CometWatch – 5 February

Today’s CometWatch is an image taken with Rosetta’s NAVCAM on 5 February 2016, when the spacecraft was 53.4 km from the nucleus of Comet 67P/Churyumov-Gerasimenko. At these closer distances the comet is now close to filling the camera’s field of view.


Lightly enhanced NAVCAM image of Comet 67P/C-G taken on 5 February 2016 53.4 km from the nucleus. The scale is 4.6 m/pixel and the image measures 4.7 km across. Credits: ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0

In particular, this image captures the rich diversity of geological features within Imhotep and bordering regions on the comet’s large lobe. Imhotep was also the focus of recent scientific studies reporting exposures of water-ice and dramatic changes in the surface features observed in the central smooth plains area.

In the background, the Wosret and Bastet regions of the comet’s small lobe are featured.

The original image is provided below:




  • A. Cooper says:

    That’s an interesting array of circles where Bastet tucks round towards the neck. All with lines tucked tight between them pointing outwards from the rim.

  • Henrik Krag-Jensen says:

    With this picture of Imhotep, I can’t help it:
    Is it possible that the roundish elevated area can have been situated in the depressed area of same shape – rotated 180 degrees and shifted somewhat to the right (areas named F and E in the 20/07/2015 post) ? Some buildup of gas pressure in an underground fracture could have pushed this area from the surface – then it rotated and fell down again

    • Ramcomet says:


      Sharp eye! I just traced the disc-slab right off my tablet screen, flipped it around 180° and placed it over the “crater”. A very good fit.

      Most of it matches very well in shape and size with the exception of a probable broken-off missing point that would have filled the top of the crater shape… Easily explained by some of the nearby rubble fragments.

      Also, the disc-slab seems to be missing a crescent shaped bite out of it to match the current crater at bottom. But that is OK, it really looks like a “glacier” flowed in from the bottom of the crater AFTER the slab blew off.

      I noticed other dark divots with the same kind of strewn out (blown out?) texture around the comet. It is good to start recognizing what the shapes and textures mean in the context of missing or moving slabs.

      Andrew, I am not so good with identifying the rotation planes. Does this possible slab movement comply with all the other objects you have identified moving along the rotation place? (Not that it is a requirement… Assymetrical outgassing could have blown the slab towards any direction)

      • A. Cooper says:


        You beat me to it with the tracing! I’ve even got he tracing paper ready for this sort of thing. I agree it fits quite well with a 180° swivel because the disc-slab is fairly symmetrical about its longitudinal axis (up/down axis in this view).

        Or rather, it’s symmetrical when you don’t include the ‘shoulder’ I mentioned that bumps out on the left. That doesn’t look as if it belongs to the disc-slab at first glance and it hasn’t even been discernible until recently due to seasonal shadowing and then orbiter distance from the comet. I always thought it was a circle because of the early pics with the shoulder in shadow. So I subtracted it in my mind’s eye and started swivelling the disc-slab 180°. It was looking good till I saw the missing crescent you mention and that made me remember the OSIRIS photo (linked in my reply to Henrik). That’s the pic with the croc head that’s the shoulder. So I looked again and saw the shoulder matching to the crescent if you do a simple translational slide.

        I’ve now found quite a few mini matches. One good one is the little 45° line of five squares just below the rebate at bottom left of the disc. It also has a v-shape below it. The line of squares and v-shape match to the depression (to the bottom of the crescent in the depression). The squares are a bit smudged but the orthogonal signature is there along with the angle of the line and the v. Also a mini match within the v.

        The direction of the slide is at 90° to the rotation plane (equator). That means it’s at 90° to the hypothesised long axis stretch vector proposed in stretch theory. The stretch vector is a tensile force vector. The recoiled crust signatures that I’ve found at Babi and Seth are also at 90° to the stretch vector and they are also around 800 metres like this slide (Part 40 on my blog). So it bears the signature of crust recoiling. It recoiled away from the long axis stretch vector as the crust unzipped via shear-tearing along the tensile force vector. In this case, the tensile force vector runs along the equator and parallel to it.

        The rotation plane movement you asked about is more for things that get launched straight upwards and travel suborbitally, landing further back down the rotation plane. I think the crunching down of that bottom lip of the disc that’s upturned, crunched the tip of the big boulder. On rebounding the boulder released the chips upwards and they floated down the rotation plane. I only say this because Cheops is exactly down the rotation plane and bears remarkable matches to the big boulder. If you mirror the crunch process to the rocks on the other side of the ‘dusty’ plain and and give them the same rotation plane angle and distance of travel, they can be traced to that line of squares mentioned above. That’s also on the bottom perimeter of the disc-slab (shoulder/crescent included). There is at least one big boulder perched next to that proposed original seating position (see red boulder field map in the linked Inside Imhotep post).

    • Dave says:

      Henrik re roundish elevated areas.

      There are several areas like this around the comet, however the ESA blog ‘Inside Rosetta’s Comet’ states that Consert ‘measurements show that there are no large cavities on the comet head. Seems slightly strange due to the huge caves we can see.

      Also you might expect some shallow craters to form near the surface based on the mechanism described to bring the subsurface ice close to the surface so that it can the sublimate into space. It seems likely that there maybe some pressurised craters near the surface if this type of sublimation is true, where sublimation either seeps through a permeable surface or even pressurises an impermeable surface to form a cavern

      However, how this can blow off or hinge off in one piece, I guess depends on the properties of the crustal material and the unnatural (to earth dwellers) conditions at the comet.


    • A. Cooper says:


      Great observation. I’ve been puzzling over this round feature for ages. I think you’ve found the answer. Except there’s a catch! I don’t think it rotated 180° but simply popped out and slid. I’ve found many signatures of long crustal slides around the whole comet. It’s a secondary signature of stretch theory but I should point out that since stretch theory isn’t mainstream and not entertained by the Rosetta team, crust sliding isn’t entertained either.

      Although this feature looks quite round, it is in fact attached at its left side (in this Rosetta post’s view) to a large expanse of crust. That would mean that for it to rotate, the whole expanse of attached crust would have to rotate with it and find a matching home on the far side of the depression. That’s a bit difficult to achieve. You can sort of see it’s attached at its left in this post’s photo. But you could conjecture that it’s an optical illusion due to overexposure and say it might be detached. However a recent OSIRIS shows definitively that it is attached:

      In this linked photo, it’s difficult to believe that what you’re looking at is the same round feature. It’s the horseshoe shape viewed in high profile. It’s just to the right of centre and is below the mesa in the distance on the horizon behind it. This side of the curve (as viewed here) is bounded by something that looks a bit like a crocodile’s head and snout (facing left). And it’s only when you adjust your brain visualisation to the exposure that you realise the open end of the horseshoe isn’t open at all but is the dust-covered end of the round feature. The attached crocodile head and extra crust lower down shows that the round feature isn’t isolated and couldn’t have rotated.

      Moreover, the linked photo suggests that the crust that the round feature is attached to is tearing away from the round feature and sliding further towards the edge of Imhotep (towards us). The direction of slide is in the same direction as you suggest for this round feature, away from the depression (the depression is out of sight directly behind the ‘horseshoe’). Looking at all the info in both pictures, it suggests the round feature popped out of the depression and slid with the rest of the crust it was attached to. And the slide was towards us in the linked photo view.

      The tearing/sliding signatures are the fact that the crocodile head is split along its length and that weird, bright, curved feature appears to have broken and slid from the edge of the crocodile head. It became curved as it slid. This degree of slide is commensurate with the displacement from the depression that you cite and wholly consistent with a wealth of evidence for crust sliding signatures I’ve found in other areas of the comet.

      So that leaves the conundrum that a pure translational match doesn’t look as compelling as your suggested rotational match. That’s what I thought at first, but on looking again from a distance and knowing that it has this attached mass to the left, I could see that the depression does indeed retain the imprint of the attached mass. It is in the form of a ‘shoulder’ to the profile of both depression and round feature that is halfway down on the left side. This corresponds to the bulging left cheek of the crocodile head. Furthermore, both depression and round feature have a sudden rebate turning to the right at 90° (in this post’s photo) which is located just below the ‘shoulder’ or croc cheek. That corresponds to the back of the croc’s head. Both rebates display the characteristic circles of the depression though not quite in perfect matching configuration but these circles are found nowhere else on the comet outside the depression. Except in this small area that matches to the depression.

      When you use the shoulder and rebate to click your round feature (now not so round) back into the depression it gets shunted upwards a bit, which explains a multitude of things, most notably why only one end of the round feature has dust on it. The ‘dust’ is what I and Marco think is flash-frozen slurry. That end fits to the slope up to the flat plain and when fitted back, the line of ‘slurry’ across the round feature is consistent with the slurry glooping down the slope towards the depression. Slurry isn’t glooping into the depression from any other place on the depression perimeter.

      This match also explains several other things including the behaviour of the large broken boulder on the perimeter. And now I’m starting to find what I call mini matches to firm it all up.

      Thanks for your input, Henrik. I think it’s a step forward.

      • A. Cooper says:

        Henrik’s round feature exhibits multiple matches to the depression. They are on its surface and so are bled-through matches, evidently a continuation of outgassing from the depression layer up through the round feature to its surface. Do I really need to point them out? They really are obvious.

        Also, the bottom of the round feature (as viewed in this post’s photo) is the source of the ‘sunset jet’ in this recent Cometwatch post:

        And it’s the source of ice in the daily cycle in this post (2nd video @ 00:29):

        Seeing as the match to the depression puts this patch of round feature ice directly over the well-documented ‘blue’ areas along the edge of the depression, it follows that even the ice signature matches. This is the only area of ice on the round feature and it matches to that very strong signature in the depression. The bled-through matches mentioned above and known porosity explain why the surface of the round feature would betray ice tens of metres above the ice source it originally sat on.

        The area on the round feature that has the ice is bent up. So is the depression where it was seated.

        The large rock in the depression has an impression directly below it that mirrors its edge profile suggesting it was pressed into the matrix by something. Its end is also chipped off along the line at which the round feature would have sat. These two factors point to the general upheaval of the round feature breaking out and away and are suggestive of it tipping as it departed from the depression.

        The Rosetta post that shows the big boulder with the compressed section is in the Inside Imhotep post, linked here (see photo under “Accumulation Basins” heading):

        The section of the round feature that would have crossed the tip of the boulder to crush it down would have been the lower-right corner in this post’s view of the feature.

        In the above linked post it mentions how the scientists could already see there was something different about this round feature:

        “Basin F is observed to be slightly different in that it is extensively fractured, with the fractures pointing towards its interior. Since this pattern is not a feature associated with collapse, it must have formed or been modified in some other way, perhaps by impact or associated with activity, perhaps even by a gas bubble rising from the interior (as already proposed by other scientists).”

        F isn’t an accumulation basin. It’s a chunk ejected from the depression. That’s why it looks different.

        • Dave says:

          A cooper re henricks round feature.

          A perfectly reasonable explanation of the round features( cow pats).

          However Consert does not seem to have found any large cavities?


          • A. Cooper says:


            I think the cavern paper uses the shape model as a starting point, which means all the visible surface caverns or pits or sinkholes are baked in. So they’re just looking for subsurface voids.

            I can see how one could divide up the lobes into such and such number of unit volumes and integrate the summed GM/r^2 forces for any particular orbiter radius from c of g and at any angle. If the comet was homogeneous the unit volumes would translate directly to unit masses and when crunched through the integration algorithm they would generate a predicted orbital path (for any initially tangential approach just prior to the algorithm kicking in and bending it into an ellipse). But I don’t know if that’s how they did it. I read the abstract at the time and I don’t think it said.

            Ianw16 read the whole paper and made this observation regarding their stance on the remaining likelihood of cavities:


            So, it seems there’s still a reasonable likelihood.

            I wonder if the unit volume they used was the several hundred metres resolution they mention. That would translate into 500-600 unit masses if all were 330-metre-sided cubes. That would explain the resolution stated.

            Any deviation from the algorithm-directed path would betray a void. But the voids on the surface would be baked into the algorithm and unit mass distribution.

            Of course, you may mean to say that the depression from which Henriks pancake came must have been a hidden void with a gas pocket before it ejected the pancake. And therefore there should be others. However, the pancake and depression are of comparable thickness, suggesting that either a) only a small, flat gas pocket dislodged it or b) it simply slid under the strain of a stretching core (according to the stretch hypothesis). Or a combination of the two maybe.

            I don’t think you need big gas pockets to dislodge then slide due to stretch because less work is needed to dislodge than to dislodge and catapult. Dislodge and catapult involves extra energy lifting through the gravity field and accelerating sideways through a given time/distance to achieve enough speed to clear the depression- with precious gas escaping round the edges at the crucial moment of lift-off. With the slide induced by stretch, the lifting out of the depression and acceleration is provided by the tensile stress force vector on the crust. Less acceleration is needed but more friction needs overcoming. But the pancake appears to have slid on the all too controversial slurry (witness the concentric lines around it in the depression).

            I suppose large cavities would be more likely if the conventional sink hole hypothesis is valid: sublimation forming a cavity followed by collapse of the roof. Stretch theory explains each and every sink hole on the comet without recourse to sublimation and collapse. The mechanisms are any one of crust rifting, crust delamination, head shear and catastrophic outgassing prior to head shear. Crust delamination would explain the one at Maftet that Robin cites below. Delamination is the riding up of the head lobe onion layers (according to stretch theory- not a view held by the mainstream).

        • Robin Sherman says:

          We have seen these dislocations before and the OSIRIS team in it’s first paper highlighted areas of the surface that appear to have been lifted and displaced. Your analysis confirms this idea again. It seems to be a fairly common process which occurs at many scales.

          As an interesting aside, the lonely mountain on Ceres also sits beside a depression of very similar shape. Only in this case the ejected lump appears to have been flipped upside down. I have a very implausible theory that this was the result of an impact on the opposite hemisphere and the shock wave got focused to a point such that this mountain sized chunk was spat out, flipped and landed next to the hole.

          The suggestion by myself, Logan and Gerald that large area, but very shallow caverns could form between onion layers maybe applicable.The pressure build up would eventually result in a chunk of surface being lifted and moved, either by the rapid expansion of the released vapour, or due to the rotation of the comet. I have already speculated that some comet caves are the result of subsurface pressure bubbles distorting the surface and the resulting explosive decompression shattering the surface to leave a cave. It would seem to be a common mechanism, but with different outcomes depending on local geology, cometary material composition/properties and probably the local gravity field. The low local gravity field at the ends of 67P might explain why such a large chunk could be moved at Imhotep.

          The impact scenario suggested for Ceres could also apply, if indeed 67P is a contact binary, the initial impact may have had focused shock waves at Imhotep as well. Just as a suggestion, maybe the comet was actually stretched to breaking point, or perhaps smashed apart by some impact, but the two bits only drifted a small distance apart and recombined later as a contact binary. The similarities in terrains on the two lobes and individual matches of fracture planes between the two lobes in my view is just too clear to be dismissed.

          The impact scenario might explain the missing chunks at the ends of the two lobes. The excavation of a large crater, ejection of large amounts of material from the opposite hemisphere and subsequent asymmetric erosion of the neck area due to shadow/illumination effects and different composition of exposed volatiles deeper inside the comet, could also follow.

          Some ideas to cogitate upon anyway.

          • A. Cooper says:


            The pancake on Maftet that you refer to (in the early OSIRIS paper) is likely a similar sliding scenario to the one hypothesised here in the comments. It’s on the edge of what I have identified as a ridden-up layer on the head lobe. Although that pancake seems isolated and matches fairly well to a hole next to it, the rest of the layer rim matches to the layer rim below it. The pancake-to-hole match the OSIRIS paper found was just one match at the end of a long line of matches. I presume the pancake either broke away from its parent layer after being yanked out of the hole or is still attached and only apparently detached. That happens all the time, like the ‘crocodile head’ attached to this seemingly round pancake in this post (Henrick’s pancake or disc-slab).

            I should say that the two matching layers on Maftet don’t match quite as well as the clearly matching layers below them (they share mini matches and delaminated holes) but the layer in question is simply the next layer up above these in the sequence. And the general look of the perimeter does match, including of course the compelling match already identified by the OSIRIS team of the pancake to the hole.

            Regarding the Ceres match, it is a very strange phenomenon. It’s much bigger than the pancakes on 67P. Your idea of antipodal focussing of forces from a meteor (or CB collision) is similar to the mountain range on Mercury being the result of the antipodal forces from the impact crater on the opposite side.

            I think a large shallow cavern would work quite well for this pancake being simply loosened enough to slide. And of course, the sudden ~5 metre slump across the flat part of Imhotep last year implies just such a longitudinal erosion along the onion layer fracture plane. That would create thin, wide caverns perhaps a few metres high. Last year it seems to have caused a slump and the slumped area was very similar in shape and size to the depression. In the case of the depression and pancake it would have been loosened and also had a tensile force vector sliding it out of the depression (if stretch is invoked).

            I think you’re right about the cave in Hapi being a former cavern. You didn’t mention that above but you mentioned it long ago and that cave (and the hump above it in Babi) features a lot in my ideas. So thanks for bringing it to our attention re being a gas pocket. Also, your talk of ‘firn’ and the associated phase diagram from way back is looking more and more plausible with the porosity going up to 70-80% and the dusty homogeneity holding sway over blocky lumps.

        • Marco Parigi says:

          Hi A. Cooper,
          Unfortunately, you really need to point them out 🙂
          Plus, this is quite a bit away from the stretch action, and on the other side of the nucleus to the fracture, and also well away from the poles. I can’t quite tell whether the direction of movement is radial from the Paleo pole. Could the Imhotep plain have been a layer stretched flat then? That is with the tension transferred from centrifugal forces at the extremities?

    • A. Cooper says:

      Henrik and Ramcomet

      Here’s my version of the suggested match. I’ve found another suggested match that is in the exact opposite direction and same distance. The common tear line to both matches is right on the equator and the slide in both cases is at 90° to the equator (in opposite directions away from the tear).

      • Gerald says:

        A. Cooper, the central of the three yellow-dotted lines in your annotations
        looks very suggestive to me to be part of the upper edge of a mass wasting zone with a sliding vector towards the bottom of the image. “Above” this edge is an almost undisturbed regolith deposit with a smooth-looking surface.
        The mass wasting left over some better-“cemented” circular/roughly cylindric features (might be old vents, or material densified by impacts).
        The mass wasting may have been triggered by sublimation-induced subsurface mass loss, reminiscent of some of the Ocala sink holes:

    • Kamal says:

      Henrik: This was the area where some action happened in June last year.

  • Mantegani Nicole says:

    Goodbye Philae You made us dreams!!!!

  • Fred says:

    I read that ESA will close the operations in the comet 67P/Churyumov – Gerasimenko doing Rosetta to land in the comet. I am totally illiterate in the technical reasons that made ESA to make that decision, but if the comet has an orbit of little duration (six years), it is not possible to maintain Rossetta in your orbit for more six years until the comet to reach the close most favorable position in the Sun? This would give much more chances to Philae to change of position and to “wake up”. So, can the instruments to run per decades? Do Rossetta have solar panels what can supply it of energy for many orbits? I think if it is possible to maintain Rosseta orbiting the comet for decades, in “stand by” operation until the most favorable position each six years waiting for a signal of Philae, why not to do that? Thank you for your attention.

    • emily says:

      Hi Fred, all good questions. Please see our earlier blog post here ( which addresses the reasons for putting Rosetta on the comet in September this year.

    • Harvey says:

      Fred. There are more extensive explanations from ESA, but in summary no, it’s not technically possible. In the outer parts of the orbit there is too little sunlight, and the spacecraft would become extremely cold and parts would fail. It can ‘hibernate’ for a while, as it did earlier in the mission, but it would not wake up this time. In addition it needs fuel to position and orientate its antennas etc, and it will run out of that.
      Philae is almost certainly now already ‘dead’, and the cold would certainly ensure it was.

  • George Nixon says:

    Are the temperatures of the comet and Rosetta as was expected at this location or slightly lower.

    • ianw16 says:

      It’s been a while since we saw any temperature data. However, what we have seen has been pretty much as expected as far as the comet goes. Nothing anomalous.
      Couldn’t tell you about the temperature of Rosetta though. However, it does carry out temperature analysis for the environment that it is within, including ion and electron temperatures. If it is anything like at Halley, then these will vary considerably, depending upon the distance from the comet, and whether it is within the solar wind free cavity or not.

Comments are closed.