An update on Comet 67P/C-G’s water-ice cycle

ESA VIRTIS infoLast week we reported on the daily water-ice cycle of Comet 67P/Churyumov-Gerasimenko that was observed in a specific region of the comet’s neck by Rosetta’s VIRTIS instrument, in September 2014.

Yesterday at the European Planetary Science Congress, animations were shown covering the period 1 August 2014 to 10 February 2015 for two different areas of the nucleus, showing the behaviour of the water-ice over a much longer period.

“We are now able to show that this cycle is common in several regions of the nucleus, depending on the illumination conditions, and hence further demonstrate that the proposed cycle is a general mechanism of water transport from depth to the surface acting on comets,” said Fabrizio Capaccioni, VIRTIS principal investigator. Fabrizio described the results in a press briefing held at EPSC yesterday afternoon.

The two movies are based on data acquired between 1 August 2014, when the comet was at a distance of about 542 million km (3.62 AU) from the Sun and 10 February 2015, when the comet was 352 million km (2.35 AU) from the Sun. They show the ice abundance at each time during the comet’s 12.4 hour day for two different regions: one focused on the Imhotep region on the comet’s large lobe, the other around the comet’s neck. For each timestamp, each data point represents the average, for a particular hour of the comet day, over the entire timeframe August-February.

The blue regions indicate the presence of ice in the uppermost surface layer, red shows no ice, and grey indicates portions of the comet that were in shadow.

“We see that in many areas on the comet’s surface the bluer regions have a short lifetime, appearing just after dawn and disappearing a few hours later,” says Fabrizio. “This indicates that in these areas the ice is not stable at the surface but is deposited during the night and sublimates away during the day. This is a clear confirmation of the proposed water-ice cycle.”

The movies also show areas where the bluish colour is persistent throughout the day, which Fabrizio explains as regions of permanent ice – these will be the subject of a future study.

Presentations on the comet’s water-ice cycle will also be given as part of the Rosetta scientific sessions at EPSC later this week, for example: Temporal variability of 67P/Churyumov-Gerasimenko nucleus spectral properties from VIRTIS-M onboard Rosetta by M. Ciarniello et al.

Browse more conference abstracts here.




  • originalJohnn says:

    Still a speculative interpretation to assert that the blue coloured regions depict ice. Water yes, from the spectrum, but without precise temperature measurement confirming that these blue regions are below 0 deg C no evidence that they are ice. It is a presumption based on the expectation of ice.

    • ianw16 says:

      Eh? So it’s water, but not necessarily ice? That leaves two options: vapour, which we wouldn’t see, or liquid, which cannot exist at these pressures.
      It’s ice.

    • sjastro says:

      It was explained to you in the previous Water-Ice Cycle thread, the IR spectra of ice, liquid and vapour are distinctly different.
      No speculative interpretation is required.

      Since the power of words hasn’t sunk in, perhaps a picture of the IR spectrum of water will.

      • originalJohn says:

        Sajstro, I posted my reply on this thread, rather too hastily, asserting the speculative nature of the ice interpretation of the infrared spectrum before i had read the comments on the previous thread you refer to entitled “rosetta reveals comet’s water ice cycle” where I posted a similar assertion. A number of people commented then in reply that the spectrum recorded exhibited an absorption peak that was unequivocally characteristic of ice. Obviously I accept this and I had glossed over it in my original reading of the description of the published paper., biased by the fact that with other comets water has been speculatively inferred as indicative of the presence of ice.

        i have no problem with the occurrence of ice on a comet nucleus surface and there are several mechanisms through which it could form. I do however dispute the fact that ice on the surface of the nucleus in any way supports the hypothesis that the nucleus consists mostly of ice or that sublimation of that postulated majority ice is the comet water source, anymore than rock or hydrocarbons on the surface would be acceptable as categorical evidence that the nucleus was made of rock or hydrocarbons, without further corroborating evidence.

        • sjastro says:

          Not only is IR spectroscopy effective in detecting and differentiating between ice liquid and vapour, the same applies in the microwave range.

          In fact the microwave spectrum of liquid water and ice is basically featureless, in the vapour phase the lines are sharp enough to allow Doppler measurements from which the speed and direction of the gas can be determined.
          Given that the vapour state of the spectrum seems to “pop out of nowhere” at the surface and travel outwards to Rosetta’s detector, its hard to see the mechanism other than sublimation.

          If on the other hand as you suggest the water is formed which then reaches the surface in the form of ice there is already a problem.

          If ice is not a substantial component of the comet and is deposited on the surface then how does it exist long enough to be covered in a layer of dust, hydrocarbons, etc in order to exist below the surface?

          Given the water ice cycle indicates that exposed ice is short lived this is a highly unlikely scenario.
          It again demonstrates the mainstream model is far more consistent then the electric comet.

  • ianw16 says:

    Very interesting videos, Emily. Would be very interesting to see the contemporaneous temperature maps of those regions. From what I’ve read/ heard, they are coming, and are unlikely to show anything different to what we saw at 103P Hartley 2. As in, high outgassing areas being no different/ slightly lower in temperature than the surroundings.
    Going to be very difficult to explain from some perspectives.

    • THOMAS says:


      It seems like you haven’t been keeping abreast of all the peer-reviewed stuff coming out on Rosetta, otherwise you’d have picked up on this highly interesting tit-bit from the abstract of the impending article “In-situ investigations of the ionosphere of comet 67P” presented at the recent European Planetary Science Congress 2015 in Nantes, France ( : ““In contrast to the often modelled scenario for a very active comet, the Langmuir probe instrument (RPC-LAP) finds electron temperatures mainly in the range of tens of thousand kelvin around this less active comet. This can be attributed to the lower density of neutral gas, meaning little cooling of recently produced electrons.”

      In other words, the plasma/neutral gas ratio is high and the plasma is hot. This is an explicit expectation of the EU model and something of an impossibility for the standard ‘dirty snowball’ model. Please note that these are actual FINDINGS, not forecasts, and they need to be addressed. Strange that they haven’t already made the headlines. Or is it?

      • sjastro says:

        Well Thomas by your logic the Earth which also possesses an ionosphere must be an electric comet.
        When a gas is subjected to ionizing photons there is a battle between ionization and recombination (where positive ions and electrons recombine to from neutral atoms). Whether one process dominates over the other depends on the density of the gas.
        In the ionosphere the density of the gas is extremely low resulting in ionization becoming the dominant process. This is due to the large time scales in which ions and electrons can exist due to the long mean free path lengths of the charged particles.
        Since ionization results in electrons being ejected at high velocities, the electron temperature is higher than the neutral gas and the positive ions.
        Ionospheres are therefore “hot” because ionization is the dominant process.
        If you actually bothered to read the abstract you would find the authors are making the same statements albeit in fewer words.
        The ionosphere around the comet is due to the expansion of sublimation gases into the surrounding space resulting in a density where ionization becomes the dominant process.”

        • logan says:

          Hi Sjastro. Is the temperature of the weakest detached H20 electron high enough as for latter breaking the molecule?

          As related in:

          • sjastro says:

            The electron temperatures in the coma are lower due to reaching thermal equilibrium with the bulk of molecules and atoms.
            Despite this they still carry sufficient kinetic energy to cause disassociation of H2O and CO2.
            This is due to the two lone pairs of non bonding electrons in a water molecule. Being non bonded, the ionization energy to remove a single non bonded electron is quite low..
            As a result these non bonded electrons can carry a high initial kinetic energy which despite losing energy through thermal equilibrium can still cause disassociation.

          • logan says:

            Thanks Sjastro 🙂

      • ianw16 says:

        Wrong. As replied to elsewhere. The ELECTRON temperature is quite high, but that is expected. Does the EU ‘model’ predict the diamagnetic cavity seen at Halley? What were the ion and electron temperatures in that region? Why were the ions and neutrals within the cavity all of cometary origin?

        As for the nonsense about the pre-impact flash lower down the page, it is precisely that; Nonsense. What sort of flash was it? Electrical? Why didn’t any of the X-ray telescopes watching the impact see anything?
        And W.T. was not the first to predict this. Pete Schultz give numerous outcomes from modelling the impact , and this included a relatively faint ‘first light’ on impact.
        I suspect that W.T. saw this prior to the impact, and then used it to fool a few people into believing his nonsense.
        Seems it worked. None of the outcomes of that impact were in any way consistent with it hitting rock. Including the ejection of water ice grains.

  • A. Cooper says:

    This is very interesting. Practically no ice sublimating on the flat Imhotep plain but abundant ice sublimating from all around the edges. Despite this, it was the Imhotep plain that sank by 5 metres so suddenly and unexpectedly while the edges remained stable. 

    This serves as evidence that the Imhotep plain is sublimating at the level of the first fracture plane down, the gases are migrating to the fractured edges and escaping from there. This supports Marco’s suggestion of deeper sublimation in a comment on the ‘Comet Surface Changes Before Rosetta’s Eyes’ blog post (18th September 2015). He said:

    “Perhaps the outgassing is not from the surface, but from somewhat below the surface. Once the subsurface is depleted of material, the surface slumps down from the point of least resistance then pulls down adjoining sections in a chain reaction growing into ever bigger circles.”

     I added that I believed his hypothesised deeper gases were forming at the level of the first fracture plane down and emerging from it at its exposed eastern edge. This is the lowest part of Imhotep, a somewhat deeper gouge where the so-called ’roundish features’ were found and described in the ‘Inside Imhotep’ Rosetta blog post (20th July 2015, describing an earlier Imhotep scientific paper).

    You can see from the Imhotep video that the largest area of ice is this sunken area at the exposed end of the first fracture plane. In the video, this area is just above the large circle near the bottom perimeter. It’s also arguably the longest-lived patch apart from some very small pockets perhaps. 

    The fracture plane runs under the flat area (the Imhotep plain). We also discussed gases and possible slurry emerging from this same lowest point in the ‘Inside Imhotep’ comments. I suggested they resulted from the escape of the hypothesised Imhotep slab. This area of heightened activity is where the slab would have hinged up on escaping and where the most damage and outgassing would be expected. I think we are seeing the low-level continuation of this process (not including slurry) as a slower seepage from under the flat area, along the first fracture plane down. That would presumbaly cause a very even erosion along the fracture plane between the first and second strata. If it then became an increasingly porous honeycomb of volatile-depleted crust, it would have to sink at some point. That point would almost certainly be on approach to perihelion. An even, porous honeycomb running along the first fracture plane would account for the very even, large-scale collapse of the Imhotep plain. 

  • dave says:

    @ A Cooper

    Interesting supposition, but are you postulating this is the cause of the so called gentle release of gas?
    Does this mean that the fierce jets we see are from a different mechanism, or does this kind of pit digging somehow end up with an explosive release?
    Your article mentions the slow deepening of the pit, does it finally explode, or just keep digging.


    • A. Cooper says:


      That last comment was referring to gentle, ongoing sublimation only. I was going to add that the two big, explosive events (March with a Rosetta blog post in April and also just after perihelion) appear to be part of the same process. I thought I’d leave that for another day but since you ask, yes, I would have thought that as 67P approaches perihelion, new conduits open up near to the exposed end of the fracture plane. That’s because the sublimation in the fracture plane would increase and pressures would rise along conduits that had been sealed off after the last perihelion. According to Marco’s thinking, if the pressures are sufficient for liquids these will explode in a BLEVE, like a boiler explosion, when the pressure drops to vacuum level. Maybe it’s just gas pressure build-up alone, then sudden release, but the latent energy in a BLEVE would give a spectacular explosion when the seal was breached. Both those two explosions appeared to come from the perimeter of the Imhotep plain. The March one looks as if it came from the lowest point on Imhotep, which I suggested, above, was the outlet from under the fracture plane and showing the most surface ice in the video. The second appears to have come from the opposite end of the flat area, what I call the plain (as opposed to the plane under it). The March one was the first recorded outburst from the unlit night side suggesting a subsurface heat source that was independent of direct insolation. At that time, scientists were trying very hard to suggest it might be a cliff catching the dawn rays. Mattias Malmer said it would be impossible because that part of the comet is flat with no cliffs and was an hour-and-a-half from sunrise.

       I could see from my knowledge of the terrain and rotation times that this was the case and that the lowest sun rays were at least several hundred metres above the surface where the outburst occurred. That location had already been in darkness for 4.5 hours and still had a large energy store to cause the outburst. That store simply couldn’t be sitting on the surface as heated crust so it had to come from the subsurface, implying a deeper heat transport process. 

      If the outburst was from the sunken area as it appears to have been, then it’s strong evidence for heat storage and consequent activity under the fracture plane, day or night towards perihelion. Several NAVCAM photos since the March outburst have shown night jets from Imhotep.

      One reason it think it’s sublimating and exiting from under the fracture plane is that the first stratum could be loose, so the vacuum is extending more readily under it in places and therefore located next to reserves of ice that are slowly warming up to the correct temperature to sublimate into the vacuum. One reason I think it could be loose is that it would have been in negative g when my hypothesised Imhotep slab departed at escape velocity. All strata down to 500 metres or more would have been experiencing negative g. The topmost one couldn’t hold on and departed. Whichever stratum was next in line was, in all likelihood, on the cusp of leaving if similarly structured. So it could have been loosened.

      As I said, this all assumes a heat transport process that goes deeper into the comet than the near-surface. The recent paper on sink holes implied this must be the case but the Rosetta blog post summarising that paper didn’t cite a mechanism as far as I can remember. The massive 5-metre subsidence of the Imhotep plain also suggests this must be happening. This is because the so-called erosion described in the other recent paper on that subject was also cited as being 30 to 100 times too fast for current theory to explain. Therefore the ‘erosion’ must have been depletion of the layers below 5 metres and possibly well below so as to give that even-looking drop to a lower level. That’s erosion too, but of a different sort because it’s invisible, subsurface depletion. Whenever erosion is mentioned in these papers in relation to anything but a sink hole, I think they mean scouring from the surface. I think that’s certainly the case for the Imhotep plain paper. 

  • ianw16 says:

    @A. Cooper
    “Perhaps the outgassing is not from the surface, but from somewhat below the surface……”

    That has been the view for a long time. See the Tempel 1 data. See also the recent report of ice in the shadowed areas of 67P.
    At Tempel 1 the impactor excavated H2O (indisputably). It was obviously sub-surface, and the overall outgassing rate could not have been supplied by the observed areas of surface ice. Hence, it must be from sub-surface reservoirs.
    We have known this for a long time. It has been theorised for even longer.
    Deep Impact crashed a projectile into a comet. What came out was H2O (indisputably). Ergo, there is ice beneath the surface. This was 10 years ago. Not sure why people are even suggesting this as a “possibility”. It is fact.

    • Marco says:

      Hi Ianw16,
      The quote you mention was from me regarding outgassing from the subsurface on the Imhotep post and I think you have it out of context. The whole “Imhotep Erosion” paper was explaining the fast “erosion” with fast surface erosion due to sublimation from very near the surface where it was getting the most thermal input from the sun. I was suggesting it was *preferentially* from the subsurface (minimum 5 metres below the surface and below that) despite the layer above it blocking direct radiative heating and presumed to be colder than the surface. I was not suggesting that subsurface outgassing was not mainstream, but that deep subsurface H20 preferential outgassing is not mainstream.

      • ianw16 says:

        Ahh, okay. Sorry, you’re right, I just didn’t have the context. Mind you, given the thermal inertia assumed at this comet and others, 5m might be pushing it. From the Tempel 1 data: “Thus, even in polar regions that receive sunlight over extended periods, to a first order thermal processing on Tempel 1 can only occur within the first few meters of the surface, while deeper layers are thermally isolated from the surface.”
        Last part of section 4 on p. 289.

        • Marco says:

          Hi Ianw16,
          The keyword here is “assumed”. The thermal conductivity of the crust has not been measured. The Imhotep “erosion” is better explained with a deeper H20 evaporation layer.

    • Sovereign Slave says:

      Hi Ian, can you link to a few of the scientific papers that indisputably prove ice ejected from the Impactor plume? I’ve read some of the papers, but as with articles like this,
      which skips around the issue, none of the scientific papers I’ve read seem to absolutely prove this either.

      • ianw16 says:

        “The Deep Impact flyby spacecraft includes a 1.05 to 4.8 μm infrared (IR) spectrometer. Although ice was not observed on the surface in the impact region, strong absorptions near 3 μm due to water ice are detected in IR measurements of the ejecta from the impact event. Absorptions from water ice occur throughout the IR dataset beginning three seconds after impact through the end of observations, ~45 min after impact…………”

        From the abstract.

        I’ve yet to see any corrections or refutations of this paper in the scientific literature, so one can assume that that when they say they detected H2O, CO2 and HCN, then we can be sure that that is indeed what they detected.
        As Sjastro explains above and elsewhere, they are unlikely to get this wrong.

        • Sovereign Slave says:

          Perhaps I’m reading this wrong, but for the spectrometer for Deep Impact, readings above 2 might not be that accurate:


          2nd page: “High quality PMT detectors may be able to yield accurate measurements at absorption values of 5 or higher. Many CCD or photodiode instruments, and lower quality PMT instruments, however, tend to become unreliable at absorption values above about 2 because too little light is reaching the detector to allow accurate measurements.”

          According to this link,

          from what I can tell at least, the equipment for Deep Impact was clearly CCD. Let me know if this is inaccurate, but if this is accurate, saying there was water ice in the ejecta can’t be said with much reliability.

          • ianw16 says:

            No, I think you may be getting confused with the optical instruments, working in visible light. The light received by the instrument was split by a beamsplitter; “The beamsplitter reflects light in the visible band to the HRI CCD, and transmits light in the IR band to the slit of the IR spectrometer. The requirements on the beamsplitter were quite stringent with a precise 1.05 micron crossover…….”
            Therefore the CCD only affects the optical images, and isn’t part of the IR measurements.
            It really doesn’t matter how the EU crowd have tried to dress this data up over the last 8 years or so, the fact remains that they detected H2O emissions from vapour, and absorptions from ice ejecta.
            Quite why one of the leading lights (WT) of that particular pseudoscience would want to obfuscate, and brush these results under the carpet, is a matter of conjecture. I have my own views on that!

          • ianw16 says:

            Further to my previous comment, I had e-mailed the lead scientist from the paper I referenced, Prof. Jessica Sunshine. She replied. The opening words of her reply were: “Ian we did indeed detect water ice. It is unambiguous.”
            So, who to believe? I know who I believe.

          • Sovereign Slave says:

            Perhaps, but the paper says that there were three instruments: the first is the HRI (which includes the spectrometer), the second is the MRI, and the third is a simple unfiltered CCD camera with the same telescope as MRI. It then clearly states that “All three instruments use a Fairchild split-frame-transfer CCD with 1,024×1,024 active pixels.” This does not seem to support your assertion. Also, when there are articles like the one linked below where, based on up close measurements and data collected, mission scientists were saying that there was little water and mainly dust in the plume, one has to wonder about assertions to the contrary. And although that article was fairly early on, it seems odd if other earthbound data is said to so radically contradict the initial up close findings.


          • THOMAS says:


            “Quite why one of the leading lights (WT) of that particular pseudoscience would want to obfuscate, and brush these results under the carpet, is a matter of conjecture. I have my own views on that!”

            In terms of who is brushing what under the carpet regarding the Deep Impact mission to Tempel 1, I’m afraid you’ve got things the wrong way round. You seem to have forgotten that the most extraordinary observation regarding the actual impact was the small initial flash which occurred a second or so before the main flash, to the utter amazement of mission scientists. There was a lot of talk about the “double flash” at the time from the top NASA people, as in Peter Schultz’s “First impressions” ( or in another NASA article They never attempted to deny the reality of the small initial flash, they simply tried to pass it off as the initial contact with the surface of Tempel 1, with the main flash a little later “presumably originating deeper within the comet”…! (Their use of the word “presumably” no doubt signifies the degree to which they themselves appreciated the absurdity of this explanation, given the 10km/sec speed of the copper impactor as it hit the comet!)

            Now the fact of the matter is that Wal Thornhill had explicitly predicted the possibility of an initial flash due to an electric discharge in a blog-post which was posted A FEW HOURS BEFORE the impact… It is no doubt one of the most extraordinary examples of a non-standard scientific prediction being very precisely and indisputably borne out by actual events. Now strangely, even though (or perhaps because) this event was so baffling, no further investigation of it was ever carried out as far as I am aware, no peer-reviewed paper was published on it… In short, it was quietly “brushed under the carpet”; to the extent that standard theory proponents such as yourself, ianw16, now even feel entitled to rewrite history by casting doubt on whether it even occurred… It did, as is proved by the “breaking news” NASA articles I linked to above and it can only be explained as an electric discharge phenomenon.

          • sjastro says:


            The irony here that while you are championing the genius of WT, you seem to have wilfully ignored or are of unaware one simple fact, WT predicted a PRE IMPACT flash.
            The links you provided show that an impact and post impact flash occurred.
            So rather than being “one of the most extraordinary examples of a non-standard scientific prediction being very precisely and indisputably borne out by actual events.” it simply reaffirms yet again the electric comet is a nonsensical theory.

            Here is something else to consider, where is the WT’s pre impact flash in the video of Philae’s descent.

            I’m sure this is ripe for a conspiracy theory that such a flash was edited out of the video as scientists are desperate to hide the truth to preserve their jobs.

      • logan says:


        No aditional H20 contained in the plume gases. Just background.

  • logan says:

    Videos gives image of super-volatiles mostly gone. Just thick water ices cocoon remaining.

    Videos also go against idea of any significant accretion post binary “contact”.

  • Robin Sherman says:

    Just a few ideas to conjure with, which may or may not be valid.

    Thermal transport seems to be the key here. Conduction is not the only heat transfer mechanism. Ices when sublimating at, or just below, the surface, generate gas travelling in all directions, gases travel into the comet as well as the vacuum of space. The latent heat released when the gases freeze and the kinetic energy they carry, is therefore going to be significant in the transfer of heat into lower layers of the comet, especially given the high porosity of the cometary material. Water one would expect to be a significant vector in this process given its thermal properties.

    The suggested accretion scenario for comet formation, resulting in “onion layers”, would suggest layer boundaries which could facilitate the seepage and migration of gases. Small voids between planetesimals, could become “seeds” for growing pockets of gas. These expand slowly as the heat trickles down from above to create pressurised gas bubbles in the subsurface layers.

    Such internal “erosion” might be thought of as analogous to the erosion of Limestone caves on Earth. Seams and pockets of more volatile ices, are “eroded” away beneath the surface creating the possibility of subsurface collapses, the Seth region sinkholes for example.

    In the very low gravity of 67P, collapses would be slow and its possible even very large areas could maintain their structural integrity and act like bellows, forcing the gas rapidly out through it’s route to the surface as a highly active, short lived outburst. Compression of the gas would also heat it and force the gases into the surrounding ice, sublimating more ice, spreading the erosion outwards around the edge of the initial collapse, such as at Imhotep. It would also enlarge and erode the route to, and exit from the surface. That outflow point could be some distance away from the collapsed area.

    A linear exit crack at the surface, which may not be perpendicular to the surface, might expand to create caves, vertical crevices as well as the ubiquitous “horseshoes”. Collapsed tunnels, or eroded vertical cracks later sealed and covered in dust, maybe the cause of “rippled” surfaces. Indeed some early images had formations that Logan called “ducting”, that could be evidence of subsurface “tunnels” now exposed at the surface and near Agilkia there look to be exposed “tunnel” openings.

    There may be occasions where a subsurface collapse compresses some gases sufficiently to liquify them before a route for the gases to escape to the surface is forced. This fluid mixed with dust and organics could form a “sludge” that then flows through ice “lava tubes” out onto the surface to rapidly freeze. The “toadstool” features for example or the apparent flows emerging from “cave” openings on cliffs and slopes. In other circumstances, the compressed and heated gas may be sufficiently confined for the pressure build up to blow chunks of cometary material off the surface, either temporarily or permanently, as we have seen.

    It seems plausible to me, especially given the low overall density and high porosity of the comet, that below the consolidated surface, there is a network of cracks, faults, tunnels and gas filled voids on multiple scales throughout the subsurface structure allowing the transport of volatile gases and heat below the consolidated surface of the comet. This network is constantly reconfigured as gases freeze, sublimation forms new channels and pockets, while other channels and voids collapse. This constant recycling of subsurface layers could be a mechanism for mixing all the components of the comet into the homogeneous consolidated material seen at the surface. Think earthworms etc. mixing garden soil as a very loose analogy.

    I would conjecture that, the higher the rate of sublimation near the surface, the greater the extent of the gas and heat transport network below. Areas where ice constantly condenses and sublimates are hence the most active. Dust covered areas one might think, build up a “permafrost” in the dust layer, thus enabling higher rates of sublimation than exposed consolidated, sintered ice. This could explain why these areas appear more active, though the higher amounts of dust available to be carried by escaping gas does make any activity there more visible to start with.

    The neck region experiences both mechanical and thermal stress to the greatest degree and therefore has the highest likelihood of developing cracks and fissures in the cometary material. It also seems to have a greater percentage of Water, improving the transfer of heat to subsurface layers and significant amounts of dust too. The thinner neck region also has less thermal inertia, “cold mass”, to refreeze gases or keep gases frozen, hence the greatest and earliest activity perhaps.

    • Marco says:

      Hi Robin,

      Thermal transport is a very important concept for any theory, such as stretch theory that requires a (relatively) warm, soft interior with a hard shell. The problem with convection is that it is pressure driven. The vacuum of space ensures that convection is necessarily outwards, thus only able to transport heat outwards unless the whole interior of the comet is under pressure (not a great deal of pressure, but it needs to be sealed bar “pressure valves” ). if you imagine a pressure cooker, outgassing can still happen through the valve, but any cracks (or even uncontrolled porosity) would break down any possible internal thermal transport as it would rush to the exits and blow them open, taking out any heat with it. Thus caves, seams, cracks and porosity are completely counterproductive to convective transport unless they are within the seal rather than on the surface.

      Conduction is the only realistic way for enough thermal energy to travel inwards effectively, certainly to below the crust. From there, as long as the internal pressure holds convection will be most effective. If there is a lot of “onion layers” all effectively holding pressure, a surface breach, crack or tunnel will only depressurise the outermost layer.

      If there is internal pressure comet-wide, all bets are off regarding sublimation-driven models.

      • logan says:

        ‘The latent heat released when the gases freeze and the kinetic energy they carry, is therefore going to be significant in the transfer of heat into lower layers of the comet, especially given the high porosity of the cometary material’.

        Fully Agree Robin. Some water vapor being ‘aspired’ at early ‘morning’, sinking heat, an latter freezing over colder, deeper layers.

        Could we bet, down to around 50m?

      • logan says:

        Hi Marco.
        ‘The problem with convection is that it is pressure driven’.

        Could be wrong, but it’s not about convection, but diffusion. When heat starts diffusion resistivity is 0, in all directions.

        Gas sinks, until pressure from lower layers balance with resistivity of upper layers.

        • Harvey says:

          Logan see my post below. The word ‘convection’ is used in more than one way.

          • Marco says:

            Hi Harvey,

            Thanks for weighing in on this. From what I gather, a great deal of research and simulation has been done on this, and for presumed materials of the comet (porous ice/dust mixture) and a great many other plausible surfaces, solar heat is not likely to go to any great depth via diffusion. Like I said you could probably engineer just the right conditions for a thermal transport system, but having them exist just by chance is a long shot. That being said, we can look for evidence of internal heat, and the outbursts are a vague evidence of that, so we could hypothesise enough heat transport deep enough for those.

      • Marco says:

        Hi Logan, Robin,

        I have looked at this in great detail, and I wouldn’t mind experts weighing in on this, but as far as I know the processes you are talking about would not have effective inward transfer of heat at all. The main issue is that on Earth we have hydrostatic equilibrium at a high pressure, and the equivalent on 67P is virtually nil, so gases do not diffuse towards the inside of the comet- they bounce back and out even if initially going inwards. You certainly could engineer a thermal transport to any depth, but it is hard to see how it would happen naturally with the materials detected.

      • Harvey says:

        There is the potential for quite a bit of terminalogical confusion here – and some rather counter intuitive physics.
        Firstly, in commonplace terms, ‘convection’ is driven by temperature induced density differences. Whether it occurs depends on critical values of things like the Rayleigh number and Grassof number etc, all of of which have ‘g’ in the numerator, so they are very small on 67P and will not exceed their critical values under any realistic conditions. So in that sense ‘convection’ is improbable on 67P
        But we also speak of ‘convective heat transport’, the movement of heat carried by moving gas or liquid, even when the flow is driven by pressure differences (say a fan) and in that sense convective heat transport could occur.
        you need to think very carefully about the nature of the gas flow. Our intuitive image of this is all in viscous flow, ‘high’ pressures, where the molecular mean free path is very short compared to the size of the passage/aperture the fluid is in.
        But in molecular flow, it’s totally different. There are no collisions, and the gas diffuses. The net flow is right if more move right than left; but large numbers can be going *in the opposite direction to the net flow*. If they then encounter a cold surface they could condense on it – and deliver their latent heat to it. Situations like this are commonplace in high vacuum systems, where things easily move *against the pressure gradient* by diffusion, it’s called backstreaming.
        Now I’m not at all sure such mechanisms are significant on 67P; they may not be. But I am sure that it needs rather careful modelling and thought, and depends on pressures, pore sizes, temperatures and materials. It cannot safely be concluded from a simple verbal argument.

        • logan says:

          Thanks Harvey. Understood. Obviously we have a complex situation here and -as you say- could even be experiencing backstream on-sun-treshold. Very fine dust at North neck’s [Maybe 50m is an exaggeration on my part].

          Document suggest that ices survive below surface, and when sun shine restart there is no ‘pressure gradient’ below them, just emptiness.

          Heat wave has to slowly travel down before a ‘pressure gradient’ is constructed. And most efficient way -given the presumed very low thermal conductivity of this material- is as a sublimation/deposition cycle [an unpressurized variation of heat pipes].

          This is, we are on pure ‘phonon’ transfer until ices are reached at document’s investigative area, from there down we are on both mechanisms [until the pressure gradient ends stabilizing].

          Just transient phenomena, but relevant as plausible vector for deep heat transfer; an issue been chewed for a long time around the blog.

          And of course, just on speculative game, Scientist’s modeling [an probably some good look on data collecting] will tell.

        • logan says:

          Hi Harvey. “but large numbers can be going *in the opposite direction to the net flow*”.

          Eventually -along many day/night cycles -of sublimation/deposition- arriving at deep surfaces of porous 67P.

          As you say, no idea about the scale of this process. Phonon heat transfer now looks a little relevant.

  • logan says:

    ‘This constant recycling of subsurface layers could be a mechanism for mixing all the components of the comet into the homogeneous consolidated material seen at the surface’.

    Think this a new line of argumentation around the blog.

    Just a little foot-note about this mechanism being more plausible to aphelion’s face than to perihelion’s.

  • Kamal says:

    Robin: What is astonishing to me about these recent discoveries is how much work day-night and summer-winter cycles can do on this comet. Wikipedia says that before 1959 67p had an orbit with perihelion 2.7 AU, so it has now had nine closer perihelia. Can we identify any phenomenon which would not have been possible at the farther perihelion and which must necessarily have happened in the last 50-odd years?

  • ianw16 says:

    I’m not sure why the ability to reply to certain comments has disappeared, so I guess I’ll have to do it here.
    Again, I’m afraid that is just wrong.
    First, let’s deal with the Earth-based SWAS early findings; they use a sub-millimeter array, which picks up an emission line (i.e. H2O vapour) of a particular rotational state of water at ~557 GHz, similar to MIRO. They were a long way from the action, and these were early results. The IR spectrometer on DI was very much closer to the action, and was able to observe the actual plume from the impact, and detect not only H2O vapour, but absorption from ice grains in the ejecta (which, obviously, SWAS couldn’t).
    Secondly, I’m afraid you are just reading the DI instrument paper wrongly. The best illustration of what I am saying is in part 2, p46, fig. 1. It quite obviously shows the IR light being split from the visible light by the dichroic beamsplitter. The optical light carries on its way to the CCD and then the CCD pre-amp, and then the CCD electronics. The IR light, on the other hand, is diverted to the IR spectrometer and the HgCdTe IR FPA (focal plane array). It then carries on to the IR electronics. Never, at any stage, does it go near a CCD! It is further illustrated in fig. 4, p51.
    This was a state of the art instrument, not something cobbled together in somebody’s attic using an old digital camera and a pair of 3d glasses!
    When the author of the paper that presented those findings says, “Our instruments are good enough to not only identify the water ice, but constrain the sizes and they are both consistent with laboratory data at the appropriate temperature.”, then that is good enough for me, not to mention the many well qualified scientists who have no doubt studied that work.
    I realise the EU folks have a very good reason for dismissing those findings (as in it kills the electric comet stone dead), but they have no scientific (or mechanical) evidence to back it up. It is purely due to their inherent belief that they can’t possibly be wrong, so everybody else must be.

    • Sovereign Slave says:

      Yes, once CAPTCHA has gotten tired of too much posting chatter under successive posts, it denies the reply link. You have to go to the last reply link shown and click and post, even if it’s several posts higher in the chain. It will still post yours at the bottom though.

      Regarding your response, seems very thorough and well laid out. As a non-scientist, I’m at the limits of my abilities to decipher the paper in question., but if all is as you say, this does represent a major bugaboo for EU.

      I’m still confused by the statements in the article about Deep Impact, and was wondering what your take on them are since you’ve looked into this so much. Especially these,

      “It’s pretty clear that this event did not produce a gusher,” said SWAS principal investigator Gary Melnick of the Harvard-Smithsonian Center for Astrophysics (CfA). “The more optimistic predictions for water output from the impact haven’t materialized, at least not yet.”

      “Theories about the volatile layers below the surface of short-period comets are going to have to be revised,” Qi said.

      • ianw16 says:

        I think the view was that the layer they impacted was ~ 1m thick. As they describe in the paper, this may be at the limit of the depth to which heat can penetrate, given that comet’s orbital period and perihelion distance. However, it is now apparent that some (possibly most) short period comets have had their orbital parameters changed by interactions with planets, mainly Jupiter. So some may have been in more distant orbits, without as much devolitilization, and therefore will have the ice closer to the upper dust layer. Others will have different “seasonal” variations”, seemingly like 67P, where the south has a very short summer, but it happens at perihelion. How long they have been in close perihelion orbits will also have an effect.
        So, they are very variable beasts!
        I should mention that the ice grains at Tempel 1 (as opposed to the vapour) were also detected by the XMM-Newton space observatory:
        The DI spacecraft also detected H2O ice grains at Hartley 2, strongly correlated with vigorous CO2 outgassing:

        Regarding the SWAS reports, they were only a week after the impact, and they were only capable of detecting H2O vapour. Observations by the Chandra and SWIFT X-ray telescopes detected charge exchange between the solar wind and the ejected material for ~ 12 days. The last estimate I saw for a value, was ~ 4.0 x 10^6 kg of H2O (about 4000 tonnes). Definitely not as much as they’d hoped, but telling results (particularly the grains) nonetheless. I suspect that if they’d held on to the impactor, and smashed it into the small lobe of Hartley 2, then things may have been somewhat more spectacular!

        Just for clarity, the three instruments on DI that had CCDs were the HRI optical instrument, the MRI optical instrument and the impactor camera.

Comments are closed.