Press Release: Landsat 8 helps unveil the coldest place on Earth
SAN FRANCISCO, CA—Scientists recently recorded the lowest temperatures on Earth at a desolate and remote ice plateau in East Antarctica, trumping a record set in 1983 and uncovering a new puzzle about the ice-covered continent.
Ted Scambos, lead scientist at the National Snow and Ice Data Center (NSIDC), and his team found temperatures from −92 to −94 degrees Celsius (−134 to −137 degrees Fahrenheit) in a 1,000-kilometer long swath on the highest section of the East Antarctic ice divide.
The measurements were made between 2003 and 2013 by the Moderate Resolution Imaging Spectroradiometer (MODIS) sensor on board NASA’s Aqua satellite and during the 2013 Southern Hemisphere winter by Landsat 8, a new satellite launched early this year by NASA and the U.S. Geological Survey.
“I’ve never been in conditions that cold and I hope I never am,” Scambos said. “I am told that every breath is painful and you have to be extremely careful not to freeze part of your throat or lungs when inhaling.”
The record temperatures are several degrees colder than the previous record of −89.2 degrees Celsius (−128.6 degrees Fahrenheit) measured on July 21, 1983 at the Vostok Research Station in East Antarctica. They are far colder than the lowest recorded temperature in the United States, measured at −62 degrees Celsius (−79.6 degrees Fahrenheit) in Alaska, in northern Asia at -68 degrees Celsius (−90.4 degrees Fahrenheit), or even at the summit of the Greenland Ice Sheet at -75 degrees Celsius (−103 degrees Fahrenheit).
Scambos said the record temperatures were found in several 5 by 10 kilometer (3 by 6 mile) pockets where the topography forms small hollows of a few meters deep (2 to 4 meters, or 6 to 13 feet). These hollows are present just off the ice ridge that runs between Dome Argus and Dome Fuji—the ice dome summits of the East Antarctic Ice Sheet. Antarctic bases sit on each of the sites and are generally not occupied during Antarctic winters.
Under clear winter skies in these areas, cold air forms near the snow surface. Because the cold air is denser than the air above it, it begins to move downhill. The air collects in the nearby hollows and chills still further, if conditions are favorable.
“The record-breaking conditions seem to happen when a wind pattern or an atmospheric pressure gradient tries to move the air back uphill, pushing against the air that was sliding down,” Scambos said. “This allows the air in the low hollows to remain there longer and cool even further under the clear, extremely dry sky conditions,” Scambos said. “When the cold air lingers in these pockets it reaches ultra-low temperatures.”
“Any gardener knows that clear skies and dry air in spring or winter lead to the coldest temperatures at night,” Scambos said. “The thing is, here in the United States and most of Canada, we don’t get a night that lasts three or four or six months long for things to really chill down under extended clear sky conditions.”
Centuries-old ice cracks
Scambos and his team spotted the record low temperatures while working on a related study on unusual cracks on East Antarctica’s ice surface that he suspects are several hundred years old.
“The cracks are probably thermal cracks—the temperature gets so low in winter that the upper layer of the snow actually shrinks to the point that the surface cracks in order to accommodate the cold and the reduction in volume,” Scambos said. “That led us to wonder what the temperature range was. So, we started hunting for the coldest places using data from three satellite sensors.”
More than 30 years of data from the Advanced Very High Resolution Radiometer (AVHRR) on the NOAA Polar Orbiting Environmental Satellite (POES) series gave Scambos a good perspective on what the pattern of low temperatures looked like across Antarctica.
“Landsat 8 is still a new sensor, but preliminary work shows its ability to map the cold pockets in detail,” Scambos said. “It’s showing how even small hummocks stick up through the cold air.”
Scambos suspected they would find one area that got extremely cold. Instead they found a large strip at high altitude where several spots regularly reach record low temperatures. Furthermore, dozens of these extremely cold areas reached about the same minimum temperatures of −92 to −94 degrees Celsius (−134 to −137 degrees Fahrenheit) on most years.
“This is like saying that on the coldest day of the year a whole strip of land from International Falls, Minnesota to Duluth, Minnesota to Great Falls, Montana reached the exact same temperature, and more than once,” Scambos said. “And that’s a little odd.”
—Credit: Ted Scambos, National Snow and Ice Data Center
The scientists suspect that a layer in the atmosphere above the ice plateau reaches a certain minimum temperature and is preventing the ice plateau’s surface from getting any colder.
A physical limit
“There seems to be a physical limit to how cold it can get in this high plateau area and how much heat can escape,” Scambos said. Although an extremely cold place, Antarctica’s surface radiates heat or energy out into space, especially when the atmosphere is dry and free of clouds.
“The levels of carbon dioxide, nitrogen oxide, traces of water vapor and other gases in the air may impose a more or less uniform limit on how much heat can radiate from the surface,” Scambos said.
Scambos and his team will continue to refine their map of Earth’s coldest places using Landsat 8 data. “It’s a remarkable satellite and we’ve repeatedly been impressed with how well it works, not just for mapping temperature but for mapping crops and forests and glaciers all over the world,” Scambos said.
“The uses for Landsat 8 data are broad and diverse,” said James Irons, Landsat 8 project scientist at NASA’s Goddard Space Flight Center in Greenbelt, Md. “And Scambos’ work is an example of some of the intriguing science that can be done using Landsat 8.”
In the longer term, Scambos and his team will try to design weather stations and set them up in the area where the record temperatures occur to confirm the data from Landsat 8 and MODIS. Currently, most of the automated weather stations in the vicinity do not work properly in the dead of winter.
“The research bases there don’t have people that stay through the winter to make temperature measurements,” Scambos said. “We will need to investigate electronics that can survive those temperatures.”
One thing this makes clear, if it took 3 years to realize what the globes lowest recorded temperature is (was?) then we have no idea what the actual “global temperature” is now or was or will be or what effect Man had or has or will ever have on it.
When the temperature sunk into my head I started looking up the boiling temperatures etc of gases. and CO2 is solid at temp < -78C. So I knew you guys would be talking about that.
So now we can pump all the CO2 to Dome A and form dry ice naturally and just stack it up like firewood. 🙂
Is there anyway to industrialize this? 🙂
rgb,
I understand what you are getting at now and agree that it is possible and maybe even plausible. This would involve very small amounts, maybe even a chemisorbed mono layer, but then there is a lot of surface area on snow. I have done work on adsorption of water vapour on various surfaces and changes in latent heat are seen as additional layers are adsorbed. However usually the relative and not absolute humidity is the important variable. More research is needed. 🙂
wms
_Jim says:
December 10, 2013 at 12:01 pm
According to this curve , the vapor pressure of CO2 at ~-92ºC is about 250 torr. I’m not sure what the nominal atmospheric pressure on the Antarctic plateau is, but given its latitude and altitude, let’s call it ~550 torr. That means that for simple thermodynamic conditions, air content of CO2 would have to be about 45% (450,000 ppm) in order to solidify as a bulk solid.
Relative to Robert’s comment about Mars, based on Mars’ atmospheric composition of 96% CO2, average atmospheric pressure of 600 Pa (4.5 torr), and polar temperatures of about -125ºC, one can see why CO2 snow forms on Mars (but just barely).
The numbers quoted above are based on some quick googling. Remember, you get what you pay for. 🙂
Brian, Willis, everyone who has been discussing CO2 condensation/solidification at such a low partial pressure. . . . interesting discussion and thoughts. Made me do a lot of thinking. . . and my mind went in a wierd direction.
Ozone solidifies at only 10 degrees lower! Since no one has been actually measuring the surface temp over the years, and since ultra low temps would create a self perpetuating sink if auto-solidification plus radiation to space of the latent heat loss occurred . . . could this be an alternative explanation for the fluctuating “ozone” hole over the continent? Could we also actually have a “CO2” hole?
This location and these unique conditions just beg for detailed atmospheric gas concentration measurements on the ground! You don’t have to release weather balloons you just have to extend a jar on a long pole out of your igloo with your 14 layer gloves on, wait a little while, remotely cap the jar, then haul it back in and sample what you “captured” with a good Mass Spec in a vacuum chamber!
And Paul Westhaver, that is where my thoughts went next. What an ideal location for an air liquifaction plant. Raw material is most of the way already there! Course the energy to ship the final product to customers would most likely make it a non-starter, but at the very least, this site is ideal for “ultra” cold research.
Another thought . . . CO2 is quite a bit more dense than the other naturally occuring atmospheric gases, so at this low a temperature, it has to have a higher partial pressure down at the bottom of depressions than what one would normally expect at this elevation. I would expect CO2 to pool underneath other gasses driving up it’s natural concentration and partial pressure. And every molecule that condenses, releases latent heat which can radiate directly to space – no water to interfere – dropping the local temp even more. I wonder if the temp limit observed might just be due to both ozone and CO2 concentrations reaching near zero, thus stopping the latent heat loss driven cooling.
Oh, a simple explanation for the temperature floor effect?
http://www.cpc.ncep.noaa.gov/products/stratosphere/polar/gif_files/time_p_t90s.png
The stratosphere even at the poles rarely drops below 178K (~-95 Celsius), so getting temperatures below the tropopause to drop further would seem to be thermodynamically unfavorable, as most of the upper troposphere and lower stratosphere should now be capable of losing energy to the surface, wouldn’t it?
rgb
Note well that there has been a recent paper suggesting that we sequester solid CO_2 in the Antarctic because a block of solid CO_2 has a very, very long lifetime at -90 C and 1 atmosphere — that was the entire point of their argument.
= = = =
a) What exactly did they suggest we do with the sequestered CO2 throughout summer?
b) What does a “a very, very, long lifetime,” mean?
quote:
=========
Bottom line, 40g of dry ice placed in an open container at -82C completely sublimated overnight, while 27g of dry ice placed in a zip top bag retained 90% of its mass.
http://wattsupwiththat.com/2009/06/13/results-lab-experiment-regarding-co2-snow-in-antarctica-at-113%C2%B0f-80-5%C2%B0c-not-possible/
========
Come on all you scientists, don’t volcanoes have an effect on global temps too. And there are a few in Antarctica. Terrestrial and sub marine.
Interesting …. possible exemplification of a ‘temperature floor’ effect involving CO2?
.
Judging from a couple Mark 1 eye-balled mid-chart values, a 20:1 ratio exists between the densities of those two levels (@tropopause and mid-stratosphere) based on pressure alone; an elevated stratospheric temperature would reduce density of that layer further (since it would not seem to be constrained) … meaning the specific heat of the stratosphere could still be less than that at the tropopause … if my logic and reading of the graphs/charts holds …
.
another opinion:
10 Dec: Russia Today: Minus credibility? Antarctic record low temperature disputed
One Russian scientist calls this into question, pointing out the new record was made using remote measurements…
Yet it would be too early and inaccurate to recognize the temperature record now, Vyacheslav Martyanov, the head of the Russian Antarctic Expedition’s logistics center, told RIA Novosti.
“It is incorrect to declare a temperature record based on satellite data,” Martyanov said, explaining that AVHRR and MODIS are measuring the so-called ‘luminance temperature’, which does not fully correlate with true meteorological conditions and must be confirmed by observations on the ground.
“Air temperature is measured according to standards, at the height of 2 meters above the ground, like they do at meteorological observing stations, therefore recognizing a temperature measured by remote sensing is unrealistic,” Martyanov said…
Still, there is a solid chance to beat the 30-year-old record, believes Russian scientist, and it could be done by Chinese scientists at Kunlun stationed opened in 2009.
“It is uninhabited in winter time, but if they put a [automatic] meteorological station there – they can register a temperatures lower than at Vostok station,” Vyacheslav Martyanovtold RIA Novosti.
Kunlun station is situated in Antarctica’s highest region known as Dome A (Dome Argus)…
Dome A is located some 600 meters higher than Vostok station and scientists believe that temperature here might fall as low as -102°C (-152F)…
http://rt.com/news/antarctica-temperature-record-questioned-922/
——————————————————————————–
So the smaller the volume of a dry ice flake, the faster it will sublime, and the slower it will accumulate new CO2 from the atmosphere. So in theory, there is a maximum size that a dry ice blob can grow to at some Temperature below the triple point. I’m guessing that the size is likely too small to be visible.
And I would completely agree. Too small to be visible is not too small to have an impact on air temperatures or on the net GHE, and the size is almost certainly variable with ionic nucleation sites. In fact, I found a recent dissertation on adsorption and desorption of interstellar ices on dust grain analogue surfaces. The physics is not simple. Bear in mind that -93 C is 180 K — that’s pretty cold.
So now we can pump all the CO2 to Dome A and form dry ice naturally and just stack it up like firewood. 🙂
You laugh, but this is a no-shit-sherlock proposal that was recently examined in a peer-reviewed scientific paper. Solar powered CO_2 precipitators in the antarctic that would deposit solid CO_2 and move it deep into the glacial ice where it would presumably sit for a long, long time, safely sequestered. I didn’t read the paper as I’ve already got several unread science fiction novels on my tablet to go through first.
rgb
Well, there you go, eh? So if a mini ice age occurs in the Northern Hemisphere again, don’t throw away your Ugg boots? Yet!
King of Cool says:
December 10, 2013 at 1:29 am
The Australian Antarctic resupply ship Aurora Australis arrived back in Hobart 07 Dec three weeks behind schedule after encountering heavy sea ice into and out of Davis base.
———————————————————————————————————-
The Antarctic sea ice index is still showing record or near record ice every day. over the last 60 days. The sea ice anomalies are above average for ice concentration also. There are two large sections of the perimeter that are 50% above average concentration. So not only is the sea ice extent still high, the concentration is also high. Is it possible that the Antarctic sea ice might start retaining a growing amount of that ice from year to year? Could this be a part of the mechanism for the cooling phase that occures with the advent of one of the long term cycles? I would think that such a late melt at those latitudes would add some extra cooling.
Max™ says:
December 10, 2013 at 4:41 am
“2. The ppm of CO2 in the hollows isn’t going to change enough that I would even hazard trying to work it out, gases just don’t work that way, Samuel C Cogar.”
—————
Are you telling me that said denser air at the surface does not contain a greater number of CO2 molecules per unit volume ….. and which is subject to being “trapped” by either fresh snowfall or blowing snow …. and subsequently being “counted” by ice core researchers …. thus resulting in a highly inflated atmospheric CO2 ppm estimate?
And I was wondering, how does one know if the “layer of ice” …. in the ice core ….. was the result of a snowfall(s) or the result of wind-blown snow?
——————-
And I liked what Robert B said so I will include the following:
Robert Brown says:
December 10, 2013 at 8:47 am
“I wouldn’t be surprised at all if the cause of the consistent “clipping” of the temperature is that CO_2 is misting out of the atmosphere at that temperature in these pockets, accumulating as CO_2 “snow” that further accretes CO_2 frost on the surfaces.
This could easily be tested, of course. Simply grab a sample of the surrounding surface snow after the temperature has held steady at the CO_2 freezing point for a week or two and measure its CO_2 concentration when it melts. ”
“The measurements were made between 2003 and 2013”
And, pray, why were they not reported until now!!! Shameful.
A sheet of heavy construction paper would suffice as shielding. Can’t get any easier than that! GK
The air pressure at that altitude is around 600~700 millibars anyways, and gases diffuse so any instantaneous increase in concentration represents a decrease elsewhere, so it is unlikely to be anything more than a brief fluctuation.
Additionally the CO2 and other gases trapped in ice cores are generally able to mix with the atmosphere freely until the snow is compressed to a certain point and traps bubbles.
Wind blown snow will probably be lighter, while the layers compacted into ice are generally from the year before and had some melting occur on top, letting water filter down inside and glue the snow together more effectively until sufficient accumulation compresses it into ice.
____________
On a lark I decided to work out what latitudes would get solar heating in the stratosphere even in winter.
Go go gadget trig! Take the polar radius to be ~6350 km, then a 90 degree tangent from the arctic circle intersects a cone from the center of the planet with sides ~6400 km long around 75 Degrees South.
Everywhere further north than there can receive stratospheric heating during the local winter, locations further south are unlikely to get direct solar heating in the stratosphere outside of spring/summer/fall.
The domes and ridges in the story are around 77~79 S, hence the regular plunge in stratosphere temperatures, and I find it unlikely to be coincidental that the stratosphere and these surface cold traps both drop to almost the same temperature before bottoming out.
To rgbatduke & G. Karst:
The plutonium-238-powered RTGs on the Voyager space probes
are still going strong!
The coldest place on earth is not quite cold enough to solidify CO2. But this discussion leads me to wonder whether solid CO2 is likely a part of the Snowball Earth geography. This potential CO2 phase transition during ice ages can answer some puzzles about paleoatmospheres.
See WUWT Oct 6, ‘2013 We Must Get Rid of the Carboniferous Warm Period
Willis’ chart makes clear how the partial pressure of CO2 affects the temperature of sublimation. Could the glaciation of the Proterozoic and Ordovician have had conditions where a CO2 rich atmosphere (partial pressures of 0.1 to 2.0 atmospheres) could lean out precipitating dry-ice onto the thick ice caps? (rgbatduke at 2:11 pm also mentions the Ordovician-Silurian transition in the context of possible CO2 deposition and sublimation).
I can imagine that once the dry ice sublimes, a positive feedback can occur by raising the CO2 concentration in the atmosphere and raising temperature to compensate for the raising of the sublimation temperature at higher partial pressures. So the Silurian and Carboniferous could have had thick atmospheres, rich in CO2. Perhaps CO2 was sublimating at the poles while plant life was burying carbon into the massive coal beds of the Carboniferous age.
The source for all that carbon in Carboniferous coal has been a puzzle for me. The gigantism of flora and fauna in the Carboniferous and Mesozoic, particularly with respect to large fliers, leads me to challenge the assumption of near-constant atmospheric pressure. A big problem is once the atmosphere is at 1 to 2 atm in the Cambrian, how can you reach 5 atm in the later Paleozoic? Is a dry-ice reservoir on the polar ice caps a potential way of parking a large CO2 source?
The counter argument is that it would have to exist (at least on and off) almost half a billion years between Snowball (or earlier) and the Carboniferous and maybe the late Mesozoic. More CO2 would have to freeze out during the polar winters than could sublime in the summers. Latent heat budgets coupled with IR radiation and emissivity of dry ice might hold some surprises.
RE: Philip Mulholland :Oct 8, 2013 at 2:07 pm
How would it be possible for CO2 concentrations to fall during an ice age when growing seasons are short and growing regions are restricted to the lower latitudes? Do we have algal blooms in the tropical oceans sequestering carbon. If so, how does the CO2 return when it warms? Can that much more CO2 be dissolved in the oceans with today the bulk of the oceans are only a couple degrees above freezing?
Is it possible that CO2 was solidifying on the high ice caps as recently as the Paleocene Ice Ages? Solid CO2 would acting as a reservoir to quickly sublime CO2 back into the atmosphere when the warming is started?
The solidification of CO2 at partial pressures less than 0.0004 atm may require colder surface temperatures than can be attained, even on a 3-4 mile high ice plateaus in the depth of an ice age. Given the cold temps reported today, is it out of the question?
(Thinking out loud…) Alternatively, could CO2 snow nucleate on cold-dust particles a mile or two in the atmosphere? Then fall to bring a colder regional downdraft to the surface? Hmmm…. we’d have to see layers of dust particles left behind after the CO2 sublimes. We probably don’t see that.
Max™ says:
December 11, 2013 at 11:24 am
“The air pressure at that altitude is around 600~700 millibars anyways, and gases diffuse so any instantaneous increase in concentration represents a decrease elsewhere, so it is unlikely to be anything more than a brief fluctuation.
Additionally the CO2 and other gases trapped in ice cores are generally able to mix with the atmosphere freely until the snow is compressed to a certain point and traps bubbles.
Wind blown snow will probably be lighter, while the layers compacted into ice are generally from the year before and had some melting occur on top, letting water filter down inside and glue the snow together more effectively until sufficient accumulation compresses it into ice.”
————-
Your obfuscations simply amaze me.
And it is obvious to me that you have never traveled or lived in “snow country” and thus have never experienced the “effects” of wind-blown snow, …… specifically in relation to its potential for creating “drifts” and/or accumulating mass quantities in surface depressions. .
Has nothing to do with obfuscation, gas diffusion is a thing which involves the statistics of large numbers of molecules, so there can be local increases and decreases in concentration, freshly wind blown snow hasn’t been compacted for a year like the layers which remain after a prior melt season, water filtering down into the lower layers and freezing is indeed a significant part of the process by which gas bubbles become sequestered in ice, and though it has no relevance to my point because your anecdotes regarding “the effects of wind-blown snow” don’t matter*, and anyways you’re just a little mistaken as I’ve been through all but a handful of the 48 contiguous states and experienced feet of snowfall a few times in my life.
*”The plural of anecdote is not data.”
“…. freshly wind blown snow hasn’t been compacted for a year like the layers which remain after a prior melt season, water filtering down into the lower layers and freezing is indeed ….”
And just what is the temperature of the snowpack that resides on the surface of glaciers in Antarctica?
Is it >1C or 33F?