The ups and downs of a comet’s surface

This time last year little was known about Agilkia, the area chosen as Philaes landing site. One year on and the situation has changed, thanks to remote and in-situ measurements that have contributed to building up a picture of this iconic area on the comet.

As can be seen from the many results reported in this blog, Rosetta is providing unrivalled information about the surface of comet 67P/Churyumov-Gerasimenko. One team has been looking in particular at the region around where Philae touched down in order to put the lander’s results in their correct global context.

The Agilkia area on comet 67P/Churyumov-Gerasimenko. The circle indicates where Philae touched down for the first time on 12 November 2014. The dashed line marks the comet’s equator. The large depression is the Hatmehit region. This image is a composite of five frames from the OSIRIS Narrow Angle Camera.

The Agilkia area on comet 67P/Churyumov-Gerasimenko. The circle indicates where Philae touched down for the first time on 12 November 2014. The dashed line marks the comet’s equator. The large depression is the Hatmehit region. This image is a composite of five frames from the OSIRIS Narrow Angle Camera, acquired on 14-15 September 2014. (An unannotated version of this image is available here.)Credit: ESA/Rosetta/MPS for OSIRIS Team MPS/UPD/LAM/IAA/SSO/INTA/UPM/DASP/IDA

Agilkia was the name chosen for the 1 km^2 landing ellipse at which Philae was targeted. Last year, as Rosetta drew closer to the comet, smaller and more distinct terrains within this ellipse became apparent. Fiorangela La Forgia, of the University of Padova, Italy, and colleagues have studied images from OSIRIS, the science camera on Rosetta, to define these areas according to their geological appearance.

On 12 November 2014, Philae touched down as planned in Agilkia, within a region named Ma’at. (Comet regions, defined according to their geomorphology, have been named after ancient Egyptian deities.) It then rebounded, and after two hours drifting about 100m above the comet surface, it settled into its final position, named Abydos and thought to be located on the other side of the Hatmehit region, close to the border with Bastet.

From OSIRIS images, we see that Ma’at, and the neighbouring region Nut, are mainly covered by smooth deposits of fine-grained material. The size of these grains was revealed by the ROLIS images from Philae to be of the order of a few centimetres. Although the depth of this layer is unknown it is probably highly variable across the comet: a 35 m-wide impact crater in the Ash region shows that the depth there may reach several metres, whereas in other places, the view across walls of more compacted material suggests that the layer is thin.

This layer may be the result of ‘airfall‘, which is produced when dust is ejected from the surface but lacks the necessary velocity to escape from the comet. Instead it falls back, dusting the surface. There are a number of longitudinal dune-like structures visible in the Agilkia area, which seem to indicate the drift-paths for this airfall. This is unusual on comets and may be the result of a prevailing ‘wind’ direction in the comet’s gas activity.

Fiorangela and her colleagues mapped the local gravity across the Agilkia area and conclude that the dunes are not caused by dust piling up under the effect of gravity because the regions concerned are almost flat.

The Agilkia area also contains some pits. These are smaller than another population of pits already observed on the surface of the comet, which may arise from the surface collapsing to form sinkholes, and which sometimes generate jets of material that spray from the comet.

One suggestion is that the smaller pits are related to ice. Several bright spots have been seen on the smooth dusty plains that run across 67P/C-G’s surface. The most obvious interpretation of these is that they are icy chunks of comet material that have been partially buried by the airfall. Subsequently, as the comet approached the Sun, the added energy has caused them to sublime. This ejects the fine dust particles too, leading to the excavation of the pits.

Geomorphological map of the area referred to as Agilkia. The main geological units have been identified on the basis of their shape and structure. Credit: Image courtesy of F. La Forgia et al., 2015

Geomorphological map of the area referred to as Agilkia. The main geological units have been identified on the basis of their shape and structure. Image courtesy of F. La Forgia et al., 2015

There are also widespread outcrops, such as cuestas (hills with a steep slope on one side and a shallow slope on the other side), terraces and steep walls, around the region where Philae touched down, as well as many boulders strewn across the surface.

Philae is thought to have finally come to rest in the Hatmehit region, close to the border with Bastet. Although the exact location is still a matter for ongoing investigation, Hatmehit itself is a circular depression that is covered with dune-like features and boulders of various sizes. The nature of this depression is still unknown. From their study of the region, Fiorangela and colleagues conclude that it is not an impact crater but could be an area that sank following the sublimation of sub-surface ices.

As well as looking at the form of the landscape in Agilkia, the team also looked at the photometric properties, to gauge how much light the surface reflects at different wavelengths. The simple answer is: not a lot. The landing site shows an average reflectance of just 0.96% in the orange band (649.2 nm; one of five filters on the OSIRIS camera).

The team reports that there is a marked similarity between the geological units and the reflectance. In particular, smooth deposits have the highest reflectance compared to the outcropping material. These readings are compatible with the surface of the comet being composed of organic material (as reported by the VIRTIS team earlier this year) that shows only small local variation. Nevertheless, Fiorangela and her colleagues suggest that this variation must be the result of small differences in the organic compounds that are present because the combined reflectance and colour variations cannot be explained by surface texture and grain size alone.

Put together, these results present a growing picture of the overall geological and compositional properties around the Philae landing site. As such, they will allow the science teams to put the specific details of the lander’s measurements into a global comet context.

This blog post is based on the paper Geomorphology and Spectrophotometry of Philaes Landing Site on Comet 67P Churyumov-Gerasimenko by F. La Forgia et al., published in the Astronomy and Astrophysics special issue on Rosetta mission results pre-perihelion. The images used in the study were acquired between 1 August 2014 and 12 November 2014 when Rosetta was between 735km and 10 km from the comet surface.



  • A. Cooper says:

    “Hatmehit itself is a circular depression that is covered with dune-like features and boulders of various sizes. The nature of this depression is still unknown. From their study of the region, Fiorangela and colleagues conclude that it is not an impact crater but could be an area that sank following the sublimation of sub-surface ices.”

    Hatmehit can’t have sunk. The strata emerge across the crater end on at an angle. That leaves very little scope for sinkage. You can see the tell-tell lines of the torn-away strata. They form V-shapes in the dust (not just the two obvious ridges but ultra thin spidery lines). These lines can be traced over the rim of the crater to the obvious strata on the south pole. They are one and the same. This means that the strata emerging end-on at Hatmeit (at least half a dozen) were torn away. That’s why the crater rim is stepped. The steps follow the emerging end-on strata. If it’s not an impact crater (it clearly isn’t) and can’t have sunk much, if at all, the obvious conclusion is that these strata carried on above the current Hatmehit surface but were torn away. Stretch theory predicts this along with a host of other corroborating evidence. The mechanism for the loss of the Hatmehit slab is of course spin-up, being at the extremity. Part 12 of the stretch blog, linked here in January 2014 explains it all although, having discovered the Hatmehit V’s and their link to the south pole strata it seems it was a Hatmehit cone rather than a slab. That’s why when viewed from Bastet the head lobe appears to be tapering towards the rim.

    The reason for the small extent of the existing depression is that all these layers at Agilkia that apparently have sink holes are onion layers that rode up over each other during the stretch. That’s why they form V shapes along with those across Hatmehit, all nested together, all the way to Serqet. The small depression is simply the top end of one of the ridden-up onion layers. The one beneath it was the first of the torn onion layers hence the short drop.

    Hatmetit isn’t circular. If you look from above, it is “rhombic” as Marchi et al said. Or, I would prefer to say diamond shaped. The diamond is oriented in line with the diamond shape of the body (logan’s “quadrature”) because both stretched together before the head sheared from the body.

    All the above will be explained in Part 28, which will be out in a few days. This comet is simply bursting at the seams with evidence of stretch.

  • Deb Gustafson says:

    Thanks for the updates, love learning about anything in space!!! 😉

  • Sovereign Slave says:

    “This layer may be the result of ‘airfall’, which is produced when dust is ejected from the surface but lacks the necessary velocity to escape from the comet. Instead it falls back, dusting the surface.”

    Detailed analysis of this region with reference to ejected dust, but still nothing that addresses the dust jets. Still seems to be the biggest elephant in the room, with no apparent or plausible explanation. Perhaps it’s already been addressed in one of the released papers, but seeing as the jets are such a glaringly obvious feature of the comet, what is the explanation, or even some new guesses? Especially when detailed and up close analysis has obviously been done on cometary landscape, yet no reference to dust jet vent holes or other likely features for sublimation driven jets.

  • logan says:

    ‘Dunes’ clearly showing a ‘shock’ from the East 🙂

  • logan says:

    Should be noted that -on ‘drawing’ this pattern- the blanketing material’s kinetics is interlocked to dragging Coma, and not to ballistic models.

  • logan says:

    Not only ‘dunes’. Everything seems ‘shocked’ from East side at this beautiful Agilkia mosaic.

    [5.3 Mb]

  • logan says:

    This blanketing pattern forgives West walls an angled pits.

    Speculating that, as long as the ‘jetting’ is gentle, and the [evolving] blanketing remains over gas outlets. Contributes to spin speed,

  • ianw16 says:

    Stupid captcha thing not working again;
    Really not sure what you are getting at there. The gas from sublimation entrains the dust. The dust is what you see, due to it reflecting sunlight. The gas is invisible in visible light.
    So, for instance, when we saw the pictures from the neck region several months ago, it was obvious that sublimation was occurring there due to the dust jets we saw. When MIRO looked at that region it saw the associated H2O. The gas expands at higher velocity than the entrained dust. Some of the dust will therefore likely fall back to the surface.

    • Sovereign Slave says:

      Yup, I think captcha is trying to weed out the posters that actually have a life, lol.

      So, not sure how my point escapes you, Ian. But I’ll try to break it down into bite sized pieces:

      1. I think that pretty much everyone would agree at this point that columnized dust jets have been established as a self evident observational fact (as opposed to a theory, or simulation, or mathematically expressed conjecture, or hypothesis), similar to saying that it’s a self evident observational fact that the sun produces light.
      2. Gas expands in all directions unless constrained. So the speculation has been that things such as vent holes or other cometary features could account for sublimation driven columnized jets (otherwise, why would so much of the dust travel at such high speeds, and all in the same direction and from the same surface area? Or, put another way, why doesn’t ALL the dust fall back to the comet?).
      3. This article clearly indicates that detailed and up close observational analysis has been done of the comets surface, yet after a year of study, still no attempt at all to shed some light on perhaps the most obvious and agreed upon feature of the comet that is still without any plausible explanation.

      So my point is, why not?

      And if you have the step-by-step, detailed answer to columnized jets of very fast moving dust caused by sublimation, please do tell. And please connect it to what the related surface features on P67 look like so the ESA scientists can perhaps include those discoveries (or lack there of) in their next paper about the comets surface.

      • ianw16 says:

        “No plausible explanation….” Really? Since when? If the explanation is that it is some kind of electric woo, then I would agree with you. It isn’t plausible.
        And I think you are getting mixed up regarding the expansion velocity of the gas and the velocity of the dust/ ice grains.

        Geologic control of jet formation on Comet 103P/Hartley 2

        A simple image search on Google will show all of the features mentioned in that paper. No doubt, as Rosetta gets closer to the comet, it will be able to view the regions from where they know these jets originated.
        There is nothing particularly mysterious or implausible about the formation of these outbursts.

        • Marco Parigi says:

          Hi Ianw16,

          Your explanation just doesn’t cut it for me, and doesn’t address what SS is questioning.

          The fairest thing that can be said about the current state of knowledge about columnized jets is that it is still an open question (but within the realms of conventional sublimation sources of outgassing)

          I am very sympathetic to the current knowledge of physics, but columnized jets are just NOT an expectation derived from our knowledge of that same physics. Columnized jets are only an expectation because they are on every comet that has been visited.

          There is an obvious denial that something very strange is going on with comets, and as SS said … “being consistently “surprised” can sort of be considered a reliable means of reverse prediction. .”

          There’s no point attacking a pseudo-science SS is no longer invested in at all. There is a new paradigm that will be found. EU certainly isn’t it I have promulgated a paradigm which includes stretch theory, but corollaries also proffer explanations for columnized jets.

          Not sure if it will come to anything, but looks promising.

        • Dave says:

          Re Ianw16 – No plausable explantion…….

          You are the only one using the word Woo, your reply to SS answers something else, not his questions.

          No one doubts sublimation of ices, but there is plenty of expalnation to come if it is to explain the various mechanisms, on and off the surface of the comet.

          Your referances to other comets are rarely useful, we already know from our trip alongside this one, that they are not all the same, there have been striking differences with this comet compared with others, notably the flavour of the water and the release of free oxygen.

          In addittion we are up close and for a long time so we are able to observe the before and after in real time.

          There is a lot of help on this blog from experienced scientists to keep the science honest and this is appriciated, but the constant refusal to recognise that there is an issue with sublimation, in the sense that a lot more explanation is required before it becomes anything else than a catch all. This is not good enough after more than a year beside the comet.

          I am sure the project team are working on the issue, but if you dont have enough information to propose some explanation, then it would be better to say you dont know and we need more iinformation, or a better model and add that overal sublimation is the best bet we have today.

          To constantly not see gaps in the expanation can look like head in the sand mentality, which I guess thats how you feel about the woo.


        • Sovereign Slave says:

          Ian, posted a reply, but may not make it through moderation, so here is the meat of my original reply:

          The paper you linked proposes the slumping model, even though they say that slumping has never been directly observed as a cometary feature. They then create a possible model to try to give some kind of credence to this theory, which of course is based on an incredible amount of assumptions, is very limited in scope, and had limited positive outcome. They even talk about many of the assumptions they used, and used the word “assumption,” along with many other frequently used words like likely, may represent, could represent, implies, suggests, inferred, appear to, etc etc. Other than that, the first half of the paper is basically just saying that there are jets, and that they come from virtually every cometary feature imaginable.

          I believe there are more things to be learned from the jets through the ongoing on location presence of Rosetta vs a comet flyby that produced some fuzzy surface pictures.

  • logan says:

    Hi Sovereign Slave: don’t have another idea of strongly collimated jets.

    Really need NOAA there for dust vectors.
    [If something can be said about it, -curiously- is that coma returns are a little more predictable]

    For sometime expected a ‘Great Endorsing’, but timing talks of disagreement on model preference.

    Papers will be, soon 🙂

    • Sovereign Slave says:

      Thanks Logan, can’t wait to see what the papers say. As Harvey points out, sublimation seems to satisfy possible answers some of the activities of the comet in a broad general way, but struggles badly with explaining/modeling details or making predictions, unless being consistently “surprised” can sort of be considered a reliable means of reverse prediction. .

    • Sovereign Slave says:

      Logan, what a hoot! “Great Endorsing” indeed. Funny how the papers do tend to read like endorsements. And yes, no doubt some feverish mental gymnastics occuring over the jet issue.

      • Ramcomet says:

        Hi SS,

        Actually there was an article a couple months or more back with crossectional graphics on how the team thinks the jets formed from relatively shallow sublimation pits, but I don’t know where it is now.

        Anyway, that was “something”, not ” nothing”. Certainly a far cry from “no attempt at all”, that you keep asserting.

        If you find it, post it up and I will do the same if I come across it again.

        I really do try to read EVERYTHING ever posted here. Hours and hours of course, but then at least when I say something I feel I am standing on solid ground, at least as much as a layman can, anyway.


        • Sovereign Slave says:

          Hi Ramcomet, yes, please try to find the paper. I put a coveat in my original post that I wasn’t aware of any very recent papers, but if there is one, would like to read. Can’t help but feel that understanding the jets is perhaps the biggest key there is to understanding comets, yet there seems remarkably little information output about them from ESA so far, and what I have read from past papers has not been convincing at all. And if the paper your refer to has to do with shallow pits, sounds quite similar to Ian’s linked (and highly suspect) paper, especially when columnized jets seem to be observed originating from other cometary surface features besides shallow pits.

          • Ramcomet says:

            SS, OK, will look.

            Yes it was a graphic of a shallow pit, (about square crossection), but not sure how deep the dust in the bottom is, perhaps being more transparent to the stream from a “deeper than shown” pit?

            Also the gasses come in from the side strata, which might help collimate jets, or seemed to do so, in the image.

            Finally, Ian’s link to Geologic Control of Jets above, mentions how the jets are more collimated with more CO2 and less H20, I believe. Comet 103P/Hartley 2’s jets are mostly fan shaped I guess, as a result.

            Wow! Comet 103P/Hartley 2’s geography with “needles” sure is different than 67P! Wonder if the CO2/ H20 ratio accounts for that too?

            That graphic I mentioned is in this blog from Rosetta team, somewhere!

          • Ramcomet says:


            Ian’s link also describes how the needles and knobs may have formed in the middle of widening pits, the flow characteristics of side ways venting gasses meeting in the middle and possibly collimating upwards?

          • Gerald says:

            A meadow with a tree with withered leafs, autumn, stormy with wind roughly from one direction, gust, “collimated jet” of leafs.

            Storm invisible, leafs visible.

            Comet: A spherical comet (for simplicity) with a dust deposit, near perihelion, gas emanating from the comet with about 700m/s, concentrically away from the comet,
            outburst under dust deposit, “collimated jet” of dust.

      • logan says:

        Clear by now I’m not native speaker. Hope that word is not taken in a bad sense. Junior Doctors need endorsement from Senior Doctors, when going on risky, diffusely defined cases. Endorsement in the sense of the document being tasted -and approved- by Scientists with slowly ‘cooked’ expertize 🙂

      • logan says:

        Neurons starting to spark again 🙂 Finally got what you mean, Gerald.

        Tree being the leaf dispenser.

        What would be that isolated, strongly perimeter -ized dust dispenser?

        What would be desolated ‘meadow’ equivalent, surrounding that ‘dust dispenser’, at 67P surface?

        • logan says:

          This tree model depends Harvey’s ‘viscous’ limitations. Is classic aerodynamics keeping the leaves within turbulence. But here we have a molecular scenario.

  • logan says:

    “…Most of the compact particle detections occurred at attitudes and longitudes where the spacecraft was in view of the comet’s neck region of the nucleus, the so-called Hapi region.”

    Always had difficulty imaging the compact grains traveling from down under, at sublimation line. Plausible thermal stress as cause of grain detachment at neck’s cliff.

    “The integrated [micron and submicron] dust mass flux collected from the Sun direction, that is, particles reflected by solar radiation pressure, was three times higher than the flux coming directly from the comet nucleus.”

    Wow, the coma ‘charge’ of dust is impressive, considering how little object Ducky is. The additional two charges are remnants from past perihelions? [Think micron and submicron dust should be ‘washed away’ when Coma retreat, at aphelions]. Could that additional charges be generated from bigger particulate?

    “The speed of these particles, having masses from 10 −10 to 10 −7 kg, ranged from 0.3 to 12.2 m s −1 . The variation of particle mass and speed distribution with respect to the distance from the nucleus gave indications of the dust acceleration region.”

    There is a ‘dust acceleration region’ according to at least GIADA Team. Cold Sep14-Feb15 period, accelerated by what? 🙂

    “We determined the speed confidence intervals by means of the bootstrap method and obtained (2.5 ± 0.8) m s −1 , (3.0 ± 1.0) m s −1 , and (4.3 ± 0.9) m s −1 for 10, 20, and 30 km, respectively. These results suggest a possible speed
    increase between 10 and 30 km.”

    Congratulations to V. Della Corte, A. Rotundi, M. Fulle, E. Gruen et al. on publication of “GIADA: shining a light on the monitoring of the comet dust production from the nucleus of 67P/Churyumov-Gerasimenko” 🙂

    • Sovereign Slave says:

      Logan, you’ve inspired me to read through all the papers listed on the ESA website, though will take awhile and I’ll not understand much of the science lingo. However, your last point is extremely interesting, and don’t remember hearing anything about it on the blog. Here is the full text:

      “We classified the dust speed data into three subgroups according to the cometocentric distance: up to 15 km (labeled as 10 km), from 15 to 25 km (labeled as 20 km) and from 25 to 35 km (labeled as 30 km). Each set of data was grouped into speed bins (Figs. 9b−d). We fitted the data with Maxwell distributions due to the data skewness. We determined the speed confidence intervals by means of the bootstrap method and obtained (2.5 ± 0.8) m s-1, (3.0 ± 1.0) m s-1, and (4.3 ± 0.9) m s-1 for 10, 20, and 30 km, respectively. These results suggest a possible speed increase between 10 and 30 km.”

      Now, as I’m not a scientist (or, as sjastro would put it, I’m someone who lacks the skills or the compulsion of learning the maths and physics necessary to be one 🙂 ), but it does seem to be saying that dust leaves the comet at one speed for up to about 10 km, then it actually INCREASES speed starting at around 10 km. But, if that’s NOT correct, please let me know.

      Now, in the Summary, they address this issue starting just above the large equations, but I can’t tell if they are simply outlining what they found, or are including some kind of explanation. So, any input from the science types (or Emily) would be appreciated. Seems this is unusual activity, and sort of big news, and yet it’s not something I remember seeing on any of the posts before yours, Logan. Perhaps it’s easily explained and anticipated behaviour, but you apparently don’t think so.

      Here’s a direct link to the paper:

    • Gerald says:

      “There is a ‘dust acceleration region’ according to at least GIADA Team. Cold Sep14-Feb15 period, accelerated by what?”
      I’ve been understanding this always as due to drag by evolving gasses.
      The very first separation of grains from the ground may be more complex, with short-distance effects (van-der-Waals, electrostatic/photoelectric, mechanical/thermal stress, fracture propagation, etc.) involved.

      • Sovereign Slave says:

        So, you’re saying that the dust is a certain speed leaving the comet due these gas, etc, fine. But it’s my understanding that whatever top speed the dust achieves, it should remain at that speed. Or put another way, explaining why the dust is a certain speed the first 10 km doesn’t explain why it accelerates afterwards, though it might explain a deceleration if dust were moving toward the comet. Acceleration indicates some force acting on the dust far beyond the scope of the comet’s direct influence that’s causing the dust to accelerate. So back to Logan’s original question, what is causing the dust to actually accelerate?

        • Gerald says:

          Sovereign Slave,
          once again, the drag by evolving gas.
          The gas expands with about 700 m/s away from the comet. The dust moves only with about 10 m/s. So you have a supersonic “storm” dragging the dust. The dust never gets the top speed of 700 m/s.
          Over more extended periods of time, you get additional effects by solar radiation, mainly on the smallest grains.

  • logan says:

    If neck’s material is from cliff origin -on precedence, Then cliff is showing us, directly, the water ice hiding substance.

  • morganism says:

    SS, if you want to read, most sci results linked in these posts.

    the collimation effect is also discussed in one of the threads as being the result of 3 or more jets intersecting, and forming a helical? structure just by the interaction.That has been observed before.

    I have to agree that shallow pit sublimation is not cutting it for me either.

    It also is interesting that most of the jets seem to be emanating only from areas that have a dust blanket next to or inside a crustal feature.

    I am wondering if the jets formation mechanism is linked more to the layering/onion skin structure exposed on the neck region wall, and visible on most of the big boulders
    and crater walls. It may be that it takes exposed surface heat and transfers it laterally , and stores it under the blanket. Or more likely that the temp differential stress between the two causes crustal fracture right at the border between the insulated and un-insulated areas.

    Remember that carbon dust is one of the best insulators in materials science. It was proposed as heat insulation for a solar probe at one time IIRC.

  • logan says:

    This is the point: micron_ and sub-micron erosion is happening not on, but above surface 🙂

Comments are closed.