New perspectives – CometWatch 6 February

Today’s CometWatch entry presents the NAVCAM’s first single frame image capturing the entire comet nucleus since leaving bound orbits last week. The image was taken on 6 February from a distance of 124 km to the centre of Comet 67P/Churyumov-Gerasimenko. The 1024 x 1024 pixel image frame therefore has a resolution of 10.6 m/pixel and measures 10.8 km across.

NAVCAM image of Comet 67P/C-G taken on 6 February from a distance of 124 km to the comet centre. In this orientation, the small comet lobe is to the left of the image and the large lobe is to the right. The image has been processed to bring out the details of the comet’s activity. The exposure time of the image is 6 seconds. ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0

NAVCAM image of Comet 67P/C-G taken on 6 February from a distance of 124 km to the comet centre. In this orientation, the small comet lobe is to the left of the image and the large lobe is to the right. The image has been processed to bring out the details of the comet’s activity. The exposure time of the image is 6 seconds. Credits: ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0

The image provides a stunning contrast to the recent close-up images, offering new perspectives on the extent of the comet’s activity. Indeed, the jets emanating from Hapi (the neck region) extend towards the edge of the frame in the upper right. Adjusting the intensity scaling, as we have done in this image, also emphasises the nebulous ‘glow’ of activity that appears to be coming from all over the sunlit surfaces of the nucleus. Bringing out the jets also highlights the large amount of background ‘noise’, which includes material ejected from the comet.

Today, 11 February, Rosetta is about 100 km from 67P/C-G and moving closer ahead of its 6 km flyby from the surface of the comet on Saturday – closest approach is at 12:41 GMT (13:41 CET). The NAVCAM is scheduled to take images 1-2 hours before and after closest approach, which will be downlinked to Earth Sunday/Monday.

The original 1024 x 1024 pixel image frame for today’s entry is provided below:




  • Ross says:

    Any images of the prominent vents which ‘must’ exist to produce these two spectacular jets in the neck region?

    • Andi Rossini says:

      This is what has baffled me from the start. The high-resolution images of the surface must be sufficient to conclusively identify these supposed source vents that are powerful enough to eject collimated and filamentary jets of gas and dust at supersonic velocity into the vacuum of space, even at hundreds of millions of miles from the sun. So where are these images?

      At the same time, I listen to Rosetta Mission staff stating that those beautiful sand-dunes we see in the dusty regions of the comet surface are likely produced by ‘wind’ from these same theorized vents, this time somehow blasting parallel to the comet surface, again in the vacuum of space. Are we now supposed to re-invent the physics of the theorized behavior of neutral gases in a vacuum?

      Can there not possibly be other forces at play? Why are the academic halls so averse to even considering an alternative theory (and there are credible alternatives), when plainly the standard model has failed every possible test as it applies to comets? I was under the assumption, as I was taught in the sciences, that if your theory does not match observation, you change your theory. That is, if your theory predicts one thing and observation proves it to be another, it is your theory that must be abandoned for one that would have predicted the observation. And you certainly don’t go about introducing ad hoc amendments to the very physical axioms you based your disproved theory on in the first place, just to salvage it. Consider what the philosopher and professor Karl Popper said; “Whenever a theory appears to you as the only possible one, take this as a sign that you have neither understood the theory nor the problem which it was intended to solve.”

      I say this with all due respect to those involved in the Rosetta Mission, because in the end I do admire the work they have accomplished. I am in awe of the feat of engineering and cooperation that went into actually landing on a comet and I applaud the efforts of everyone engaged in the endeavor. However, I find it very difficult to take seriously the circular and cloistered reasoning of many I have heard as they attempt to explain away obvious contradictions between theory and observation.

    • Dr Shekelsberg says:

      They are created by sublimation of water ice that erupts through the frozen surface. Don’t worry, comet science has been well understood for decades so there is plenty of facts about comets available.

      • logan says:

        Hi Dr. Shekelsberg. As a workaround for the fonts problem you can disable temporally Javascript. You can activate later to contribute.

        • logan says:

          Another workaround would be to change WebBrowser. Opera show gray. My mozilla show black. Mozilla’s let you force default fonts, don’t know from others.

  • logan says:

    Hi Emily and H. NAVCAM team. Until now I was convinced that jet curvature was dominated by 67P spin. Not so sure now…

    • Graham Hall says:

      if the material is moving away very fast (circa 700m per sec according to Miro) then it would reach the edge of frame position in less than a minute. Curvature would be slight as the rotation is comparatively slow. The further away Rosetta is the more the curvature would become visible, assuming the jets remain collimated at distance.

      These current jets are apparently nothing compared to the activity later on, although that will be from the current dark side. At the moment Hapi valley is the most active area, but if that is in shade later on will the outgassing be as dramatic and limited to a similarly few number of regions. Its unlikely that the is a equivalent valley on the dark side!

      • THOMAS says:

        @ Graham Hall

        “if the material is moving away very fast (circa 700m per sec according to Miro) then it would reach the edge of frame position in less than a minute.”
        Supposing a length of 7 km, it would take precisely 10 seconds, and the comet rotation could thus have absolutely no detectable effect on the jets, so that particular standard theory mechanism just doesn’t work.

    • Robin Sherman says:

      Hi Logan. I tend to think this image lends credence to the comet’s spin being the cause of the curvature in the duct columns.

      Lets assume the particles of dust are not moving that fast say a few metres per second. A reasonable assumption as they don’t travel that far before their motion becomes random. The surface of the comet is also moving, the further away from the rotation axis the faster it moves. A rough estimate of the height of the dust columns emerging from the lobes, would be about 500 to 1000m and if we take a plausible speed of 5m/s for the dust it would take 100 to 200 seconds to travel to the top of the plume. In that time the surface will have moved as much as 50 – 100 metres.

      For the dust columns emerging from the Hapi region, the surface is close to the axis of rotation and so the speed at which the area of surface producing the column moves is far less. This has two effects, one the columns are more focused and concentrated and two, the energy of the sublimating gas is not dispersed over such a large volume and so the dust has more energy and goes faster and further.

      A hosepipe spraying water is a simple analogy. If you hold the end of the hosepipe relatively still, most of the water will end up in the same spot some distance away. If you move the hose pipe across your body in front of you, the water is spread out over a larger area, curves and does not travel as far.

      This is what we see on the comet. Those columns further from Hapi curve more, are less, concentrated, disperse sooner and don’t travel as far. For the dust columns in the neck area, the source becomes more of a “point” source”, narrower, less dispersed columns form and more energy is concentrated into the column, they stay coherent for longer and travel further.

      The evidence from the science team is the Hapi region has a higher percentage of volatiles in its composition, as evidenced by the bluer slope in the IR spectra, so gas pressures to start with may well be higher. A double whammy if you will. The columns are more focused and being powered by a higher gas pressure.

      Whether this is actually the case is another matter, but that is how it appears to me. When the energy reaching the subsurface increases and penetrates deeper, then we might see evidence of subsurface cavities and proper jets as the pressure of trapped gases forces gas and dust through cracks and vents, possibly at supersonic velocities.

      What we have seen up until now seems to be little more than variable amounts of diffusion through the dust layer from the low volatile content layers just beneath it. The deeper the energy penetrates below the surface, the higher the amount of volatiles available to sublimate and the greater the activity, is the theory. We await developments to see if this is indeed the case.

      • THOMAS says:

        @ Robin Sherman

        “A rough estimate of the height of the dust columns emerging from the lobes, would be about 500 to 1000m and if we take a plausible speed of 5m/s for the dust it would take 100 to 200 seconds to travel to the top of the plume.”

        What makes the “5m/s” speed for the dust “plausible? It starkly contradicts the only data we have been given on jet speed, which is 700 m/s, i.e. 140 times greater… The “5m/s” speed for the dust is not at all plausible, it is simply *required* to make your hose-pipe model for the curved jets work. Do you consequently believe that the more spectacular curved jets in the thread were also travelling this slowly?

      • logan says:

        Hi Robin. Thanks for the answer.
        “…a plausible speed of 5m/s for the dust it would take 100 to 200 seconds to travel to the top of the plume”.
        Really I declare myself neophyte on the issue. Tend to agree with you in that gas as dust has orders of magnitude different speeds.

        • Harvey says:

          There is a lot of confusion here regarding the speed of the *gas* – which we cannot see – and the *dust* which we can.
          The gas speed is set by gas dynamics, and one might reasonably expect ‘choked flow’, and a gas speed roughly limited to the speed of sound in the gas by formation of a shock wave. The only direct measurements I’ve seen, by MIRO, or the Doppler shift of the water vapour, are of this order, for the *gas*.

          But the dust has to be accelerated by viscous interaction with the dust. Broadly that means the mean free path in the gas must be less than the particle size, and that will fail some very short distance from the comet as the pressure falls. The acceleration phase is very short, and very inefficient, because of the low pressure. It will be nothing like ‘a bullet in a gun barrel’.

          So it is reasonable to think the dust velocity may be much lower than the gas velocity; to quantify that would need a lot of detail of the geometry etc which we don’t have. It’s also possible there will be a correlation between particle size and velocity.

          However, it probably can be deduced quite easily from the observed distortion of the dust jet with distance due to rotation; it could also be done by correlating the position of fluctuations in intensity in the jet.

          Have there been direct deductions of *dust* velocity I’ve missed? – entirely possible.

          • Harvey says:

            Sorry, ‘……interaction with the *gas*’

            It’s interesting to put some rough numbers into this.
            Ice vapour pressure
            Mean free paths can be found here
            They are given at atmospheric pressure, 10^5Pa
            They scale inversely with pressure.
            So it turns out that at pressures of the order of the ice vapour pressure at comet temperatures, ~Pa, the mean free path is order microns, tens of microns.
            The pressure will fall very rapidly away from the comet.
            Which suggests to me that acceleration will be very inefficient, and dust velocities much less than gas velocities.
            (Note that achieving choked flow does not depend on absolute pressure, but on upstream/downstream pressure *ratio* which has to exceed a critical value of order 2 depending on the gas, and the geometry has to be a good many mean free paths in size. So the gas can still achieve ~ sonic, but the dust probably is far slower.)

          • logan says:

            Thanks for the guiding, Harvey. You have commented before about a ‘limit’ where viscosity ends to be relevant. Is that related to the triple point of water vapor?

            Bare tables a little too much for my knowledge level.

  • logan says:

    Some day at this perihelion 67P is going to suffer a catastrophic event at base of neck. Former ‘night’ side? 😉

    • Harvey says:


      Not directly to the triple point.

      For our purposes, there are two flow regimes, ‘viscous’ and ‘molecular’; everything we are used to, turbulence, vortices, shock waves etc belongs to the viscous regime. Most people have no experience of molecular flow directly.

      Which applies depends on the ratio of the mean free path, MFP, to the physical scale of the system being considered.

      The MFP a is simply the mean distance a molecule travels before it hits another molecule. So it depends fairly weakly on temperature and what molecule it is, but strongly on pressure; the higher the pressure, obviously the shorter the MFP; its a simple reciprocal. For some ‘typical’ gas at 300K and one atmosphere, 10^5Pa, the MFP is around 0.5nm.

      What pressure should we use?
      Well one sensible value would be the saturated vapour pressure SVP of ice at comet temperatures. That is of order a few Pa. So the MFP is of order tens of microns. Now if you argue there is a lot of CO2, the vapour pressure would be a lot higher; but actually we can’t expect to get the full SVP, because there is a great big pump (the Universe! 🙂 ) sucking it away.
      So a MFP in the order of microns, tens of microns, is about right, dropping rapidly as you move away from the comet.

      If the size of your ‘system’ is small compared to this, molecular flow; if large, viscous flow.
      So dust particles are probably borderline, but will easily become molecular regime as the pressure drops. But anny aperture, ‘nozzle’ on a scale of mm or more is viscous, shock waves can form etc.

      The viscous flow regime we are used to. But in molecular flow, the random thermal motion becomes very important. The molecules are acting essentially independently, and don’t ‘know the other molecules are there’ so to speak, because they don’t hit them.

      This difference is crucially important in the design of vacuum systems, and in understanding what might be going on on 67P. A good example would be what happens to any dust, scrap of foil insulation etc in a vacuum chamber. As you start to pump down starting at one atmosphere, anything like that gets ‘blown around’ – but quickly, as the pressure drops, nothing moves. The low density gas can’t transfer enough momentum to the particular to accelerate it, and by the time you get to a Pa or so (crude vacuum) small items would be locally in molecular flow; nothing moves, despite gas still being pumped out of the chamber rapidly.

      • Harvey says:

        Sorry, *pressure* dropping rapidly as you move away, MFP *rising* of course. Not very far from the comet the MFP proble reaches metres.

        • logan says:

          Thanks a lot.
          “For some ‘typical’ gas…” Our breathing air 🙂

          So, at 67P, all ‘flying’ dust and particulate out of the proximity of a jet source is essentially in a ‘Free Path’. [I am deliberately ignoring the statistically far possibility of that particle having electric charge].

  • logan says:

    Now, this is the most of the madness…

    Could be error camouflage matrices on my very old LCD screen.

    Could be lightly ‘ordered’ depositions on NAVCAM lens.

    Could be .JPG file format limitations.

    But, “…the large amount of background ‘noise’, which includes material ejected from the comet”, looked in a very ‘conjunct’ setting of mind, doesn’t look kind of… ‘globularly ordered’?

    Ejected particle’s location doesn’t look absolutely random to me. Isn’t that phenomena expected, before the bow shock formation?

    Ducky has a soft, foamy solar wind front, until now.

  • Jacob Nielsen says:

    “perfectly collimated beams”? more like a bunch of hastily aligned floodlights. It’s just dust hosing into space, no magic tricks.

    • THOMAS says:

      @ Jacob Nielsen
      “no magic tricks”


      There is nothing magic whatever about curved electric discharge arcs collimated by the electric fields which spiral around them, as required by the basic laws of physics. Such arcs are produced routinely in the laboratory and are a standard feature of the electromagnetic activity on the Sun’s surface. The fact that they are being observed on the surface of 67P proves the electrical nature of comets and hence of the cosmos in general.

      • Jacob Nielsen says:

        @THOMAS, I was merely pointing to the impression of a gradual dispersion of the ejected dust… Regarding your theory about the driving forces: I do think you are making too big extrapolations from observations with (some) visible similarities. The properties of the sun just does not compare to that any comet: sun is a rotating super conductor creating immense magnetic fields. No other body in our solar system has those properties. Besides, I do not see any real similarity between “dust hosing” on a comet approaching sun, and something like coronal mass ejection.

  • THOMAS says:

    Those extremely powerful straight jets emanating from the least sunlit region of the comet, Hapi valley, apparently extend for at least five kilometres if they are perpendicular to our line of sight and considerably more if they are tilted to a significant degree away from or towards

    Does anyone know how the standard model might still account for them by the putative mechanics of sublimating ice? Any offers will be very welcome, since the comet’s activity is expected to increase exponentially over the coming weeks and months and it will be extremely interesting to be able to match precise standard theory explanations against what we shall soon be actually witnessing…

    • marcoone says:

      This dust is in the foreground with orbiter.
      Orbiter is 100km, dust and gas are at 100km and more.
      This is already coma and tail , for me

      • THOMAS says:

        @ Marcoone,

        Rosetta may be 100 km from the nucleus here, but it’s actually deep inside the coma. You can’t take a photograph of a house if you’re sitting in the kitchen…

        To give some idea of the distances involved: this image of the comet, taken from Earth by the VLT in Chile way back in September, when activity was presumably much less than it is now, already shows a coma which was “estimated to be 19,000 km long”.

        The jets we see in today’s NAVCAM image are actively *feeding* the coma, otherwise it wouldn’t be there.

      • logan says:

        Hi Marcoone. You are right. Most of this particulate/dust should be near to camera. I guess ‘background’ was used as a figure to contrast to the main subject: the jet aura.

        Is 124km beyond of the actual solar wind front?

    • Judy says:

      The ISSI book, Heat and Gas Diffusion in Comet Nuclei, which I’ve got access to via my library card at the local state university, has this to say:

      High pressures may develop in dense, fine-grained nuclei. The
      peak pressure always occurs near the source of the gas, be it sublimation
      or gas released from crystallization of amorphous ice, provided this source
      is sufficiently deep. Near the surface the pressure is low since the comet’s
      environment is practically a vacuum. When the pressure is exerted by gas
      released from crystallization, the peak occurs near the boundary between
      crystalline and amorphous ice, at a depth of a few tens of metres, typically,
      declining gradually toward the surface and toward the centre.

      So that would be a model that applies to your question. Their model is of a highly porous, largely homogenous nucleus.

      It’s worth reading the book, if you can get access. There’s a lot of very interesting and illuminating information in it.

    • Steven says:

      We have witnessed these filamentary tails for decades, extending millions of miles – directly in opposition to how neutral gas and dust behaves in the vacuum of space.

      • martinvolwerk says:

        The cometary tails consist of draped magnetic field and plasma, they are not the jets that you see in this picture.

  • THOMAS says:

    Explanations would also be welcome as to why the fainter jets to the right of the main central jets are curving away to the right. There seems to be a distinct pattern emerging.

  • Rod says:

    Which ice is sublimating, the crystaline near the surface or the amorphous ice deeper down?
    Are the boulders and exposed cliff faces they appear to come from all dried out?

    • Graham Hall says:

      The surveys have not revealed ice (whether CO, CO2 or H2O) on any of the observed surfaces so yes the surface is denuded of ices. The cliffs are probably made of relatively thin crusts of compacted material, covering a different structure below. The Hapi valley material is different in composition to the material on the other sunlit areas of the comet and this might suggest that the crust surface has been lost from the neck area and a different “softer” material underneath sublimes more readily.

      • logan says:

        Hi Graham. Thanks for the answer.

        I have nothing to argument with you, have to accept MIRO data about gas speed. Also accepting that dust ejected trough big nozzles should have the time to be accelerated to high speed by that gas. Nowadays, that’s not the case of surface or near surface sublimation. There I should accept Robin’s modeling.

        “…Hapy Valley… Its unlikely that the is a equivalent valley on the dark side!”
        I would prefer it to exist 🙂

        • THOMAS says:

          @ Logan

          “dust ejected through big nozzles should have the time to be accelerated to high speed by that gas”

          Sorry, Logan, but according to the basic principles of hydrodynamics, “big nozzles” are too big to accelerate anything. Unless, of course, you hike up the upstream pressure exponentially and that’s ruled out by the CONSERT data findings (because of the proven absence of voids larger than 2m in diameter).

          • logan says:

            You are right Thomas. I should say ‘Long Nozzles’. A rifle has longer range than a pistol, on equal ammunition. Cryo-volcanoes have ‘long nozzles’.

          • logan says:

            CONSERT data has so few non consolidated readings as to be able say just one, or two affirmations of a unknown corner of Ducky. Why THOMAS takes this drops of info as enough to say what is absent at 67P’s insides?

          • THOMAS says:

            The “just one, or two affirmations” from the CONSERT team were unambiguous: a homogeous interior with no voids larger than 2m in diameter (if that…). They wouldn’t have stated this if they hadn’t been sure of their findings.

          • Marco says:

            I think long nozzles would explain a lot actually. apart from being consistent with CONSERT data, and enabling faster gas flows through the nozzle, it also gives more opportunity for the gas to pick up and accelerate dust before the sudden pressure drop on exit of the nozzle. Ok, the deeper areas are shaded and insulated from the Suns heat, but so is Hapi Valley. Whatever can make Hapi Valley hot may also work for something in the interior of the comet. I think you are brilliant, Logan.

    • logan says:

      Hi Rod. My personal bet: Deep ice (the little that remains) at surfaces close to gravity center. Superficial (and smaller amounts of deep) ice at surfaces far from it.

  • rod says:

    Does the main jet (not just visually, but in volume) of particles, follow the line of the axis of rotation?
    If it does, and there are much more particles being expelled there than elsewhere, and it is darker there, that’s likely more than a coincidence isn’t it?

  • logan says:

    Hi Rod. As of today, I would like to believe that jet sources near to actual gravity center of Coraline are energized [mostly] by interaction of magneto-[atmo]-sphere ‘MAS’ with nucleus material.

    Gravity center just happens to be the most common visitor of the magnetic axis. [Not the zone most magneto-resisted].

    Also wanting to comment that ‘MAS’ geometry and density are dynamic and far from a spheric shape.

    The magnetic axis will be drawing a small and energetic circle around South Pole [And also around North Pole] on perihelion.

    • logan says:

      ERRATA: Should say:

      “I would like to believe that… The sun-comet axis will be drawing a small circle around South Pole [There goes the radiative, superficial energy]. And the magnetic axis will be drawing an equatorial ‘neck’ [There goes the magneto-resistive, profound energy]. That’s the correct geometry at perihelion.”

  • Solon says:

    “The exposure time of the image is 6 seconds.”

    Doesn’t mean much to me without some kind of comparison to something familiar, like my own camera. Is there some way to come up with what exposure settings I would use on my camera to obtain such an image? An f/0.7 lens, ISO 256,000 and a week long exposure maybe?

    • Harvey says:

      There is a publicly accessible description of the OSIRIS a camera here

      The F number is easy, for the narrow angle camera it is F8. The relatively high F number will be largely because the system uses off-axis mirrors, which tend to force fairly high F number. There may also be a depth of field reason not to go too low in F number, I’ve not done the sum.

      It’s hard to make a comparison to ISO (though possible if I had more time!) but one can make a general point.
      The CCD is 2048 square, 13um pixels, so about 26mm square, not so very different to an up market full-frame DSLR sensor. The technology is basically similar.
      The big difference is that the Rosetta CCD is certainly operated at low temperature. This greatly reduces the ‘leakage current’ and is of particular benefit for long exposures. The leakage is probably reduced by a factor of perhaps a couple of hundred by operating cold. The actual improvement is only the square root of that factor, so maybe 4 stops in camera terms better. There is a further complication regarding when readout electronic noise, and how that reduces with temperature, becomes the limiting factor I’ve not looked in to.
      But as a VERY very rough comparison, an up market full frame sensor DSLR at F8, but with a sensor maybe four stops more sensitive, might the right sort of area.

  • logan says:

    Ducky’s neck looks more tropical than equatorial.

  • logan says:

    The enlightening posting from Claudia on albedo theme on October give us too an example on a plausible former equatorial ‘neck’.

  • logan says:

    From the same Claudia’s posting on albedo, I can now say that Asteroid Steins looks like a Ducky’s head.

  • Harvey says:

    Logan is correct re CONSERT.
    They only got very limited data.
    Their report is absolutely clear.
    For a very restricted number of paths, which happen to be rather tangential and close to the surface, they did not see any large voids on a scale somewhere in the low metres.
    They simply did not get data on the deep interior of the comet, there was not enough time and Rosetta was not correctly positioned.

    • THOMAS says:

      One of the paths was mentioned as having penetrated I km below the surface. The interior has been described by Rosetta researchers as “homogeneous”. Mission scientists have presented the limited CONSERT data acquired as being nevertheless representative of the interior of the nucleus.

      • Harvey says:

        1km path length *in the comet*; but not 1km ‘below the surface’. The path was tangential, as the locations clearly show. They also had poor S/N on the longer paths, which is worrying for ‘through centre’ data which might be acquired later if Philae ‘wakes up’. They may need a lot of integration time. Because loss is exponential with distance, say 4km can be enormously harder than 1km.

        This is not about ‘what model is correct’. It’s about being careful not to over interpret results. When, if, we get through-centre CONSERT data, who knows what it will show. For the moment, CONSERT a data is really rather limited.

Comments are closed.