Uh, oh – Ice cores used for paleoclimatology may have been altered by bacterial action from within

From the UNIVERSITY OF YORK and the “that’s gonna leave a mark” department.

For the first time scientists have directly observed living bacteria in polar ice and snow — an environment once considered sterile

For the first time scientists have directly observed living bacteria in polar ice and snow – an environment once considered sterile. The new evidence has the potential to alter perceptions about which planets in the universe could sustain life and may mean that humans are having an even greater impact on levels of CO2 in the Earth’s atmosphere than accepted evidence from climate history studies of ice cores suggests.

Section of the West Antarctic Ice Sheet ice core sample with a dark ash layer. Credit: Image courtesy Heidi Roop.

Gases captured and sealed in snow as it compresses into ice can provide researchers with snapshots of the Earth’s atmosphere going back hundreds of thousands of years. Climate scientists use ice core samples to look at prehistoric levels of CO2 in the atmosphere so they can be compared with current levels in an industrial age.

This analysis of ice cores relies on the assumption that there is limited biological activity altering the environment in the snow during its transition into ice. Research reported today in the Journal of the Royal Society Interface, which has directly observed microbial activity in Antarctic and Arctic snow, has revealed that the composition of these small samples of gas trapped in the ice may have been affected by bacteria that remain active in snow while it is being compressed into ice – a process that can last decades.

Lead author of the research Dr Kelly Redeker from the Department of Biology at the University of York said “As microbial activity and its influence on its local environment has never been taken into account when looking at ice-core gas samples it could provide a moderate source of error in climate history interpretations. Respiration by bacteria may have slightly increased levels of CO2 in pockets of air trapped within polar ice caps meaning that before human activity CO2 levels may have been even lower than previously thought”.

“In addition, the fact that we have observed metabolically active bacteria in the most pristine ice and snow is a sign of life proliferating in environments where you wouldn’t expect it to exist. This suggests we may be able to broaden our horizons when it comes to thinking about which planets are capable of sustaining life,” Redeker added.

Research conducted in laboratories has previously shown that bacteria can stay alive at extremely cold temperatures, but this study is the first time that bacteria have been observed altering the polar snow environment in situ.

The researchers looked at snow in is natural state, and in other areas they sterilised it using UV sterilising lamps. When they compared the results the team found unexpected levels of methyl iodide – a gas known to be produced by marine bacteria – in the untouched snow.

Cutting-edge techniques enabled the researchers to detect the presence of gases even at part-per-trillion levels, one million times less concentrated than atmospheric CO2 concentrations.

The researchers worked on sites in the Arctic and Antarctic and took precautions to limit the impact of sunlight and wind, using tarpaulins to protect their sample sites and positioning themselves on the middle of a glacier away from soil and other forms of polar wildlife which might contaminate the snow.

The results of the study also suggest that life can be sustained even in remote, cold, nutrient poor environments, offering a new perspective on whether the frozen planets of the universe could support microorganisms.

With more research, astrobiologists working to identify planets in the universe with temperature levels that could allow for the presence of liquid water may be able to expand the zones they consider potentially habitable to include planets where water is found as ice.

“We know that bacteria have the potential to remain viable and metabolically active at low temperatures for hundreds to thousands of years,” said Redeker. “The next step is to look further down to see if we can observe active bacteria deep in the ice caps,”


The paper (preprint): http://eprints.whiterose.ac.uk/124852/1/Redeker.pdf

“Microbial metabolism directly affects trace gases in (sub) polar snowpacks” is published in the Journal of the Royal Society Interface.


Concentrations of trace gases trapped in ice are considered to develop uniquely from direct snow/atmosphere interactions at the time of contact. This assumption relies upon limited or no biological, chemical or physical transformations occurring during transition from snow to firn to ice; a process that can take decades to complete. Here, we present the first evidence of environmental alteration due to in situ microbial metabolism of trace gases (methyl halides and dimethyl sulphide) in Polar snow. We collected evidence for ongoing microbial metabolism from an Arctic and an Antarctic location during different years. Methyl iodide production in the snowpack decreased significantly after exposure to enhanced UV radiation. Our results also show large variations in the production and consumption of other methyl halides, including methyl bromide and methyl chloride, used in climate interpretations. These results suggest that this long neglected microbial activity could constitute a potential source of error in climate history interpretations, by introducing a so far unappreciated source of bias in the quantification of atmospheric+derived trace gases trapped within the Polar ice caps.


166 thoughts on “Uh, oh – Ice cores used for paleoclimatology may have been altered by bacterial action from within

  1. I remember after the Salby lecture in Edinburgh some dimwit who claims to be a physicist criticised him heavily for suggesting that CO2 levels in the ice-core might not reflect the actual level that was in the atmosphere.

    Seems like Salby is yet again vindicated and someone, if they had any integrity, will yet again admit they are wrong.

    • Scottish Sceptic,

      What the theory of Salby implied was that CO2 levels at the peaks were order(s) of magnitude higher during the peaks, which diffused over time. But each interglacial, that should have been another order of magnitude higher, as the dissusion had twice the time to spread over the lower levels and the lower levels were even (much) lower, until even negative, which is not possible…

      • True, Salby proposed a completely different method. However, why are you saying that it would have implied negative levels? The changes would have flattened to a constant until you have no signal but just noise.

      • Ben,

        If Dr. Salby was right, the CO2 levels were 10 times higher in the previous interglacial, that is 3000 ppmv during roughly 10% of the full period, spreading over the glacial periods, that is 90% of the time. As we measure about 300 ppmv during an interglacial and below 200 ppmv during glacial periods, from the original 3000 ppmv some 2700 ppmv CO2 must have migrated over 90% of the period, or about 300 ppmv extra. Thus the original CO2 levels were negative…

        Even if the peaks were (much) lower, each peak in the past must have been higher and higher, to result in the same peak today, or they should be flatter and flatter over each 110,000 period back in time, thus reducing the original CO2 levels in the glacial periods.

        In reality, the CO2/temperature (proxy) ratio between interglacial and glacial periods is the same about 8 ppmv/K over all periods and there is no measurable CO2 migration at -40ºC in the (c)oldest ice cores.

      • Bart,

        Do you have any indication that CO2 in ice at -40ºC still migrates to any detectable level? Not much water left at that temperature between the crystals, except around dust inclusions, which are quite low in Antarctic ice…

      • While I think Salby’s unpublished speculations seem to require some very large stretch of the imagination, I don’t follow how you get to negative CO2.

        If this research is correct and CO2 was “even lower than previously thought”, then your 8ppm/ degree needs revising.

        I always find starting a scientific claim with ” in reality” to prove you are right is very unconvincing. It’s the equivalent of calling yourself a “realist” and imagining that means everything you say is correct.

      • Greg,

        The 8 ppmv/K is what is measured for every interglacial/glacial ratio in CO2 and δ18O,δ.
        There is no observed change in ratio over the past 800,000 years, thus either there was no migration or the peaks were increasingly higher in the past, as the migration had multiples of 110,000 years to level out.

        CO2 in the ice cores, if it migrates, does spread over adjacent layers over very long periods, with little chance to escape to the atmosphere. Thus if Dr. Salby was right with his claim that the interglacial peaks were 10 times higher than measured (a claim he didn’t repeat in later presentations), then that was originally 3000 ppmv. As we measure 300 ppmv today, the difference of 2700 ppmv must have spread over the glacial period, or average 300 ppmv over a full period. But we only measure 180-200 ppmv during glacials, thus IF the original peak was 3000 ppmv, the original CO2 levels in glacial periods before diffusion must have been minus 100 ppmv…

        Even if the original peak was only 10% higher during the previous interglacial, then the peak in the interglacial before that must have been 21%, as the migration speed depends of concentration and the duration doubled, ending at the same concentration in the ice core. Thus for each peak farther in the past, the original CO2 level must have been higher and higher…

    • First off, they should recognize that the experts on ice cores say that 30–50% of the CO2 is lost during extraction of ice cores as there is massive decompression and extensive micro fracturing. Back calculating the CO2 with an average of 40% losses and you have atmospheric CO2 the same as or higher than CO2 today.

      With the new bacteria factor, we could assume some of their contribution might be countered by the losses. We are back to CO2 concentrations that are egregiously and alarmingly low if the are <300 ppm.

      Nonetheless, we have a CO2 trend, not anything close to the absolute concentrations that the IPCC "scientists like to pretend they are. And the trend changes clearly show a lag of about 800 years that follows changes in Earth's climate.

      It is ingenuous to assume that this might reflect a greater influence of man's activities on atmospheric CO2. It is clear that human emissions have been growing logarithmically while the atmospheric CO2 has been going up linearly. We are having no effect on CO2 concentrations, regardless of what the ice cores tell us. The oceans hold up to 50 times the CO2 of the atmosphere, which dwarfs our human input.

      • Higley – they don’t measure the CO2 content of the ice core. They measure the CO2 content of the air trapped in the ice core. Losing air during handling should not affect the CO2 content of the remaining air. Bad logic there, son. It’s not like the 0.04% of CO2 will leak out and the 99.96% of N2, O2 and Argon will miraculously stay in place.

        Salby’s point was about diffusion (I think) between air bubbles in adjacent layers of ice. Probably true, but “orders of magnitude” seems like a bit of a reach. He’s not a very good poster boy for the skeptical position.

      • “It’s not like the 0.04% of CO2 will leak out and the 99.96% of N2, O2 and Argon will miraculously stay in place.”

        Think it through more carefully. Assume it was originally 0.3% CO2, 99.7% other, and it will try to equilibrate with today’s 0.04% CO2, 99.96% other. It does, indeed, affect the CO2 portion disproportionately.

      • Bart and Smart Rock,

        – An ice core from the depth is stored for over a year at the drilling site at -20ºC for relaxation of the ice. The ice expands with some 50% during that time. If clathrates are present, O2 and N2 clathrates decompose first and if that forms cracks, the high pressure of 99.7% N2 and O2 will escape first. CO2 clathrates decompose (much) later, whithout giving any appreciable pressure, If there was a change in CO2 / air ratio, that would give too high CO2 levels, never too low.

        – As CO2 levels found in ice cores are about between 180 and 310 ppmv and the outside world during drilling and extraction up to measurement time are 350-400 ppmv, if there was any migration during storage, the CO2 levels would go up, not down. There are no differences on the same ice core, re-examining after many years of storage.

      • Different gases will expand to different degrees and therefore leak out at different rates. Unless you are taking into account varying levels of emission from the bubbles the gas concentration measurements will be inaccurate.

      • It depends what this “leak out” claim is supposed to be physically. I doubt it is an open mixing with surrounding air allowing diffusive mixing.

    • Another assumption that is part of the analysis of ice cores is that all of the CO2 in gas bubbles are extracted and not lost to ice surfaces or clathrates, etc. any time from formation to analysis. To my knowledge, no one has validated this assumption by reproducing cores under realistic pressure, temperature and atmospheric levels. Of course, it would not be possible to replicate anything but short times of formation or storage.

      • R. Shearer,

        As said above, O2 and N2 escape first, CO2 later, so if clathrates are present and/or cracks in the ice, that would lead to too high CO2 levels, never too low. The same for solubility: CO2 is more soluble in any liquid water than O2 or N2, thus leading to too high levels, not too low if that was a problem at all.

    • Anyone that has any experience with wintertime snowfall ….. and possesses common sense thinking and intelligent reasoning skills, …… knows for a fact, …… that the “quantity” of entrapped air molecules in a layer of snow residing on the surface is highly (99.999%) dependent upon “the rate of accumulation” of said snow layer.

      Blizzard snowfall conditions will quickly create an extremely thick layer of snow with a large quantity of entrapped air molecules.

      Clear or cloudy skies and high wind conditions will also create an extremely thick layer of snow with a large quantity of entrapped air molecules, depending upon the size/consistency of the snow particles..

      A light snowfall occurring over several days or weeks will slowly create a layer of snow with variable quantities of entrapped air molecules in various locations.

      The leading edge of a “windblown” snow drift is more highly compact than the trailing edge of said drifting snow.

      And when drilling “ice core” holes one just has to settle for “pot luck” of what the drill hole yields.

  2. The quantification of the alteration will certainly incite quite a debate. Blown over by one side and under by the other, I bet.

    This is quite interesting and does answer one of the many questions I’ve always had about ice cores and paleoclimatology.

    Thanks for the post!

  3. …Respiration by bacteria may have slightly increased levels of CO2 in pockets of air trapped within polar ice caps meaning that before human activity CO2 levels may have been even lower than previously thought”

    Doesn’t that imply CO2 sensitivity may be much lower than assumed?

      • and after finding thriving living organisms is previously sterile environments they are quite certain that there’s no way that life could have managed to propagate from the dirty soil outside the edges of large glaciers or snow fields to the now certifiably sterile middle of those same ice and snow fields.

        There’s simply no way that microbes could have gotten inside their tents.


      • Not only is it worse than we thought, it is worse than we thought back when we thought it was worse than we what we originally thought…etc.

    • The paper, however claims to have found trace amounts of methyl iodide, NOT CO2. It is an unwarranted stretch to state that microbes could be accounting for increased CO2.

      Damn good thing that all this unsettling science has been settled.

      • Also…

        1) The primary sources of methyl iodide in the hydrosphere are algae and kelp – CO2 consumers.

        2) Methyl iodide is a compound that dissociates when exposed to even visible light. Ultraviolet destroys it even faster – which means that they were destroying already present methyl iodide in the irradiated sample, and not in the “control.”

        3) Very shoddy “science.”

      • LOL. Thanks, Nick.

        Some day they may try to get science undergrads to write the press releases instead of “media studies” AWG zealots with NO scientific training or knowledge.

    • Yes, especially since 2H and 18O concentrations in present-day snow are distinctly lower than levels in the ice of previous warm periods, when CO2 might have been even less “than previously thought”.

    • Sorry. Just means that end is coming sooner than previously thought (again) AND it will be WAY more painful.

  4. I love the jump to the conclusion that CO2 may have been lower. They couldn’t possibly entertain the idea it may have lowered CO2 levels.

    • It sounds like the bacteria produce co2. Trapped bacteria would elevate the measurements of co2 ppm in a bubble. This would mean that actual co2 levels of the atmosphere would have been lower than what we measure.

      • They haven’t classified the bacteria yet. However, I would agree that it’s more likely that they consumed rather than produced oxygen, as photosynthesis would be much more noticeable.

      • Perhaps, but when the bacteria dies what is left behind? Only CO2? What chemical reaction occurs between those molecules and any CO2 after 10000 years? Does some or most of the CO2 recombine? I don’t think they attempted to find out. It is no longer a pristine sterile environment, therefore ALL assumptions derived from that are now null and void. It could be more it could be less why should we assume either.

    • Since CO2 is far more soluble in water than N2 and O2, we also have to allow for differential gas diffusion through the ice.

      • All of those gases you mention; and in particular the CO2, since that’s what climatologists are interested in knowing above all else, have a segregation coefficient that describes the redistribution of solutes between phases of a solvent.

        Everybody (well almost everybody), is well aware that when water containing various salts and other dissolved molecules changes phase to a solid phase (ice) or to a gaseous phase (steam), that the dissolved molecules redistribute themselves between the two phases.

        Gases for example follow a Henry’s law relationship between the liquid water phase, and the gaseous atmospheric phase, so that warmer water gives up CO2 to the atmosphere , while colder water, will accept more CO2 from the atmosphere into solution.

        At a moving liquid/solid interface, most impurity molecules are preferentially excluded form the solid phase, preferring to remain dissolved in the liquid phase.
        This is an extremely valuable method of purification of crystalline materials, as in some cases, dissolved impurities can be rejected down to six or seven nines purity (99.9999 or 99.99999 %) by crystallization. I happen to know of a company (now defunct) that used to make its own seven nines purity Gallium metal out of scrap gallium arsenide material including grinding dust waste simply by crystallizing it after simple chemical cleaning solvents were used to remove sheer gunk.

        And when using that same purified Gallium along with pure arsenic to form a liquid GaAs melt and a subsequent gradient freeze Horizontal Bridgeman crystal growth process, the remaining impurities in the melt are selectively swept along as the solid/ liquid interface moves to the final butt end of the resulting single crystal ingot, which is then cut off and recycled once more.

        So when snow on the surface melts and refreezes, while compacting to ice, which may take years or decades to complete, it is a fairly safe bet that the dissolved impurities in the water will not all be found in the solid ice.


    • One question that I would have is, “Can bacteria form whole colonies in such a way as to create vein structures within the ice, cracks, micro-canals, or some other type of transport networks that might increase diffusion?” And could these combine with OTHER factors to reduce ice stability over hundreds of thousands of years? Could a somewhat stable process have occurred, involving these and other factors, in a very consistent fashion to give the consistent readings regarded as facts today?

      Remember, hundreds of thousands of years — AIR, ICE — fluid change.

  5. Another bad starting assumption throws decades of work into question. Before beginning actual work, always question every assumption and try to find a way to prove them!

    • Don’t jump to conclusions, we haven’t heard Michael “Tree Ring Counter” Mann’s take on this. Now, he’s a bit busy in that he is also now apparently a polar bear expert, so we may have to wait, but hey, counting tree rings isn’t much different than counting polar bears, so counting ice rings can’t be too hard for a genius, can it?

  6. “meaning that before human activity CO2 levels may have been even lower than previously thought””

    Funny I thought life was pretty much extinguished below around 180ppm, when did that happen and why didn’t we notice it?

    • I was thinking the same thing. It’s hard to imagine C3 plants surviving at lower CO2 concentrations than at a level they were already stressed and nearly suffocating in.

  7. Oh Boy!

    Here is another opportunity for the individuals with letters following their names to apply adjustments to the long term record. I foresee the CO2 levels in the past being adjusted downward giving a noticeable steepening to the slope of the line in the present.

  8. Again with another WUWT article, this is not a new controversy.

    I looked into this seriously at least seven years ago, when any thought of entertaining doubt about ice-core records automatically qualified professionals in any related field as intellectual criminals. Seemingly this was one area of doubt that would get any skeptic burned at the stake.

    Anyhow, here’s the article I did specifically on this topic, at least seven years ago:


    Glad to see the controversy reborn with some new sparks in 2017.

    • Robert,

      Please, let the late Dr. Jaworowski rest in peace, together with his ideas about ice cores. Completely outdated and completely wrong. See:

      Ice cores from Greenland are not suitable for CO2 levels, as they may produce in-situ CO2 from the reaction of sea salts and frequent acid dust from nearby Icelandic volcanoes. Antarctic ice cores have orders of magitude less dust deposits and volcanoes are far away…

      • I would think that Antarctic ice cores are no further away from Antarctic volcanoes, than are Greenland ice cores from Iceland volcanoes.


      • George E. Smith,

        The distance from mid Iceland to mid Greenland is about 1100 km, from Mt. Erebus to the South Pole about 1400 km, thus not that much difference.
        More important is the difference in types: Mt. Erebus is a Stromboli type volcano, whith near continuous small eruptions, seldom reaching the higher atmosphere, while Icelandic volcanoes can and do reach the stratosphere.
        As far as I remeber, katabatic winds prevail in Antartica, blowing volcanic dust away from the continent.

        Anyway, dust deposits in Antarctic ice are (an) order(s) of magnitude less than in Greenland ice.

      • Many other Antarctic volcanoes, much closer to ice core extraction sites, may have been active in the past. Also, it is important to know, what type of bacteria authors of this article have discovered. If at least some of those are anaerobic, for example, the whole assertion about them producing additional carbon oxide may be in question. And what about bacteria working at it now, in surface layers?

      • Alexander Feht,

        Indeed, there may have been several of the now dormant volcanoes that were active in the past. Point is that the amounts of dust deposits in Antarctic ice cores are very low, including bacterial strains which, not by coincidence, are at the same depth as the dust. Even if all CH4 or N2O was used/produced by bacteria at the cost of CO2, that would give a difference of less than 1 ppmv in the CO2 levels.

  9. I take it the study wasn’t from Climate Science as someone actually bothered to check their controls and assumptions.

  10. You mean to tell me they are just now checking and finding this out! What Amtrak management team is running this operation?

  11. Great…they discovered aerobic bacteria in ice cores….at this rate they are several hundred years away from discovering what happens when it goes anaerobic…reduces CO2

  12. Bacteria can oxidise or reduce carbon. But they can’t create carbon. The only way they could be altering CO2 is if some carbon-containing substrate is present, which presumably can be tested.

    • Some of the microbes themselves are made of carbon and bring the carbon Nick :-)

      These ice-organisms are adapted to growing on the ice surfaces and within a labyrinth of channels and pores that permeate the ice floes.

      Black smokers at the bottom of the ocean work the same way

      The bacteria provide carbon for the worm and in return the worm provides oxygen for the bacteria (to oxidise the hydrogen sulphide)

      It is very common for symbiotic relationships to develop in extreme enviroments.

      • Read the article Nick it’s in the title “Algae and bacteria”, get it there are two species in the ice and in smokers it’s another two one being worms. Yes the residues should have been detected if a scientists does his job which is what this paper does.

        So the obvious step for anyone to do is ask the climate scientists what steps they did to eliminate this.You being an advocate won’t but any real scientist would because they need to recalibrate their results if they didn’t. Some in the anti-cagw extreme will argue this only got published because it is going to take make historic CO2 levels lower.

        Neither of you two crazy extreme groups care about the science you just care about the result.

      • “Read the article”
        Your article is about sea ice – a totally different situation. This paper does not detect residues. It detects traces of microbial metabolism (not microbes or their remains) in snow, not ice. They speculate that the microbes could still be alive in the ice. Neither of their snow sites were where people drill for ice cores; one was on Svalbard, not far from the town of Longyearbyen, the other was on Signy Island, at just 60°S. They did no analysis of icecores. The paper itself does not say that CO2 levels in icecores may be affected; just methyl halides.

      • Again Nick you get microbes at 2.8Km down in what we call rock. The thing that was expected to stop them was temperature if you have an extreme species you will have microbes living there with ease. Science is about checking controls :-)

      • “Nothing can function if it is literally frozen solid.”……..

        They are called psychrophiles, and yes, they do

    • Dust and pollen, plenty of carbon there for a microorganism. And you say there are no worms, but up until now you were certain there were no microbes either.

    • Nick Stokes what don’t you understand of: “These results suggest that this long neglected microbial activity could constitute a potential source of error in climate history interpretations” ?

      • No he isn’t accepting it, he is in denial :-)

        The funny part is based on this work the CO2 levels would be lower in the paleo record so it’s not clear why other than Climate Scientists made a mistake which as a CAGW activist he doesn’t want to admit.

    • I’m with Nick on this one. If bacteria are present in such a small quantity to have been undetected for so long, I’m questioning whether there are enough of them to meaningfully change the results. After all, even if they are self destructive and end up eating each other to stay alive, eventually converting all bacterial carbon into CO2 (hyperbolic, I know, but work with me), would it meaningfully change CO2 concentrations? Maybe a ppm or two, but these are scarce bacteria with a very low volume of carbon to begin with.

      Now, it could throw off the third or fourth significant figure, but the more I think about it, the more I think it will be a widening of the error bars instead of a large step change.

      • Don’t be so quick to judge. They discovered it with gas detection technology down to the parts-per-trillion levels, and it takes 10+ years for snow to compact into ice, rendering the gases “trapped”, AND they discovered these microbes in snow that has already been compacted into ice, so there’s no guarantee the process doesn’t continue after compaction. Their test presumably only took a few weeks to a month at most, and they detected gas changes at the parts-per-trillion level… now multiply that by 50-100 per year, for 10 years, you’ve got parts-per-billion changes, in 10000 years time, you’ve got levels that are in the parts per million levels, or in the same orders of magnitude of CO2.

      • Jeremy, once it’s trapped, there’s no carbon for them to eat. It’s a closed system. They starve themselves. It doesn’t matter how many millions of years, it won’t go up once all the food is eaten.

      • You miss microbes and algae can move thru the ice, it’s not solid to them there are channels and pores that they can move thru. Even on land you can pick up deep microbes at 2.8Km down the upper limit is expected to be 7km and that is what we call rock.

        As you see back in 1926 it was assumed they were buried up to 300 million years ago on later works showed the suckers can get down there thru the solid rock and are pretty much disconnected from the rest of life as we know it.

        The temperature was what they thought excluded the suckers in the ice core and it looks like they have found extreme types. I imagine it will fire some in that area to go study what species they have and how far down they go.

      • LdB,

        The bacteria types were already examined in the 2004 study above and even the most extreme types only survived at -40ºC by reducing their “life” to only DNA damage repair. That is all they could perform.

      • It is already accepted that microbes can replicate down to at least -39degree C

        Microbial activity has been measured in soils frozen below −39 °C

        There is contention that there are colonies down as far as -87degree C but they need more deep core samples.

        They are actually half expecting them to turn up in space like on Mars.

      • I did check your claim that they were only repairing but they are saying they are replicating as in the colony is getting bigger.

      • LdB,

        Microbial activity doesn’t mean that the microbes multiplicate. It only means that they survive and repairing their DNA is the last remaining activity before it finally ends.

        Further, you can’t compare bacteria in ice with bacteria in permafrost: in permafrost the bacteria have plenty of food, in ice hardly any. The only energy they can use in ice is by “burning” the traces of CH4 and NH4 and using CO2 as carbon source. The quantities involved are less than 1 ppmv, even after 140,000 years of “life”, just surviving the -40ºC, that is all.

      • Again you are just circulating a myth based on science understanding circa 2004. Work with several extreme temperature psychrophiles in 2008 lead to the realization an overlooked extreme subsurface habitat that requires more microbiological study are mile-deep ice cores.

      • I should actually single out Paul Buford Price who back in 2000 argued in a section “Evidence for Microbial Life in Deep Antarctic Glacial Ice” that there were veins in Vostok Ice that extreme temperature psychrophiles might find a suitable habitat and how one might search for them (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC15584/). He was probably one of the earliest to realize what is now accepted.

      • LDB,

        Please read more carefully what is written there:

        In the accretion ice, with an age of a few 10^4 years and a temperature a few degrees below freezing, the carbon and energy sources in the veins can maintain significant numbers of cells per cubic centimeter that are metabolizing but not multiplying. In the 4 × 10^5-year-old colder glacial ice, at least 1 cell per cm3 in acid veins can be maintained.

        The whole article is about the part of the ice core that is near the bottom of the glacier and much “warmer” than the bulk of the ice core, due to earth warmth and there is a supply of methane as energy source from lake Vostok below. Even then, the author speaks of “maintaining” life, not multiplying.

    • That was my first thought too Nick. There’s no carbon for the microbes to create CO2, unless the microbes cannibalize their own bodies, which is an unlikely trait from an evolutionary standpoint.

      Even if one accepts that stretch of logic, the amount must be small.

    • Umm, doesn’t air contain carbon, therefore, couldn’t a microbe capture that carbon and therefore modify the amount of CO2. Why is a carbon-containing substrate needed?

      • Carbon in air is CO2 (with tiny amounts of methane, and even less CFCs etc. So no scope for conversion to CO2.

        But I think the fatal flaw is that the microbes would be frozen solid. They can’t do anything.

      • “But I think the fatal flaw is that the microbes would be frozen solid. They can’t do anything.”

        ..and todays word is psychrophiles

    • My thoughts too, Nck.
      If bacteria populations are to grow, where does their structural carbon come from?
      If from CO2, then CO2 will decrease.
      If from impurities, what are they?
      But so much interpretation cannot proceed until relative abundances of reactants are known, as a minimum.
      Also some of the early chemical determinations of CO2 in air should not be dismissed lightly, as some seem to have been. There were some good early analysts. Geoff.

      • Geoff,

        The most extreme bacteria just survive at -40ºC with the only activity of repairing their DNA from (frost) damage, no population growth possible. The Calvin cycle uses CO2 as carbon source and the oxydation of NH4 to N2O as energy source. Even if all N2O was from bacterial activity, that means less than 1 ppmv CO2 used…

        Problem with the early CO2 measurements was not the skill of the analysts or the methods (+/- 10 ppmv for most of them), problem was where was measured: midst of towns, forests, over, in between and under growing crops… Only samples taken on seaships or coastal with wind from the seaside can be compared to modern measurements. These are around what the high resolution ice cores show.
        Unfortunately, there are no such measurements in the period 1940-1950, where the historical CO2 data show the highest “peak”, which doesn’t exist in any other dataset.

      • @ Ferdinand,
        But Mauna Loa measurements, next to an active volcano, are honky dory?! And whatever happened to the notion of CO2 being a “well mixed” gas??

      • AGW is not Science,

        Some days the wind blows from the fumaroles in the direction of the station at Mauna Loa, that is seen as an extra variability in the measurements. These data are not used for daily to yearly averages (but still available as raw data). The average is not influenced to more than 0.1 ppmv if you include or exclude these and other outliers. See the graph of only raw data from Barrow, Mauna Loa and the South Pole, compared to Linden/Giessen, a modern station not far from where the longest historical measurements were taken, which are the base for the 1942 “peak” in CO2 in the late Ernst Beck’s compilation of historical measurements:

      • Forgot to add:

        whatever happened to the notion of CO2 being a “well mixed” gas

        CO2 is a well mixed gas, except near huge sources and sinks, that is in the first few hundred meters over vegetation. For 95% of the atmosphere CO2 levels are within +/- 2% of full scale. I call that well mixed…

  13. Larger organisms can live in a glacier – Mesenchytraeus the “ice worm” for example. I am unaware of any observation in the Antarctica, but Arctic ice cores may be influenced.

  14. The temperatures in this snow and in the ice itself is often -30C, or -40C or even -55C. Some cores near the coasts or on mountains may have higher seasonal temperatures but in most ice-sheet cores, there is no bacterial activity going on.

    Anthrax spores etc. can survive for centuries at these temperature and be reanimated later but there is no biologic activity going on in the meantime

  15. Why this timidity?
    Just go for the jackpot.
    Atmospheric CO2 was ZERO before the industrial revolution.
    Anyone saying different is a DUNAYA!

  16. Sorry, this is not “new”, years ago (2004!) allready researched.

    The dependency of CO2 and other gases is mainly a matter of temperature and availabability of food: in coastal ice cores relative active, in the mid-Antarctic ice cores (average -40ºC), only “surviving” by DNA repairing.

    As they have no food, except some dust, they use an alternative carbon cycle, using CO2 with the production of N2O out of NH4 and O2 as energy source.
    The total produced N2O at the places of the highest increase of dust and microbes is about 0.25 ppmv. If that was all from synthesis at the cost of CO2, then the drop in the real CO2 levels at these places was 0.25 ppmv. Just negligible…

    See: http://www.pnas.org/content/101/13/4631.full.pdf
    Item K is specific for the Vostok ice core.

  17. I am not a scientist, but this study brings up a question in my mind:

    Then does this study (assuming it is valid) possibly alter the previous conclusion that temperature in the Earth’s paleohistory goes up and down BEFORE CO2 does? In other words, does this possibly change the timing issue regarding the rise and fall of CO2 and temperature?

    Thanks in advance for all replies.

    • FWIW, the late Ernst Beck was of the belief that it did. (he also believed that microbes reduce co2 levels, not raise them) He thought that the apparent 800 year lag in ice cores was in reality an artifact due to microbial activity…

  18. It’s unlikely that this would have affected the mapping of CO2 throughout the ice ages because the correlation of CO2 to temperatures is well established (with temperatures leading CO2 levels by hundreds of years). If bacteria had queered the ice cores, the CO2 levels would be seen as simply increasing with age.

    • ~graph courtesy of (the one and only) ferdinand engelbeen

      This would have to be the kill shot graph/argument for the validity of ice cores. Ice cores of varying depth and resolution all show us essentially the same CO2 levels throughout the holocene. (any critique of ice cores should be given in the context of this simple fact)…

      • Hang on Afonzarelli:

        To me this graph merely shows that nearly all the CO2 measurement methodologies are based on a similar set of assumptions, definitions and fudge factors.
        The kinky Hocky Stick trace should send out warning signals.
        To date not enough is known about the relative diffusion processes taking place in a mixed gas bubble trapped in ice over thousands of years to enable any definite conclusions to be reached.

      • cognog2,

        CO2 and other measurements in ice cores are direct measurements in enclosed air with the same equipment as direct measurements in ambient air. No assumptions, definitions and fudge (*) factors available, only calibration mixtures of exact known compostion.

        If there was any measurable diffusion of CO2 in ice, that would give enormous differences in ice of average -22ºC (coastal), -40ºC (inland) or in between.
        All these ice cores overlap each other for the same periods, despite enormous differences in accumulation rate and temperature.

        See for a good introduction of the methods used:

        (*) In stagnant air, which is the case for air in firn before the pores are fully closed, the heavier isotopes and molecules tend to increase towards the bottom. That is compensated for by measuring the 15N/14N ratio in the bubbles and applying the change in that ratio to the enrichment of CO2 in core air. As the correction is not more than 1% of the CO2 values, that doesn’t have much influence as “fudge factor”.

      • “Hang on Afonzarelli”
        He’s right on the issue of bacteria. It’s very unlikely that those different cores would be so similarly affected by bacteria. Including the hockey stick.

      • Cog, critiques are fine. i’m just saying that they should be done in the context of that one singular fact…

        …methodologies are based on a similar set of assumptions, definitions and fudge factors.

        Regardless of assumptions, defintions and fudge factors, how is it that differing cores, with differing locations and differing resolutions all give the exact same result? For all the problems that ice cores allegedly have, how is it that at the end of the day they all give us the same result? (how do those assumptions, defintions and fudge factors do that?) What i’m suggesting is that the various ice cores corroborate one another. Whatever methodologies are being used are working as evidenced by the close corroboration of the various cores. (if those methodologies were flawed it would be evidenced by a lack of corroboration)…

      • “Regardless of assumptions, defintions and fudge factors, how is it that differing cores, with differing locations and differing resolutions all give the exact same result?”

        For the same reason that water heated in a variety of pots to different temperatures will all show room temperature after several hours. They all dissipate heat until they reach equilibrium with their surroundings.

      • Some of the things corrected for
        absorption of gases on the surface of snow and ice crystals,
        separation by gravity and molecular diffusion of the gases in the ice column,
        alteration of gas composition by formation of air hydrates at great depths in the ice sheet or by presence of drilling-induced fractures or thermal cracks in ice samples,
        alteration by chemical interaction between gases and ice on long time scales.

      • “how is it that they all give the same result” It’s called curve fitting. How come Michael Mann’s bogus hockey stick matched all the ones that came after it? The same argument was used by Mann that the other model’s matching his proved his correct. Even though it has been shown that his produces a hockey stick with any randomized data. All the other “scientists” said this is what temps look should look like and adjusted away to make it so.
        I honestly have no idea what goes into ice core measurements, but I agree with the healthy skepticism..

      • Robert B

        – absorption of gases on the surface of snow and ice crystals.
        Not an issue as measurements are done under vacuum or even complete sublimation.

        – separation by gravity and molecular diffusion of the gases in the ice column.
        Based on solid physical laws and less than 1% of CO2 levels

        – alteration of gas composition by formation of air hydrates at great depths in the ice sheet or by presence of drilling-induced fractures or thermal cracks in ice samples.
        Ice samples with cracks are avoided for measurements and CO2/air hydrates decompose partly during relaxation during storage and completely under vacuum at measurement time.

        – alteration by chemical interaction between gases and ice on long time scales.
        Only an issue for Greenland ice where high deposits of sea salts react with acid dust from Icelandic volcanoes. No reactions with ice itself, only some solubility in the liquid-like ice surface, but removed under vacuum.

      • ironargonaut and Bart,

        CO2 measurements are direct measurements of air in the bubbles of the ice, not the result of data “massaging”. The only problem is the resolution, which is between 10-600 years, depending of the local accumulation rate.
        Mann’s (and other’s) work is a questionable interpretation of proxy data, nothing to do with direct measurements.

        There are no constraints in the form of maximum or minimum levels of CO2, thus the “boiling” example is of no use here.
        If ice cores with extreme differences in temperature and accumulation rate show the same CO2 levels (+/- 5 ppmv) over the same average time period, then the data are independent of these two factors (including bacteria, which don’t do much work at -40ºC, not even over 140,000 years).

      • Bart, good point… But, what does this equilibriation look like? Various pots of water cool down to the surrounding air temperature via the 2nd law. What would the mechanism be for reducing co2 levels found in higher resolution cores to the levels found in lower resolution cores? Keep in mind that at closing time you end up with a certain co2 concentration, that’s all. (a bubble at closing is like any other bubble at closing) Differing resolutions produce differing concentrations at closing. What process of equilibriation could then make them ultimately match?

        iron, yes, what i’ve presumed here is an absence of fraud…

      • “What process of equilibriation could then make them ultimately match?”

        I don’t know, Fonzie. I suppose I could speculate, but I try to avoid speculation. That’s all this is, really: speculation. We may think we know everything there is to know about the processes that occur over millennia in the ice cores, but until we can run end-to-end experiments, we do not actually know.

        In my view, the ice core records do not match the dynamics we see in modern records. I trust the modern records more than I do the ice cores, and thereby have confidence they will one day be discredited.

  19. Another climate science paper that will necessitate countless articles in the press, television news items, interviews, commentary and punditry, maybe a documentary or two, and most of all: more climate science papers.

    I’m not saying it’s entirely a racket just because it plays and plays and pays and pays. I’m sure that consideration never crosses the minds of beneficiaries for even a moment – it’s a happy accident, as it were, albeit an entirely predictable one.

  20. https://en.m.wikipedia.org/wiki/Bacteria

    Now hold on just a dolgarned minute…

    What type of micro bacteria were found? Some actually use carbon dioxide and would reduce the Carbon Dioxide in these ice cores. That elevated methyl Iodine was found suggests carbon dioxide may be used in the cellular structures of those microorganisms and other gases like methane used by other microorganisms creating a cyclic affect in the trapped bubbles in the ice, by one type feeding the other type. So knowing the quantity of these organisms trapped in what quantity of the ice has to be determined…before any affects could be determined on how it effected the Carbon Dioxide in the ice cores.

  21. For the first time scientists have directly observed living bacteria in polar ice and snow – an environment once considered sterile.

    Not quite.

    P. B. Price and R. C. Bay (2012) Marine bacteria in deep Arctic and Antarctic ice cores: a proxy for evolution in oceans over 300 million generations Biogeosciences, 9, 3799–3815. doi:10.5194/bg-9-3799-2012

    Abstract. “Using fluorescence spectrometry to map autofluorescence of chlorophyll (Chl) and tryptophan (Trp) versus depth in polar ice cores in the US National Ice Core Laboratory, we found that the Chl and Trp concentrations often showed an annual modulation of up to 25 %, with peaks at depths corresponding to local summers.

    “Using epifluorescence microscopy (EFM) and flow cytometry (FCM) triggered on red fluorescence at 670 nm to study microbes from unstained melts of the polar ice, we inferred that picocyanobacteria may have been responsible for the red fluores- cence in the cores. Micron-size bacteria in all ice melts from Arctic and Antarctic sites showed FCM patterns of scattering and of red vs. orange fluorescence (interpreted as due to Chl vs. phycoerythrin (PE)) that bore similarities to patterns of cultures of unstained picocyanobacteria Prochlorococcus and Synechococcus.

    “Concentrations in ice from all sites were low, but measurable at ∼ 1 to ∼ 10^3 cells cm^-3. Calibrations showed that FCM patterns of mineral grains and volcanic ash could be distinguished from microbes with high efficiency by triggering on scattering instead of by red fluorescence.

    “Average Chl and PE autofluorescence intensities showed no decrease per cell with time during up to 150,000 yr of storage in glacial ice. Taking into account the annual modulation of ∼ 25 % and seasonal changes of ocean temperatures and winds, we suggest that picocyanobacteria are wind-transported year-round from warmer ocean waters onto polar ice. Ice cores offer the opportunity to study evolution of marine microbes over ∼ 300 million generations by analysing their genomes vs. depth in glacial ice over the last 700,000 yr as frozen proxies for changes in their genomes in oceans.”

  22. A Google Scholar search turned up all sorts of reports of bacteria in ancient polar ice core strata.

    Here’s another good one (my bold):

    Lijun Ma Catharine M.Catranis William T.Starmer Scott O.Rogers (1999) Revival and characterization of fungi from ancient polar ice Mycologist 13(2), 70-73;
    doi: 10.1016/S0269-915X(99)80012-3

    Abstract: “Glacial ice provides a unique global source of micro-organisms that allows study of both contemporary and ancient fungal diversity. The aim of this research was to revive and/or characterize fungi and fungal DNA entrapped in ancient ice cores from Greenland.

    “Two main methods were developed to detect these micro-organisms in glacial ice: (1) Culturing to obtain viable isolates which were characterized by microscopy, PCR (polymerase chain reaction) amplification and DNA sequencing, using fungal ribosomal DNA internal transcribed spacer primers; and (2) PCR amplification and DNA sequencing directly from glacial ice.

    Hundreds of viable fungi (as well as bacteria) were isolated, and DNA sequences were obtained from Greenland ice cores up to 140,000 years old. Some sequences derived directly from ice melt resembled those of contemporary species, while others exhibited little similarity to well-characterized present-day fungi.

    “Species diversity varied among the ice cores. Some fungi remain viable for over 100,000 years in ice. This leads to the possibility that a species can appear to become extinct, but may be capable of reappearing at a later point in time.”

    • Yes there were papers from back in 1993 that also picked them up. It’s basically the same situation as when they picked up microbes down 2.8Km in the crust if the microbes have evolved to tolerate the conditions they will be thriving.

  23. Read the draft paper. There is one very big item missing, without which it is not possible to exclude mere physical/chemical reactions between known trace gasses and water phases. They sampled trace gasses at ppt (!!!) levels that naturally exist in the air thanks to biological activity. What they did NOT do was take corresponding snow samples after getting the trace gas samples, gently melt them, midrofilter the meltwater, and use a microscope to show there really were snow bacteria. And, then culturing those bacteria in 0C meltwater with nutrients to show thentrace gas productiin at higher quantities because of more benign conditions. Without bodies (and maybe some living ones), I am very skeptical. Finding bacteria is what Leeuwenhoek used the newly invented ~1590 by Jansen microscope for in the first place. Why such a simple clinching experiment was not done is beyond comprehension—unless the whole thing was suspect from the beginning.

  24. Haven’t you seen Star Trek (2009) or Star Wars? Everybody knows frozen planets are packed with huge dangerous animals. A few bacteria in an ice core is like nothing, really. These scientists get excited with the simplest things.

  25. If they’ve found one form of Bacteria in ice, how do they know there are not others that use up CO2 and leave O2? Wouldn’t sunlight be the best source of energy to a bacterium trapped in nearly-frozen water? And wouldn’t sunlight indicate that some form of photosynthesis would be more likely than respiration?

    • Yogi Bear,

      Not much to hide:
      If they use the grating technique, ice is crushed under vacuum and water vapor removed in a cold trap (-70ºC).
      Another technique sublimates everything under vacuum (99% recovery for CO2) and freezes everything, including N2 and O2, cryogenic. Then each part is evaporated separately and measured via a mass spectrometer, including the isotopic composition.
      Both methods show the same CO2 levels.

      See slide 6 of 8 at:

      • A systemic error is likely built into the analysis, one way or the other. In the early days of ice core analysis, results greater than 1000 ppm were regularly obtained. These were eventually thrown out and high results were blamed on contamination, poor sampling, etc. Eventually, they developed methods that consistently gave the results that they expected.

      • “Both methods show the same CO2 levels”

        Dry extraction; 99% extraction efficiency, but it’s all mixed together.

      • That got chopped short by wordpress, second try…

        “Both methods show the same CO2 levels”

        Dry extraction; 99% extraction efficiency, but it’s all mixed together.

      • R. Shearer,

        The early methods of extraction needed huge parts of the ice core and contamination with drilling fluid were quite often. That was what Neftel said in his description of the Siple ice core: where huge values were found drilling fluid had contaminated the sample. Measurements of several parts of the same sampling depths did show huge differences in CO2 levels, including “normal” values in line with other depths. A new core was drilled many years later and no contamination or high values were encountered.

      • The dry extraction is only measuring the air bubble content while the sublimation is measuring the total gas content, so how can they be the same? And the dry extraction doesn’t show how much CO2 the ice has absorbed, plus the error margin of +/- 38% for CO2 is huge.

      • Yogi BEar,

        Both methods extract air, not only CO2. The grating technique recovers some 80% of all air, the sublimation technique over 99%. Both give the CO2 content as a ratio between CO2 and air, which is essentially the same, no matter if you measure in 1 air bubble or 80% of all air bubbles ot 99% of all air bubbles…

  26. A huge variety of microbes occur in water lakes under the ice sheets in Antarctica (see link below). Most of these live in anoxic, very very saline water. In many cases this water is warmer than expected because of microbial respiration and these biological communities have survived and evolved isolated from the atmosphere for hundreds of thousands of years. Clearly there is also ‘life’ within the ice (or has been) and of course these biota will have changed the gas contents and ratios within the ice. NASA has been onto this for many years as the fauna offer insights into extraterestrial life forms. https://www.nature.com/news/lakes-under-the-ice-antarctica-s-secret-garden-1.15729

    [saline water ?? .mod]

    • Yes mod, very saline water is typical of Antarctic under-ice lakes. They are often also highly sulfidic. This is mostly due to the respiration chemistries of microbes living in an oxygen-free environment in layered/stratified communities within the water. Some under-ice lake waters may approach the Archean – paleoproterozoic sea chemistry when microbes were the only living form and survived without oxygen or very low oxygen levels.

      [Thank you for the clarification. .mod]

  27. Wouldn’t be surprised if this study will (or was designed to) be used to get rid of the pesky temp-CO2 lag in ice cores

  28. Photosynthesizing cyanobacteria are widespread in Antarctica, including on the ice near the Pole (Carpenter, E.J., Lin, S. and Capone, D.G., Bacterial activity in South Pole snow. Appl. Environ. Microbiol. 2000, 66: 4514-4517).

    If they occur below the surface (though still in the photic zone) they might well lower the amount of CO2 by photosynthesis.

  29. “Uh, oh”

    Why politicise it? Agenda?

    [“uh oh”, as in it means they’ll have to check the science based on ice cores to be sure they aren’t affected. No politics in that, except in your mind. -Anthony]

    • The header would have meant that without adding “uh oh”. In the context of your creative headlines uh oh sounds a lot like a dog whistle.

      [and your comments sound a lot like the kool-aid talking -Anthony]

  30. “No pollen in Antarctica. Not much carbon-containing dust either.”
    – Nick Stokes, December 20, 2017 at 11:45 am


    Absolutely written in stone. No, wait. Scratch that.

    Slartibartfast is a Magrathean, and a designer of planets. His favourite part of the job was creating coastlines, the most notable of which were the fjords found on the coast of Norway on planet Earth, for which he won an award. While Arthur Dent and Ford Prefect were on ancient Earth, they saw Slartibartfast’s signature deep inside a glacier in ancient Norway.


    Yes. Written deep inside a glacier.

  31. Hm. This seems to confirm one of my thoughts about wholly rhinos, mammoth and other megafauna extinction.

    What we have in the ice core is at moments close to CO2 starvation for C3 plants (180 ppm and lower)
    There are also studies that showed some trees suffered from CO2 starvation during the ice age.
    This would mean C3 plants especially the short lived like forbs may have suffered.
    -> C3 plants + CO2 starvation – these plants become rare and cannot sustain the population feeding on them
    We also know the megafauna existent until about 40k-20k years ago was eating mostly forbs (C3 plants)
    no plants => megafauna starvation
    Probably extinction of the megafauna that is so much blamed of humans may have been a result in changes in the environment cause by CO2 starvation

    What is interesting is that even if CO2 concentration started to rise at the beginning of the Holocene it seems it was too late for the C3 plants, the plains in Siberia which were covered with forbs are now covered with other plants (there was a study confirming this too). This may have been the case in other areas too.

    Of course humans may have played a role too, but a population that was already pushed at the survival limits could more easily go extinct.
    Certainly this idea will not get too much traction with the ‘CO2 is evil’ crowd…

  32. Just a minute though – They’re measuring the CO2 level of air that’s been in contact with ice? Does any of the ice melt whilst extracting the air? CO2 is highly soluble in water, so any measurement of levels in the trapped air would be totally meaningless, would they not?

    This could be relevant:

    When you think a bit about this, it’s questionable if anything can be deduced from CO2 levels in trapped air. The levels in the original air could have been higher or lower.

    • Ian,

      Sea ice and snow glacier ice have little in common: there are very small quantities of carbonates in glacier ice, so these are no carbon sinks at all.
      The measurements of CO2 in ice are performed under vacuum. Any liquid water, including any CO2 dissolved in it is removed and water is trapped in a cold trap at -70ºC, where very little liquid-like water is left at the surface.

  33. So the discrepancy is in Parts per Trillion to the Parts per Millions of CO2 and this is somehow significant? When the charts show past CO2 and Global Temperature never matched over the history of Earth. The CO2 was not considered to be a factor of Climate Change. And the biological timelines with massive increase and decrease of flora and fauna show us that CO2 levels played a significant roll in their existence and demise as climates changed from warm to cold. No matter what the extent of CO2 in Ice Cores have shown over history. Live and dead and fossils of flora and fauna thereof show us what was and is significant.

  34. I’m not lichen the idea of tiny critters invading our treaty-protected polar environments and corrupting the ice cores, places where life has no right to be.

Comments are closed.