that the best current estimate for humanity to have a shot at keeping global average temperatures below a 2oC increase is a carbon budget of 565Gt. So if all the permafrost melted, we’ve blown that budget twice.

What this paper did, was ask the above long list of experts on soil, carbon in soil, permafrost and Arctic research three questions over three different time scales.

  1. How much permafrost is likely to degrade (aka quantitative estimates of surface permafrost degradation)
  2. How much carbon it will likely release
  3. How much methane it will likely release

They included the methane question because methane has short term ramifications for the atmosphere. Methane ‘only’ stays in the atmosphere for around 100 years (compared to carbon dioxide’s 1000 plus years) and it has 33 times the global warming potential (GWP) of CO2 over a 100 year period. So for the first hundred years after you’ve released it, one tonne of methane is as bad as 33 tonnes of CO2. This could quickly blow our carbon budgets as we head merrily past 400 parts per million of CO2 in the atmosphere from human forcing.

The time periods for each question were; by 2040 with 1.5-2.5oC Arctic temperature rise (the Arctic warms faster than lower latitudes), by 2100 with between 2.0-7.5oC temperature rise (so from ‘we can possibly deal with this’ to ‘catastrophic climate change’), and by 2300 where temperatures are stable after 2100.

The estimates the experts gave were then screened for level of expertise (you don’t want to be asking an atmospheric specialist the soil questions!) and averaged to give an estimate range. For surface loss of permafrost under the highest warming scenario, the results were;

  1. 9-16% loss by 2040
  2. 48-63% loss by 2100
  3. 67-80% loss by 2300
Permafrost melting estimates for each time period over four different emissions scenarios (from paper)

Permafrost melting estimates for each time period over four different emissions scenarios (from paper)

Ouch. If we don’t start doing something serious about reducing our carbon emissions soon, we could be blowing that carbon budget really quickly.

For how much carbon the highest warming scenario may release, the results were;

  1. 19-45billion tonnes (Gt) CO2 by 2040
  2. 162-288Gt CO2 by 2100
  3. 381-616Gt CO2 by 2300

Hmm. So if we don’t stop burning carbon by 2040, melting permafrost will have taken 45Gt of CO2 out of our atmospheric carbon budget of 565Gt. Let’s hope we haven’t burned through the rest by then too.

However, if Arctic temperature rises were limited to 2oC by 2100, the CO2 emissions would ‘only’ be;

  1. 6-17Gt CO2 by 2040
  2. 41-80Gt CO2 by 2100
  3. 119-200Gt CO2 by 2300

That’s about a third of the highest warming estimates, but still nothing to breathe a sigh of relief at given that the 2000-2010 average annual rate of fossil fuel burning was 7.9Gt per year. So even the low estimate has permafrost releasing more than two years worth of global emissions, meaning we’d have to stop burning carbon two years earlier.

When the researchers calculated the expected methane emissions, the estimates were low. However, when they calculated the CO2 equivalent (CO2e) for the methane (methane being 33 times more potent than CO2 over 100 years), they got;

  1. 29-60Gt CO2e by 2040
  2. 250-463Gt CO2e by 2100
  3. 572-1004Gt CO2e by 2300

Thankfully, most of the carbon in the permafrost is expected to be released as the less potent carbon dioxide, but working out the balance between how much methane may be released into the atmosphere vs how much will be carbon dioxide is really crucial for working out global carbon budgets.

The other problem is that most climate models that look at permafrost contributions to climate change do it in a linear manner where increased temps lead directly to an increase in microbes and bacteria and the carbon is released. In reality, permafrost is much more dynamic and non-linear and therefore more unpredictable, which makes it a pain to put into models. It’s really difficult to predict abrupt thaw processes (as was seen over 98% of Greenland last summer) where ice wedges can melt and the ground could collapse irreversibly.

These kinds of non-linear processes (the really terrifying bit about climate change) made the news this week when it was reported that the Alaskan town of Newtok is likely to wash away by 2017, making the townspeople the first climate refugees from the USA.

The paper points out that one of the key limitations to knowing exactly what the permafrost is going to do is the lack of historical permafrost data. Permafrost is in really remote hard to get to places where people don’t live because the ground is permanently frozen. People haven’t been going to these places and taking samples unlike more populated areas that have lengthy and detailed climate records. But if you don’t know how much permafrost was historically there, you can’t tell how fast it’s melting.

The key point from this paper is that even though we’re not sure exactly how much permafrost will contribute to global carbon budgets and temperature rise, this uncertainty alone should not be enough to stall action on climate change.

Yes, there is uncertainty in exactly how badly climate change will affect the biosphere and everything that lives within it, but currently our options range from ‘uncomfortable and we may be able to adapt’ to ‘the next mass extinction’.

So while we’re working out exactly how far we’ve opened the Pandora’s Freezer of permafrost, let’s also stop burning carbon. 

Unprecedented: Melting Before Our Eyes

The volume of Arctic sea ice is reducing faster than the area of sea ice, further speeding the arctic death spiral.

WHO:  Seymour W. Laxon, Katharine A. Giles, Andy L. Ridout, Duncan J. Wingham, Rosemary Willatt, Centre for Polar Observation and Modelling, Department of Earth Sciences, University College London, London, UK
Robert Cullen, Malcolm Davidson, European Space Agency, Noordwijk, The Netherlands
Ron Kwok, Jet Propulsion Laboratory, California Institute of Technology, Pasadena, California, USA.
Axel Schweiger, Jinlun Zhang, Polar Science Center, Applied Physics Laboratory, University of Washington, Seattle, Washington, USA
Christian Haas, Department of Earth and Space Science and Engineering, York University, Toronto, Canada.
Stefan Hendricks, Alfred Wegener Institute for Polar and Marine Research, Bremerhaven, Germany
Richard Krishfield, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts, USA
Nathan Kurtz,School of Computer, Math, and Natural Sciences, Morgan State University, Baltimore, Maryland, USA.
Sinead Farrell, Earth System Science Interdisciplinary Center, University of Maryland, Maryland, USA.

WHAT: Measuring the volume of polar ice melt

WHEN: February 2013 (online pre-published version)

WHERE: American Geophysical Union, 2013, doi: 10.1002/grl.50193

TITLE:  CryoSat-2 estimates of Arctic sea ice thickness and volume (subs req.)

Much has been written about the Arctic Death Spiral of sea ice melting each spring and summer, with many researchers attempting to model and predict exactly how fast the sea ice is melting and when we will get the horrifying reality of an ice-free summer Arctic.

But is it just melting at the edges? Or is the thickness, and therefore the volume of sea ice being reduced as well? That’s what these researchers set out to try and find out using satellite data from CryoSat-2 (Science with satellites!).

The researchers used satellite radar altimeter measurements of sea ice thickness, and then compared their results with measured in-situ data as well as other Arctic sea ice models.

A loss of volume in Arctic sea ice is a signifier of changes in the heat exchange between the ice, ocean and atmosphere, and most global climate models predict a decrease in sea ice volume of 3.4% per decade which is larger than the predicted 2.4% per decade of sea ice area.

Sea ice area minimum from September 2012 (image: NASA/Goddard Space Flight Center Scientific Visualization Studio)

Sea ice area minimum from September 2012 (image: NASA/Goddard Space Flight Center Scientific Visualization Studio)

The researchers ran their numbers for ice volume in winter 2010/11 and winter 2011/12, and then used the recorded data sets to check the accuracy of their satellites (calibration for my fellow science nerds).

The most striking thing they found was a much greater loss of ice thickness in the north of Greenland and the Canadian Archipelago. Additionally, they found that the first year ice was thinner in autumn, which made it harder to catch up to average thickness during the winter, and made greater melting easier in summer.

Interestingly, they found that there was additional ice growth in winter between 2010-12 (7,500km3) compared to 2003-08 (5,000km3), which makes for an extra 36cm of ice growth in the winter. Unfortunately the increased summer melt is much greater than the extra growth, so it’s not adding to the overall thickness of the sea ice.

For the period of 2010-12 the satellite measured rate of decline in autumn sea ice was 60% greater than the predicted decline using PIOMAS (Panarctic Ice Ocean Modeling and Assimilation System). Most researchers when seeing results like that might hope that there’s an error, however when measured against the recorded data, the CryoSat-2 data was within 0.1 metres of accuracy. So while astounding, the 60% greater than expected loss of sea ice thickness is pretty spot on.

The researchers think that lower ice thickness at the end of winter in February and March could be contributing to the scarily low September minimums in the arctic death spiral, but the greatest risk here is that the ever increasing melt rate of ice in the arctic could take the climate beyond a tipping point where climate change is both irreversible and uncontrollable in a way we are unable to adapt to.

Visualisation of reduction in arctic sea ice thickness (from Andy Lee Robinson, via ClimateProgress)

Visualisation of reduction in arctic sea ice thickness (from Andy Lee Robinson, via ClimateProgress)

So as usual, my remedy for all of this is: stop burning carbon.