Rosetta tracks debris around comet

This blog post is based on the papers “Orbital elements of material surrounding comet 67P/Churyumov-Gerasimenko,” by B. Davidsson et al, and “Search for satellites near comet 67P/Churyumov-Gerasimenko using Rosetta/OSIRIS images,” by I. Bertini et al, which are both accepted for publication in Astronomy and Astrophysics.

Ever since its approach to and arrival at Comet 67P/Churyumov–Gerasimenko, Rosetta has been investigating the nucleus and its environment with a variety of instruments and techniques. One key area is the study of dust grains and other objects in the vicinity of the comet.

Earlier this year, an analysis of measurements from GIADA – Rosetta’s Grain Impact Analyzer and Dust Accumulator – and images from the OSIRIS camera revealed hundreds of individual grains, either bound to the comet’s gravity or flowing away from it. These included small grains as well as much larger chunks, with sizes ranging from a few centimetres to two metres. Lumps up to four metres in size were also identified by NASA’s EPOXI mission in the environment of Comet 103P/Hartley 2 after its flyby of this comet in 2010.

A new study based on OSIRIS images has now built on these previous detections of cometary chunks, using dedicated observations to perform a dynamical study and determine, for the first time, the orbits of four pieces of debris, the largest of them half a metre in size, in orbit around 67P/C-G.

“Previous studies were based on a handful of images of a given field, and this was sufficient to detect chunks of material and say that they are moving. However, to determine their trajectories and demonstrate whether they are truly bound to the comet, we need dozens of images taken over an extended period of time,” explains Björn Davidsson, an OSIRIS scientist at Uppsala University, Sweden, and lead author of the paper reporting the new results.

Four image mosaic of comet 67P/C-G, using images taken on 10 September. Credits: ESA/Rosetta/NAVCAM

Four image mosaic of comet 67P/C-G, using images taken on 10 September. Credits: ESA/Rosetta/NAVCAM

To follow the motion of the cometary debris in fine detail, the scientists monitored a patch of the sky with the OSIRIS wide-angle camera (WAC), which has a field of view of 12 x 12 degrees – over 700 times the area of the full Moon as seen from Earth. Observing over a thirty-minute interval on 10 September 2014, they obtained 30 images, one every minute, with an exposure of 10.2 seconds each.

Incidentally, these observations were performed just a few hours before the manoeuvre that would place the spacecraft on its first bound orbit around the comet. At that time, Rosetta was 30 km away from the comet centre.

When Davidsson and his collaborators later inspected the images, they identified four debris pieces with sizes ranging between 15 and 50 centimetres, making their way against the stellar background in the sequence of images. The chunks appear to move very slowly, with velocities of a few tens of centimetres per second, and are within four to 17 kilometres of the comet.

“This is the first time that we could determine the individual orbits of such pieces of debris around a comet. This information is very important to study their origin, and is helping us understand the mass loss processes of comets,” says Davidsson.

It seems that some of these chunks may have been accompanying the nucleus of 67P/C-G for quite a while.

In fact, three of these pieces appeared to be bound to the comet’s gravity, moving on elliptical orbits, in agreement with what the scientists had expected. However, the paths covered by the grains over the 30-minute long monitoring were too short to enable a unique determination of their orbits, so they cannot exclude that these three chunks are in fact on unbound, hyperbolic orbits.

As for their origin, the chunks might date back to the last time that the comet reached its closest point to the Sun, the perihelion passage in 2009, when they were driven away from the nucleus by very strong sublimation processes. But since the gas drag was not sufficient to free them from the gravity of the nucleus, they lingered in the realm of the comet rather than dispersing into space.

“This study proves that comets can eject such large chunks of material and that these may also remain bound for long stretches of time as the comet swings around the Sun,” says Davidsson.

On the other hand, one of the pieces of debris is definitely following a hyperbolic trajectory, which will see it soon depart from the comet’s surroundings.


The trajectory of the outbound chunk (identified with the letter ‘B’ in the paper) found around Comet 67P/C-G. The chunk is seen moving against the background of fixed stars. This sequence shows ten consecutive images taken with the OSIRIS wide-angle camera (WAC) on 10 September 2014.
The images span 1.9 x 2.1 degrees, showing a portion of the full WAC field of view. Each image was taken with an exposure of 10.2 seconds, with 60 seconds between the beginning of each exposure. Transient point sources are also visible, likely due to cosmic rays, while the long streaks visible in certain frames are caused by dust grains that happened to be close to Rosetta during the exposure.

“The outbound trajectory of the fourth lump was a surprise: it suggests that the cloud of debris these objects belong to, bound to the comet since its last perihelion, had already started to dissolve in September 2014, when the comet was 3.4 AU (about 500 million km) from the Sun,” he adds.

This is likely the result of increased activity, causing outgassing from the nucleus and pushing the chunk away from the comet’s gravity.

One of the three bound lumps also has an interesting trajectory that appears to cross the comet’s nucleus, hinting that it might have been ejected shortly before the observations.

This possibility is as intriguing as it is puzzling, since the comet was still at a very large distance from the Sun at the time for sunlight to cause enough sublimation and release such a large chunk of material from the nucleus surface.

More sets of similar images were collected after last September, and they are being analysed to identify and study the trajectories of other chunks as the comet got closer and closer to the Sun. However, it will be virtually impossible to recover and identify the same chunks in later images.

As far as Rosetta is concerned, the lumps of cometary material detected by Davidsson and colleagues are too sparse to pose any hazard to the spacecraft operations.

But what about much larger lumps of cometary material, several tens of metres across? Such satellites have been detected around many asteroids and other small bodies in the solar system. Is there any evidence for similar ‘companions’ of Comet 67P/C-G?

Ivano Bertini from the University of Padua, Italy, led a study to look for such satellites around the comet, reporting their results in another paper tto be published in Astronomy and Astrophysics. The team used images that were taken with the OSIRIS narrow-aperture camera (NAC) in July 2014, prior to arrival at the comet, to inspect the comet’s large-scale surroundings at high resolution.

After careful examination of these images, the scientists found no evidence of satellites around 67P/C-G. The upper limits set by these measurements indicate that no chunks larger than six metres were found within distances of 20 kilometres from the nucleus, and none larger than a metre at distances between 20 and 110 kilometres from the nucleus.

Finding such a large satellite around the comet could have brought additional information to constrain the formation of this curiously shaped body. However, the analysis of Bertini and collaborators does not exclude the possibility that 67P/C-G might have had such a companion in the past, that could have well been lost given the harsh events that characterise a comet’s life.


OSIRIS: The scientific imaging system OSIRIS was built by a consortium led by the Max Planck Institute for Solar System Research (Germany) in collaboration with CISAS, University of Padova (Italy), the Laboratoire d’Astrophysique de Marseille (France), the Instituto de Astrofísica de Andalucia, CSIC (Spain), the Scientific Support Office of the European Space Agency (The Netherlands), the Instituto Nacional de Técnica Aeroespacial (Spain), the Universidad Politéchnica de Madrid (Spain), the Department of Physics and Astronomy of Uppsala University (Sweden), and the Institute of Computer and Network Engineering of the TU Braunschweig (Germany). OSIRIS was financially supported by the national funding agencies of Germany (DLR), France (CNES), Italy (ASI), Spain (MEC), and Sweden (SNSB), and the ESA Technical Directorate.



  • Martin Schoenmaker says:

    quote: “This possibility is as intriguing as it is puzzling, since the comet was still at a very large distance from the Sun at the time for sunlight to cause enough sublimation and release such a large chunk of material from the nucleus surface.” end-quote

    Isn’t it possible that the sudden activity reported in April-May 2014 was responsible for the ‘ejection’ of boulders from 67P/C-G?

  • Steve Goodey says:

    Do these chunks pose any threat to Rosetta?

    • Gerald says:

      Not to be expected by collision. But they might contribute to confusing the star tracker.

  • Interesting report, thx. Let’s see how many new lumps will appear during the forthcoming months.

    Might it even be possible that Philae somehow would leave the surface again?

  • A.Cooper says:

    I wonder if the authors know about the 170-metre rock on Site A that was unequivocally detached from the comet and sent into a suborbital trajectory before landing further back down the rotation plane. That would answer a lot of the questions these authors have about the past behaviour of the comet.

    Marco has linked the 170-metre rock before but he didn’t link this 200-metre one that detached in the same manner and floated half a kilometre across Site A in a suborbital trajectory:

    • Marco says:

      Yes. I challenge Harvey and Gerald to look at that annotated image, and tell me why it is not visually compelling…
      If you can accept that this monolith has broken of from its seating position and moved, have a read of the blog post that explains what caused it to move in the direction it did, why it stayed intact, and in its original orientation, just turned upside down…

      • Gerald says:

        Hi Marco, some boulders are likely to have moved. But the candidate you marked isn’t convincing.
        By a 180 degrees rotation you may see similar triangular areas above and below the “sat here” area.
        Besides this the “rock C” area is well-connected with its surroundings, not likely to have been floating around as one fragment.
        Some of the boulders on the smooth areas, however, are candidates for having been displaced, particularly looking at the context, which suggests a cometary version of a past rockfall.

        • A.Cooper says:


          “Some of the boulders on the smooth areas, however, are candidates for having been displaced, particularly looking at the context, which suggests a cometary version of a past rockfall.”

          I presume you mean the rocks on the smooth area of Site A, which I mentioned in the original comment. Rock A is unequivocally matched to the Site A crater rim: it shares a sliced crater with its seating position along with many other features. If its displacement was as a result of a rock fall, it would have had to fall in a direction that was almost exactly perpendicular to the gravity vector. It could only fall (and then roll or slide onto the flat area) if it came from the higher parts of the crater rim. But it definitely came from the edge that’s level with the flat crater floor. Cliff falls cannot explain the 170-metre displacement of rock A. Another mechanism has to be the cause.

      • Marco says:

        Hi Gerald,
        There is a distinct difference between “similar triangular areas” and matching features that challenge our belief that they could be just chance.

        The length, and shape matching is a start. If you measure the feature and the seating area they are close within measurement error.

        Secondly, the tip,is curved down in the feature, and curved up,in the seating area.

        Thirdly, ridges and bumps on the feature has matching ridges and divots in mirror fashion.

        Fourthly, the vector that the base end of the feature points at to the tip is an uncanny match to the same vector made at the seating position. They are pointing in the same direction in three dimensions.

        Fifthly, we are looking specifically for large boulders lifted at the same time as the slab that was removed to make the amphitheatre. The slab reached escape velocity but these boulders did not and they rose and fell in position, the comet slowly rotating beneath, translating their position in a specific direction, rock C being the third one as a confirmed match moved in the same direction from the seating position.

        Rock C is only “well connected” with its surroundings at a casual glance. A good look at different images of this area shows that the similarity with its immediate surrounds is superficial and only because it did not land on the flat area of the amphitheatre.

  • originalJohn says:

    The escape speed is around 1 metre per second. Are the authors suggesting that the jets of hypothesised gas with a velocity of hundreds of metres per second somehow picked up and moved these up to half a metre boulders ( of very low desity, 0.4 g/cc) at a creepingly low speed of less than 1metre/second enabling them to remain gravitationally bound to the nucleus. Have they considered other more likely explanations.

    • Gerald says:

      They have also looked for tidal disruption (near Jupiter in 1959) and impacts.

      • Marco says:

        Hi Gerald,

        I didn’t think the tidal disruptions would be significant for the 1959 pass at around .05 AU.

        Unless there was also significantly higher spin than now, at least…

        • Gerald says:

          Hi Marco,
          that’s probably why they made their conclusions the way they did.

    • Ramcomet says:

      Why not? Sure, on the outskirts of the blasts, there should be descending orders of magnitude pressure, off center, of course slowly launching some tangential pieces just below orbital escape velocity.Why not? Meanwhile, main gas velocity from center stage explosions have sent main slabs out of orbit, long gone. This is not a laboratory with *just* one precise directional measurement in your much needed but unnecessary nozzle and chamber scenario, it’s nature happening in every way possible. Watch any explosion and chart the debris.

  • simon says:

    Very possible indeed . That short burst of activity was probably not caused by the heartingproces of the crust but by other causes like an impact , a cometquake or something else .

  • Erik says:

    My deepest thanks to the scientits who have made this

    Most sincere thanks and congratulations to all of them

  • Sovereign Slave says:

    This is exciting science, and it’s so very impressive that the Rosetta science teams have packed so many instruments and ways of analyzing and capturing data about the comet into such a small package, two packages actually. Is also exciting that virtually every new finding seems to raise more questions than it answers. So am wondering, if comets are supposed to be made of just dust and ice, what are these “chunks” made of? Surely not frozen volatiles, which should have sublimated off long ago, especially if they’ve been orbiting thE comet for awhile, as the article speculates. And if sublimation leaves gas and dust only, these chunks really shouldn’t exist for very long, nor the chunks and boulders of all sizes on the comets surface. Any small chunks of comet should sublimate down to dust pretty quick, but this doesn’t seem to be the case. So if they do not contain frozen volatiles, what are they composed of?

    • Gerald says:

      The analysis of the refractory material from the comet’s surface is still to be published, probably in a few weeks in the Science magazine.
      Chunks which don’t make it into a stable orbit or escape may fall back to the comet and contribute to the refractory material on the surface. So I tend to assume, that there should be a compositional similarity between the chunks and the analysed sutface material at the first touch-down location.

      The few hints published thus far point towards a high abundance of a large number of different organic materials, aliphatic and aromatic compounds, many of which made of carbon, hydrogen, oxygen, and nitrogen. This doesn’t rule out the presence of anorganic material, like silicates. It seems to be very diverse in composition.

      Take fine grains of frozen volatiles, soot, asphalt, basaltic material, mix and stick all together, and then remove the volatiles by sublimation.
      I guess, the closest you can get on Earth are carbonaceous chondrites of type CI:
      This is just a rough idea, and will be obsolete with the publishing of the expected papers.

      • dave says:

        Gerald re analysis of refractory material,

        I know we don’t know the average thickness of the surface layer, but wont the density of Carbonaceous Chondritesstructures that have been found earth be a bit of a problem for the quoted density of the comet.
        Admittedly we don’t know how much of it there is on the comet or what the structure inside the comet consists of.
        Hopefully Phillae & rosetta can team up again and improve the look inside the comet to give us some clues.

        • Gerald says:

          The average material of 67P wouldn’t survive the entry into Earth’s atmosphere.
          The best we can hope for on Earth, is finding some of the most robust fragments of cometary or prestine material.
          Some carbonaceaous chondrites might be related to that.
          But it’s true , that there remains a large gap to fill in.
          The only feasible way to get access to representative material of the early solar system appears to be visiting a comet, as Rosetta and Philae are doing now.

          (There is a factor of 4 between the bulk density of 67P and chodrites of type CI:

    • Ramcomet says:

      Sovereign, I thought you started out with a wonderfully appriciative intro, but then…

      Heeeeeere comes the pitch!

      But how can you miss this? The article refers to DUST grains, chunks, and by extension, dust lumps. And, why not?

      Have you ever freeze-dried primordially formed dust-ice structures that may be composed of hollow fullerine-like cages and/or strangely mixed cathrates (which are also cage like structures – even crystalline, formed slowly, with negligible gravity in a cold vacuum over billions of years), and have you found the sublimated residue to be temporary, as in, “… these chunks really shouldn’t exist for very long,…” No, you haven’t, because no one has.

      Eye of the beholder and all that, but please take off the earthbound blinders: Synthetic NASA Aerogel is light as air but has structure to hold it together.

      And as for the Champagne Cork density analogy of 67P, if you slowly bake a cork to charcoal, it can hold together in the shape of a cork. It’s not necessary for ALL to dissolve into tiny free floating dust grains. Even dust bunnies hold together if you just look behind the old EU cupboard.

      Sorry but your ” if- then” speculative arguments just don’t hold any water, pun intended. You are fishing is all!

      • Sovereign Slave says:

        Ramcomet, you say I’m fishing like it’s a bad thing, LOL. No doubt guilty as charged, though it seems we’re all sort of on the same fishing trip, but some are expecting to catch different fish than others.

        But regarding your counter-pitch post, the irony of your reply obviously eludes you, as beautifully presented in your statement “Eye of the beholder and all that, but please take off the earthbound blinders: Synthetic NASA Aerogel is light as air but has structure to hold it together,” along with the rest of your own earthbound, and even man-made, examples of explanations (“baked corks,” “dust bunnies,” why don’t we just throw in twinkies for good measure).

        However, in re-reading the 24/06 post about detecting what may be surface ice, it says “The (ESA) team also turned to laboratory experiments that tested the behaviour of water ice mixed with different minerals under simulated solar illumination in order to gain more insights into the process. They found that after a few hours of sublimation, a dark dust mantle a few millimetres thick was formed. In some places this acted to completely conceal any visible traces of the ice below, but occasionally larger dust grains or chunks would lift from the surface and move elsewhere, exposing bright patches of water ice.” So, I’ll concede my fishing expedition has no doubt reached barren waters, pun intended.

        • Ramcomet says:

          Sovereign Slave,
          Well said, and you are obviously a gentleman. Looking back, I got a little carried away and my tone was unnecessary. And still, you kept to the high road. I will have that Twinkie now. Thank you.

Comments are closed.